UCLA CENTER FOR HEALTH POLICY RESEARCH HEALTH ECONOMICS AND EVALUATION RESEARCH Final Evaluation of California’s Whole Person Care (WPC) Program DECEMBER 2022 UCLA CENTER FOR 3, HEALTH POLICY RESEARCH 62: Racine tts to aire ev Final Evaluation of California’s Whole Person Care (WPC) Program Nadereh Pourat, PhD Emmeline Chuang, PhD Brenna O’Masta, MPH Leigh Ann Haley, MPP Xiao Chen, PhD Weihao Zhou, MS Menbere Haile, PhD UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 This evaluation was supported by funds received from the California Department of Health Care Services (contract number 17-94448). The analyses, interpretations, and conclusions contained within this evaluation are the sole responsibility of the authors. Acknowledgments The authors would like to thank Denisse Huerta, Kelly Taylor, Wafeeq Ridhuan, Nadia Safaeinili, Dahai Yue, Ammar Bhaiji, Christine Lo, and Michelle Pham for their hard work and support of WPC program evaluation activities. Suggested Citation Pourat N, Chuang E, O’Masta B, Haley LA, Chen X, Zhou W, Haile M. Final Evaluation of California’s Whole Person Care (WPC) Program. Los Angeles, CA: UCLA Center for Health Policy Research, December 2022. UCLA Center for Health Policy Research | 10960 Wilshire Blvd. | Suite 1550 | Los Angeles, CA 90024 | t:310.794.0909 | f:310.794.2686 | healthpolicy@ucla.edu Rana ates emits UCLA Center for Health Policy Research December 2022 Health Economics and Evaluation Research Program Contents EX@CUTIVE SONIA ercrecsoncererssensron examin nuenenrommmnnnmrnn Tener 27 WRG Programm OVERvVieWs..crsecrccnrosavensrneonononsnonstensevnvavsenvesnusonensannsns cedenenseneson trnnnnensnesrtesnenntonotee 27 Evaluation Methods........ccscssessssessseesesseessesseessssssesssessessscsssassessssssesssessessseessasseeneasscensaeseeneseens 27 RESUltS scsscnconnmemornnmnmonnnmanimnnnmmnn anon ARENDT 28 Structure of WPC Pilots. Health Information Technology and Data Sharing Infrastructure .......ceessssesesesssesssenseeee 29 WPC Enrollment Size, Patterns, and Trends WPC Services Offered and Delivered sicissississcsveesovisisssirsaviiasveesieesivereesssessscssaseaseseevecaetes 31 WPC Care Coordimation ........cccecsssessessseesessssessscsesssessessssesssessesssessescsssnssssvensssssenseseneaeeteee 32 WPC Quality Improvement, Program Monitoring, and Stakeholder Engagement.......... 33 WPC and COVID-19 Enrollee Demographics, Health Status, and Prior Health Care Utilization... 34 Better Car Gacrccsrnaninrnae aio ninn aera 35 Better Health LOWER COStS s cccssccscccscosessscisesvennvsnis eensvens seaveasntanseeverenanea nnnesieeetseveneceaeoiasensi cies pensenseenaectaeces 40 Homeless WPC Enrollee Services and OUtCOMES ......ssesecsesseseeeeeestesseseeseesesseessessestenesee 43 WPGC Transition to CalAIM nuns nnn enna RIES 45 Implications Chapter 1: Introduction ..........cccecesessesessesssesesesesssscsessssessesssesssassesasscseeassesncacsssnsecsvensaneseesanereeeatens 47 WERE Program scene wnnswsnn emanate eee en ean me 47 WPC Goals.. WPC Lead Entities Target Populations, Services, ANd REPOrting...........seceeeceececeteeseeeeeeseeeeseeseeeseeseseestaeteesenetateeeate 51 WPC Funding and Pilot Payment Methodology ...........::sscsssessesssesssseseesssessesesessescssessesseeseseeees 52 UCLA Evaluation Conceptual Framework........cccscesssesssessesseseecseeeseesseeeseaeseeeseeeeseeeseeeseseeesseessenssenseeeseeteeese 54 EValUstion QUESTIONS sis nnusnsenannianin mame: 56 Reyer) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Data SOULCES ......cseseseseseseseesesesesesesesesesesesesesessensensscsnsnsesssesssesssessseseassesesessseasseasaeseseaenees 57 Analytic Methods Chapter: 2::Structure:of WPC Pilots svsccsescssicssveessisssvesvessvessasscsssusesaveeasevevevavensvensesutvsasverresvetenactes 61 Organizational Structure.. Tareet Populations cnnmascnmmannnmonmnoa nm aT. 63 Pati ?S AIDS cccssessraverssemssssncrnrmnercesn serra ennaeameeno inn reT TEASER 67 Chapter 3: Health Information Technology and Data Sharing Infrastructure .......ceeeeeeseeeeeees 82 Data Sharing Agreements and Enrollee Consents Data Sharing Platforms and Tools to Support Care Coordination ........cssseseeneeneeneeene 86 Use of Incentives to Promote Data Sharing.........cccsssesssessssssessessssssssssesssessessssenesesseeesees 90 Challenges Related to Data Sharing and Reporting. Successes in Data Sharing and Reporting .......:.cccsscssseseseesseseteeetetssetsesssesseeereestensseeseesenaee 92 Chapter 4: WPC Enrollment Processes, Size, and Patterns .......cccceceseeeeseeseseeeeeeteeseseteeseeerseteeee 95 WPC Processes for Identification, Engagement, and Enrollment of Eligible Medi-Cal B@NGTI CITIES scccstccrsavessvesseseseesivoceveressuaweaevecsntaveneue rT ecatCesTTeaaNiN ata OREN T RCN EAN SNEaSa NaN OREN 95 Identifying Prospective Enrollee@s .....c.cssssssscesscesseesseseseeseessestsessseessesessssssseesaeaseeaseeaeaeaenees 95 Determining Eligibility... cccececeseeseeeeeeseeeeseseeseseeseaeeeesenecsesecseaeeaeseeecaeeeseeetseeeaeies 100 Enrollment Approach Enrollee Engagement and Retention.........ccscssssesssesssesseessssssesssesseessesssessesseesssesereneee 103 Challenges and Successes... WEPC Enrollment Sizeand Patterns vavccccormnimannennnanmin manna 106 Enroll Metit: Size scccassssssescsarsceanenrcinwesnne wissereesieraiarentcusireT sins ranTsC aN Sect ioe TeNCTEEE 106 EnrollmentsPatterns:.......casesscoarenunecerscreusmonsossaessenveonsennanssxansrmenavenmeneanenracanentaessensssnsoness 110 Disenrollment. Enrollment Size and Patterns by Target Population .........cccssessesessesseeeseeeeeeteeteeeee 114 Chapter 5: WPC Services Offered and Delivered ...........cscccscssessseesssessessseseesssessacseeessesneeeseaeeeeas 118 WPC Services Offered .......ccscsssesssssseesesestessseessssseesssssvesssesssssseasesssesisasscsssssvesssesueassnensasseenenesees 118 WPG SERVICES DelIVEF Ed seen cramamenenerwenenwunncneme eA 122 QUE ACN svssssssvnssesssanaconsoxeccavevsrsonswenesvuvavarasvcasoventaxsnanvaxontevancesevonunsanvuueenssumvesenvaveuaretaesenes 122 UCLA Center for Health Policy Research nee lyr) Health Economics and Evaluation Research Program Care Coordimation .....c.ccccccsssessssesesessseessssssessscsesssesssssseesssssesssessasseseasseesssesreesseseeesees 123 HOuSirg SUBPOR wisinenanennaannmna RE 124 B@retitiASSiStanCe sccvisisssssovsveresvevscxecesvovasessvevssonseaerceansvscvecevasauaoataaeravevoueaeivesaebateNaNaNIeaRs 126 Employment Assistance.. Sobering Centers sncranunmonnnannmeanmam EET 129 Mdicall RES BIt@ tescsssercscesscacsunemnemenencoanen renner eneumonr Renae oT 130 MANS PORATION .coneseccssexscenraxsevesne ccrssecescennnomsenaeasuennexenredontenansntecseanusnecenenniaepenensmnmsenereeenn 131 Health Education ......ccccccssessessseesesessesssessssseessssseessssssesssesiessseassssseessssseesssesiessseaseesseentes 132 Legal SENICES wisscssssiicenesinummnnneinannnun ana aimandnnamenencT 134 ReFENtry SERVICES ssscssosvesssnesveossssvesevensvassvoeasesavassesssceanscevinssuvaraevvenonentstensescusvevestouessesasteRs 135 Services without Enrollment. WPC Expenditures and Payment for WPC Services.........sssssessesssessesesessssesresssessesestsnssssrenesesnes 136 Chapter 6: WPE Care COON at OM a isecescoscnmencusvencnasr avecennenrernnieermennercsvncssmeceaereiectsconisers ss 142 Care Coordination Infrastructure ......cssesssssssseseessesseseesessesssesseessestessessasesssetsseeseessensenseee 143 Care Coordination Processes... Care COOTdITATON Sta iN Bisse ccceernsernccernnnn aera 149 Challenges and SUCCESSES ........sssesesessseseseseseseseseseasscaesesescesscessessucesecsseenseensestsestseseeeasseaseeeeneaes 152 Chapter 7: WPC Quality Improvement and Program Monitoring .......c ccc eeeeseeseeseeseees 156 Pilot-Initiated Quality IMProveMent..........cccecseseesseessessseeseecsesssescscssscsssesssessesssecsesesseeseeesenees 156 PDSA TYP@S.csseseccssssssssscccccssssssssssssnsuusssssesseccecessssssssssssiusunssssessceceeessssssassssuununsesseseeceeeeeses 156 Volume of PDSAs Conducted by WPC Pilots, PY 3-PY 6.......:.cessssssesssesseseseeseseeeeseeereeeees 158 Technical Assistance Stakeholder Engagement on Quality Improvement Activities 0. 160 Chapter 8: WPC and COVID-19 ooo... se eeceeeeeeneeeeseeeeseseeessesecseeeseevaceeaesesseserecaeeasseeetaseetaseteseees 163 Progression of COVID-19 in WPC Countie@S........ccccce cece eseseseseeesenssessessseseeseeeeeeeeaneee 164 Impact of COVID-19 on WPC Implementation and InfrastrUCtUre .......c.ccseseeseteeteteeeeeeeees 165 How WPC Infrastructure and Processes Facilitated COVID-19 Response ..........:csceee 165 Impact Of COVID-19 ON WPC.... cece tees tees tees seescsnscenscenscesseeseemecsssessesieeeeeeeeee 170 COVID-19 Target Population ele) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Impact of the COVID-19 Pandemic on WPC Enrollment Characteristics of WPC Enrollees before and after the COVID-19 Pandemic ..........ssseeeees 174 Estimated Prevalence of COVID-19 among WPC Enrollees COVID-19-Related Health Service Use of WPC Enrollees .. Changes in Healthcare Utilization from COVID-19 ..........ccssessssssesseseseesesestesssesseseseesescsrenesesnes 178 Challenges, Successes, and Lessons Learned Related to COVID-19 ......eeeeeseseeseseeeteeeeeeteeee 180 Chapter 9: Enrollee Demographics, Health Status, and Prior Health Care Utilization ............... 183 Demographics ......cccecceceseeseseeeeseeeeseseeecseeecseseeaesecseseeeesenecsesecsesescaeecaenecaeetesieeasieeetaeteeaseeteeaeee 184 Health Status ccsissannancnnacnnnnnannainarann mn naenTS 185 Utilization: Prior tO/ENCOlIMENE, ccsssccvsissvasccsoussvoesneasocesaesscasvereveacovessesvvavsressesuanevesvevenessosessesewsese 186 Selected Outpatient Service Use Prior to Enrollment Emergency Department Visits Prior to Enrollment .........csscsssessessseessssssesesesreseseseseseeneee 187 Hospitalization Priof tO Enroll M@hit secccccnanisieiennnnnema cen annniinnnnsess 188 Chapter 10: Better Care... ccsessessssesesssesesessscscssecsseessecsescneesieesteeseeeseeeseeessetseenseeessarseesneasaensaenes 189 Utilization of Outpatient Services. Primary Care SQWICES: i. ccicssiscsesiessnnceaenassisnmenrs eesti ennNTE 190 Specialty: Care Services. -icessesccssseeeearesvecmensuvenauncsnne anevunessasaisretsenssnevenmnonecareceveterene 192 Mental Health Services ........cccccscesecessseseeseseeeseseessssssesssesssssesssasseessesseessseseesseseeeseeeees 193 Substance Use Disorder SétVvices ven ssicammnnimmarnannnminnerammnnnmmeanennenrs 195 Follow-Up After Hospitalization for Mental IIIM@SS ...........ccsesseeseesesessesesesseseseeeeeseeeeee 197 Initiation and Engagement of Alcohol and Other Drug Dependence Treatment........... 199 Trends in WPC Pilot-Reported Metrics Comprehensive Care Plan Major Depressive Disorder: Suicide Risk ASS@SSMENE ..........:.seseesseeseeeeseeereeteenereeeeseeeenee 204 Pilot Assessment of Challenges to and Impact of WPC on Better Care .........csescsesseeeeseeeees 205 Chapter 11: Better HOAl th issicrsccscsscssnamninisrmenenanwninncniniemnaniwenia cette 207 Utilization of Acute and Long-Term Care Services ........c:ccssssseesssessesssessessseesscssesssessenssesneneees 208 Ambulatory Care: Emergency Department Visits ..........cesseeeeseeeeseteeseteeeeeeestsetetaeteteenees 208 Inpatient Utilization . UCLA Center for Health Policy Research ee Tyr) Health Economics and Evaluation Research Program Long-Term Care Stays .....ecsceecseesesesessseensensesseesessessecesessesseeaeseesneeeesesesseeseseeseesenseseeneeegee 212 Better Health OUtCOMeS si iianisiessciinianinssnacnnaninsaninmintaninannacaae nie 213 Controlling High Blood Pressure ..ccisosssessssisesesscvssesevesoessvevecevavevecaversostscsateeasonesosssetenes 213 Comprehensive Diabetes Care. All-Cause Readmission Trends in Better Health Based on WPC Pilot-Reported Metrics ........cccccceeeeseeeeeteeseeeeeeees 216 Variant Metric: Decrease Jail Incarcerations (DJ) ....cccceseesseesseeseesseessessseessesssessseaseenes 219 Variant Metric: Overall Beneficiary Health... cece eseeeeseteneeseeseseseeeeseeseseteeaseeteesees 220 Variant Metric: Controlling High Blood PreSSUre......cccssccssessseessesseesseseseecseeescessesseessaeeee 222 Variant Metric: Comprehensive Diabetes Care (CDC) ....cccccscssesseseseessesssesssesssenseeees 224 Variant Metric: PHQ-9/Depression Remission at 12 Months (NQF 0719). Pilot Assessment of Challenges to and Impact of WPC on Better Health ........ cesses 225 Chapter) 2s LOWE COS tiscssaverescsserancconmavennsrerninevene een nsenvnteiia venta iceman nennenecsnerees 227 Total Estimated Medi-Cal PayMents........sccsssssssesssesssesesessseassessesssneseeeneeeeeneeeneestenreeneeeeseeree 228 Estimated Payments for Outpatient Services.. Estimated Payments for Outpatient Medications Estimated Payments for Emergency Department ViSitS........:cssessessesseesssesseessesseeeeeeenees 235 Estimated Payments for Hospitalizations... cece ceeseeeseseseseseeeseassesssesseseeeeneneeaees 237 Estimated Payments for Long-Term Care StayS........ccscessseeseesesetscseeeeseeetsesseecseeerseeeeaseeateees 238 Estimated Payments for Residual Medi-Cal PayMeNt...........cccsccsessesessesesesesssrenescsresssesneneeee 240 Chapter 13: WPC Services and Outcomes for Enrollees Experiencing Homelessness ............... 242 Approaches to Enrolling and Delivering Housing Support Services to Individuals Experiencing Homelessness and At-Risk-Of-Homelessness POpUulations.........csessseesesesectseeeeeteeeeeree 243 Identification of Individuals Experiencing Homelessness.. Outreach to Individuals Experiencing Homelessness ............:ccssseseeseereeseteeseseeeeeeneeeenees 243 HOUS Ng SUPPOrt SERVICES: ....svsrserserrsncresersnonevesensssnevenrsnsnnnstecesereseevsenventssosusancnanensnenene 245 Tracking and Retention ..........ccccsescecesesseseteeseeetseseeeeseeetseseeassecseseeesseeesseecseneeateetseseeeeeee 250 Specialized Housing Staff in Care Coordination Teams Enrollment Patterns and Characteristics of WPC Enrollees Experiencing homelessness......253 Reyer) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Enrollment Patterns and SizZ@......ccscccsesessessseesseseesssessesssestesssessessseessssseesssesressseaeasseentee 253 Demographics Trends in Pilot-Reported HOUSING Metrics.......cccccecseseseseeteseeseseseseeteeseesecsecseeseeeseeeseeeee 260 Variant Metric: Permanent Housing.. Variant Metric: Housing Services..........csccsscsssesssesssesssesteeseecsteeseecsescsesssesssessscssscsssesseeees 262 Variant Metric: SUPPOrtive HOUSING). .scccccscsccnsesscaveccssesnceeesstavedesoneticcentecctsiecstesndoreerd 263 Comparison of Adjusted Trends Between WPC Enrollees Experiencing Homelessness and their Controls, Before and After WPC Implementation ..........eseeceeenseeeeeeeseneeeetetseneeeeneeetee 264 Health Service Utilization .cccsissssesssssivsscssssesscessessiessicvaxssessversssesscessarvacssvasvsennassicacaaveecsioaty 265 Follow-Up After Hospitalization for Mental IIIn@ss .........cceeeseseeseseeeeeeteeeeeeteeastetseeeeeeeee 269 Initiation and Engagement of Alcohol and Other Drug Dependence Treatment...........271 All-CaUse REAMMISSIOMisissicsvsseevssccvvcevscsswivawernsesnicenavaaeneniieresueevenesestcaaNeaNeNeNesueNesuesARNS 273 Challenges and Successes Approaches to Address Housing Challenges .........c:scsssssessssessessssssesssssessssesssssrenesssneneies 274 Chapterc14: WRG T rainisition té! CalAl Meneses ener nena 276 Planning and Preparation for Transition .......cccscesseesseesseeseeeseessesssessseesecssecssscssecaseeeeeeeeesees 276 Technical Support for Transition Pilot Participation in Transition Planning Meetings ..........::c:ssseseeseeseteesseeeesetetseneeeenee 278 Participation in Enhanced Care Management ...........ssccsssesseesessessseeesessesssesseseseeeeeseeesee 279 Community Supports Trafisition Challénges:and Successes vaccine 285 Sustainability of WPC Goals and Pilot Innovations after WPC ........ccccceesseeseseteeseteeeeeeeeeeeeeeee 287 Inter-organizational Collaboration between WPC Partners..........cseeseesseeeeeteeeeeeeeeee 287 Data Sharing Infrastructure Needed to Support Integration of Care... Cate COOIIRAION beisccnncrnnninonrmnanrnsnnIniTA connie aC CI TITS 290 CONCIUSIONS:.1ss0r-arsxernaevrranuenennsnenonsmntonnevoneveanonssuonuniannaseitenewensng sotasonetedveterecannsavenensavens tonntconesseneee 291 Structure Of WPC PilOts.......cccccssssseeesessesssessssseessssseesssessesssesnsessesssssseesssesesssesusssseensssseenenesees 291 Health Information Technology and Data Sharing Infrastructure... 291 WPC Enrollment Size, Patterns, and Trends .........cccscsseesseesseesseesseesseessesseesseessesssessseesseeseeees 292 UCLA Center for Health Policy Research ee Tyr) Health Economics and Evaluation Research Program WPC Services Offered and Delivered.......c.ccscssssesessssesssssesssesssssesssssseesssesresssesessseessssseessesnes 292 WPC Care Coordination WPC Quality Improvement, Program Monitoring, and Stakeholder Engagement .............4+. 293 WPC and COVID-19. Enrollee Demographics, Health Status, and Prior Health Care Utilization... 294 Better CaF sesscscnacncnnisecnniensvecermnanniiannn waren a nientiiestaanniiamareeemenrceets 295 Better: Health, occccssvsssssnssosseoocovesnesnsexoavonssssstsonvonsnssnxeonsesteasscevexsnussunsunteessrbeenvoxsnosnssoveventesves tennis 296 Lower Costs WPC Enrollees Experiencing Homelessness Services ANd OUtCOMES......cccseseseseseseseseseseneees 297 Sustainabilityand Transition:to:CalAIM. nsescvvesscevseassvasvevcusvececvaasotessevousnsasovanvevevesvavensavenensioes 298 Implications... Appendix A: Data Sources and Analytic Methods for Quantitative Analysis ........:csecseseeeeeee 301 Appendix B: Data and Analyses Methods for Pilot-Reported Metrics .........:cccecseeeseteeseeeeeeeees 324 Appendix C: Data and Analyses Methods for Narrative Reports .......ccsecssssseeeseessesssesenensneneacees 364 Appendix D: Data and Analysis Methods for Lead Entity Surveys -. 366 Appendix E: Data and Analyses Methods for Follow-up Interviews with Lead Entity and Frontline Appendix F: Data and Analyses Methods for Partner SUrVeyYS .......:csscsseseseesseesseseseaseeseneasneneneers 374 Appendix G: Data and Analyses Methods for PDSA REpofts..........:ccsscsssessessesseseseessesseeesesneeeee 376 Appendix H: WPC Services Offered through PMPM Bundles and FFS Appendix |: Pilot Primary Target Populations and Reporting...........sscssscssesssessseseeesesresesesneseeee 391 Appendix J: Selected Illustrative Examples Of WPC PDSAS.........cccscecesesseeeseeseeeeeeeeeetseteeaeeeeeesees 425 Appendix K: Policy Brief Care Coordination Framework .......cssscsssessseessesseeeseeesessseesseesseaeseneaeaee 430 Appendix L: Policy Brief — A Snapshot of California’s Whole Person Care Pilot Program... Appendix M: Policy Brief - COVID=19 wcsicisssssiscsssssisesusevivascsavesasesavesavesanteardsasscassorascactaasncaveaasseess 447 Appendix N: Lead Entity Survey Instrument ...........ccccsceesesssseseesesessesssesseseseessscseesssesseesseaneeeees 455 Appendix O: COVID-19 Survey Instrument ......... ee cecsseeseeeeseeeeseseesteeeesesetseseteeseeesseetaeeeteeeetaeeees 531 Appendix P: Lead Entity and Frontline Staff Interview Protocols..........::scssssessesseseeseesesreeneenenee 536 Appendix Q: Partner Survey. Instrument...cccsissssvsssssoressesscassovonsevonsseassvensstvesesensneavenensanssanssssnsens 547 iy ply) UCLA Center for Health Policy Research aati Health Economics and Evaluation Research Program Appendix R: General Glossary. + 562 Appendix S: Enrollee Demographics, Health Status, and Prior Health Care Utilization by Target Population Appendix T: Comprehensive Community Support Offerings by COUNtY .......ccseeseeseseseseseeeneees 571 Appendix U: Pilot Specific Case Studies .......c.scsssssssesesssesesssesssessnesssesenesseessnseeseesteesenssereseneeenee 574 UCLA Center for Health Policy Research eye) Health Economics and Evaluation Research Program Table of Figures Exhibit 1: Care Related Difference-in-Difference Model Outcomes for WPC Enrollees, PY 2 to PY Exhibit 3: Cost-Related Difference-in-Difference Model Outcomes for WPC Enrollees, PY 2 to PY 6.. 40 Exhibit 4: Difference-in-Difference Findings Comparing Trends in Yearly Estimated Medi-Cal Payments per Beneficiary for WPC Enrollees and Controls ....cccccsecsesssesssessseseseseseseeeseassenesens 41 Exhibit 5: Timeline of Key Whole Person Care ACtiVities .........ccccssssssessssessessessesssesresessenesessenesess 48 Exhibit 6: Map of Participating Lead Entities and Counties in California .........ccsccssseseseseneseneene 49 Exhibit 7: WPC Pilots and Participating Lead Entities... ecsseseeeeseeeeeeteeeeteesteeteeseeetseneeeeneeee 50 Exhibit 8: WPC Universal and Variant Metrics........ccscsssssssssssssesssessesssessessseesescssesssesseseseseeesseenseess 52 Exhibit 9: Whole Person Care Budget Categories........c.csccsssssesssessssssesnessssssssssesssesseessssnescsseeeess 52 Exhibit 10: Whole Person Care Conceptual Framework.........ccscessesssesseseseenssesseassesresesssesesrenenees 55 Exhibit 11: WPC Evaluation Questions and Location of Associated FINdingSs..........:cscsseseeeee 56 Exhibit 12: Overview of WPC Evaluation Data SOUICES ........csscssssessesssessessseesessssesesesteseseessesrentiees 57 Exhibit 13: Selection of Primary Target Population by WPC Pilot, PY 6 .......cccceseeeeeseeseeeeeeeeeeee 64 Exhibit 14: Examples of Criteria Used by WPC Pilots to Assign Enrollees to Primary Target POpUlatiOns oo... cccecseecesesesesesescsessscscsescsesescscecseaeseasseesscesecseeeseeesesnseseseessestsesseesvesereseeaseraseraceeasetes 65 Exhibit 15: DHCS Pre-Specified Partner Type by Lead Entity, PY 5......ccccscsssssscsesesesesensensteneeeee 68 Exhibit 16: Selected Examples of Specific WPC Partners by DHCS Pre-Specified Partner Type and their Role within the WPC Pilot, PY 5 ....c..cccccccscsssssssscsscesscsecssssscescssecsecsessassaseaesaseasesseesseaseassaees 69 Exhibit 17: WPC Community Partners by UCLA Service-Specific Classification, PY 5..........006 70 Seer) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 18: Selected Examples of Types of Community Partners by Service-Specific Classification, Exhibit 19: Level of Partner Engagement in WPC across all Pilots, as Determined by the Lead Entity) PY 3 anid: PY Sissssssssesssvssssssnveesierevarstcansavessivcensacesseeseassianenesacannciconsenvenrsanivioveancownteastonesties 72 Exhibit 20: WPC Lead Entity Designation of Level of Partner Engagement in WPC, PY 5............. 73 Exhibit 21: Level of WPC Partner Engagement by DHCS Pre-Specified Partner Type, PY 3 and PY Exhibit 22: Level of Community Partner Engagement by UCLA Service-Specific Classification, PY 3'dNd PY Svasnmrennmnnnmanrem AT RN eS 75 Exhibit 23: Selected Examples of Challenges and Solutions to WPC Partner BUy-in .........0 76 Exhibit 24: Selected Examples of Partnership Buy-in Successes Among WPC Pilots ...........::6: 77 Exhibit 25: Type of Interaction with Partners among WPC Lead Entities and Partners, Percentages Before WPC, PY 3, and PY5 Exhibit 26: Partners’ Average Perceived Effectiveness of WPC in Achieving Goals, PY 3 and PY5 Exhibit 28: Frequency of Data Sharing Agreements with Lead Entity and Specific Types of Key Partners): PY 9 sesissscescccssiersesanssivsveevannwasceaanscavaaaanvassuveaseavtcentcavausaveanrensceseeveaciaa naauscaacaeiecauateNieesns 84 Exhibit 29: Frequency of Use of Universal Consent Form for Data Sharing by Key Partner Type, Exhibit 30: Selected Examples of Data Sharing Agreements and Enrollee Consent in WPC, PY 685 Exhibit 31: Platforms and Tools Used to Support WPC Data Sharing, PY 5.0... cece 87 Exhibit 32: Type of Staff or Partner and Access to Care Management Platform and Event-Based Notificatiotis,. PY S isissceciiavscessvessnsamaveiieiniereuenssnassmmnanmimnn nannies 87 Exhibit 33: Type of Data Accessible to Care Coordination Staff, PY 5 .....csssssseeseesesesseaeeeseeeee 88 UCLA Center for Health Policy Research ee Tyr) Health Economics and Evaluation Research Program Exhibit 34: Selected Examples of Data Sharing Tools and Platforms to Support Care Coordination 1 WRC, BY 6 ssscccwss cs sisvsreeascannsssevsranisencanccennnntnnu naman taraN tee 89 Exhibit 35: Data Sharing and Reporting Challenges Among WPC Pilots by Program Year, PY 2 — Exhibit 37: Most Common Strategies for Identifying Prospective Enrollees and Pilot Perceived Effectiveness, PY 6 ..ccccssesseseseesssensessessessensessssnsessessecsseeseessesnsesssesssesssesssessesesseseesseasseasseaseees 97 Exhibit 38: Selected Examples of WPC Pilot Strategies to Identifying Prospective Enrollees...... 98 Exhibit 39: Method for Determining WPC Eligibility Following Identification of Prospective Enrollees, PY 6... . 100 Exhibit 40: Method for Determining Eligibility for WPC within Primary Target Population, PY 6 Exhibit 41: Pilot Perceived Effectiveness of WPC Enrollment Method, PY 6 ......csscseseeeeeeeee 102 Exhibit 42: Selected Examples of Strategies for Engagement of WPC Enrollees .........scceeeeeeee 104 Exhibit 43: Timeline of the Start of WPC Enrollment by Pilot, PY 2 to PY 3 oo... ceeseeeeteeeeeee 107 Exhibit 44: Unduplicated Monthly and Cumulative WPC Enrollment, PY 2 to PY 6... 108 Exhibit 45: Total Enrollment in WPC by Pilot, PY 2 to PY 6.0... 109 Exhibit 46: Patterns of Enrollment and Disenrollment in WPC, PY 2 to PY 6.......cccseseeeseeeeeeeee 110 Exhibit 47: Length of Enrollment of WPC Enrollees, PY 2 to PY 6 ....sesssesssessseesesseseteeeeeeeeeees 111 Exhibit 48: Quarterly Disenrollments from WPC, PY 2 to PY 6....ccccsscssesssessscsesesseeeseeeeeseeteeees 112 Exhibit 49: Reason for Disenrollment from WPC, PY 2 to PY 6.......:cccesseeseseeeeseeeeseseeeeseeeeseeeeeeeee 113 Exhibit 50: WPC Pilots Reporting at Least Ten Enrollees by Target Population, PY 2 to PY 6....114 Exhibit 51: WPC Enrollee Target Population Classifications, PY 2 to PY 6......c.sscssesseseeteeeeeeee 116 Exhibit 52: WPC Length of Enrollment in Months by Target Population, PY 2 to PY 6... 117 Seer) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 53: WPC Services Offered by Pilots as Of PY 5 o.....cececssseeeseseeeeeeetseeeeeeteeseteeaseetseneeeeaee 119 Exhibit 54: Percentage of WPC Pilots Offering Each Service Group .....ccseesesesseteeteeteeeneeees 120 Exhibit 55: Number of Bundles (PMPM) and Individual (FFS) Services Offered by WPC Pilots, PY 2to PY6 «A241 Exhibit 56: Outreach Services Delivered to WPC Enrollees by Enrollment Status and Target PoptilatiGn-PY 26. PY: 6 wesmcsvesesseessvescesneasrennnnremunenerareninnin arene eT 123 Exhibit 57: Care Coordination Services Delivered to WPC Enrollees, Overall and by Target Populationy:PY'2:t0: PYG ssssscvssscasscesvcssvesvesescevecesveventvavassteontxervesriswonsevisvcasaienaessatecasoveusvavaaaveasanaese 124 Exhibit 58: Estimated Delivery of Housing Support Service to WPC Enrollees, Overall and by Target Population, PY 2 to PY 6 v..csesssesssesesesesssesssenssesssesssessacssaesseesenensaeaeasscesseesseeeenseenereneeeree 125 Exhibit 59: Selected Examples of Housing Support in WPC.........ccccsceeesceseeteeseseeseseeeeseeeteeseeeenee 126 Exhibit 60: Benefit Assistance Services Delivered to WPC Enrollees, Overall and by Target Population, PY 2 to PY 6. .127 Exhibit 61: Selected Examples of Benefit Assistance Services in WPC Exhibit 62: Employment Assistance Services Delivered to WPC Enrollees, Overall and by Target Population, PY 2 to PY 6.0... .ccceccseeseeeesceseseseeseseeseseecseseceeseceeseeassesecaeeessesecaeseeesetecaeseesseesaeetaeees 128 Exhibit 63: Sobering Centers Services Delivered to WPC Enrollees by Enrollment Status and Target Population, PY 2 to PY 6 ...ceccccccceseesssesseseneeseseeseseeceseeecseeeaenensceeceeseeesseeecseeeasesseneeeenee 129 Exhibit 64: Selected Examples of Sobering Center Services in WPC ......:.ssscsseseeeeereeeterseseeteeeee 130 Exhibit 65: Medical Respite Services Delivered to WPC Enrollees by Enrollment Status and Target Population; PY 2:06. PY.6 vecccsssenseccnsnsm ences nncesmmanusnnerinannnmnrame neni ie TY 130 Exhibit 66: Transportation Services Delivered to WPC Enrollees by Enrollment Status and Target Population; PY 2:10 PY 6 weccssssssscesiscsisanisvsesensevessvusawnsevosnscusvassessncssiaverieocsterasetasarsseteauesasssereaaeieet 131 Exhibit 67: Health Education Services Delivered to WPC Enrollees by Enrollment Status and Target Population;:PY 2t0:PY 6 .ccssssssisosssesussessseersevessivsvesiuovnieseanasseniuevieaiaveraneaeneenanaenees 132 Exhibit 68: Selected Examples of Health Education Services in WPC........ccssssseseeteteteteteeeteees 133 UCLA Center for Health Policy Research . * December 2022 Health Economics and Evaluation Research Program Exhibit 69: Estimated Delivery of Legal Service to WPC Enrollees by Enrollment Status and ‘Target Population;.PY 206: BY.6 scsi eervossresnnarmcennsnrnannnanaemnanmamneiae 134 Exhibit 70: Estimated Delivery of Re-entry Services to WPC Enrollees by Enrollment Status and Target Population). PY 2:0 PY G:scssssssvcervesressvacnvesvenssvessvesnieveiasnnsssxeniionvssitvanuesnaevevsvesaeaveressencests 135 Exhibit 71: Individuals Receiving Services through WPC without Enrollment by Pilot, PY 2 to PY 6 Exhibit 72: Program Year and Overall WPC Payments to Pilots, PY 1 to PY 6 ......csssseseeeeeeee 137 Exhibit 73: Proportion of Overall WPC Payments to Pilots by Budget Category, PY 2 to PY 6... 137 Exhibit 74: Proportion of Total WPC Services Payments under PMPM and FFS Reimbursement Methods by Pilot, PY 2 to PY 6.......cssscsssessssssssssssnesstenseeessessvesseesseesseseseasseaseeasacasaenseensasasaeneaeeeeeee 139 Exhibit 75: Average Overall Payment for Services per WPC Enrollees by Enrollment Status and Target Population, PY 2 to PY6.. .140 Exhibit 76: Average Monthly Payment per WPC Enrollees Receiving Services for WPC Services Overall and by Target Population, PY 2 to PY 6.....ccssccssessesstsesteesseesseesseeseetssesseesseasseesseseaeeeaeees 141 Exhibit 77: Number of WPC Pilots Participating in Select Data Sharing Capabilities to Support Care Coordination, PY 3 and PY 5 ......ccccccccscsssessesscssssccsscssessessssaessecsecsessscsecaessesseeseseasenssaeseeers 144 Exhibit 78: Information Systematically Collected as Part of Needs Assessment Process in WPC Exhibit 80: Number of WPC Pilots Engaging in Selected Strategies to Increase Care Coordination ME=tlaamaceoolblale-|ol lin armen meevecercrerecerrrrer reer ere creer cer erm rrtrer tern rrenrerrecry crrercererer etree 149 Exhibit 81: Types of Staff Involved in WPC Care Coordination by Pilot..........c.csssesseeseereseseeeee 150 Exhibit 82: Resources in Place to Support Staff Responsible for Care Coordination .................151 Exhibit 83: Commonly Identified Challenges in Care Coordination Among WPC Pilots, by Reportifig' Period; PY:2 16 PY 6 sncncncsmeseraneesaeennn meena 153 Sele) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 84: Commonly Identified Successes in Care Coordination Among WPC Pilots, by RepOrtirig Period, PY:2:10) PY Gv ssesesscenmeecnnnmnnanenenanmannTenencTeE 155 Exhibit 85: WPC PDSA Category Types Across Reporting Periods, PY 2 to PY6.. .158 Exhibit 86: Selected Illustrative Examples of WPC Quality and Performance Improvement and MOHItOFNEACtVIEGS? science aici moe aie aaa eas 159 Exhibit 87: WPC Pilots’ Rating of Frequency of Involvement of Stakeholders in Aspects of Quality Improvement Activities isis sc siesaneerinienien aivieenmieniiaiee Rei eee ene 161 Exhibit 88: WPC Pilots’ Rating of Extent of Stakeholder Influence on Quality Improvement ACUIVIRIES ss cessvssseseessxapexapessiasuncasssesxxessiasonsaueeysxeeewa8aun5 vee eae eig TAR Ova aN aN SRDNREU ARATE NAA TORI ORT UTTR 162 Exhibit 89: 14-Day Average Daily Confirmed COVID Cases and Hospitalizations per 100K for WPC Counties, April 2020 to December 2021. . 164 Exhibit 90: WPC Informing or Impacting COVID-19 Response by Program Element, PY 5........165 Exhibit 91: Reports of WPC Informing or Impacting COVID-19 Response by Program Element AN Extent, PY 5S ....ccccescsseesseseesessesssessssescsesssssssesssensanssecscsssssssesusensaesseesecssessessasssnsensseeseessesseneses 166 Exhibit 92: Illustrative Examples of How WPC Informed or Impacted COVID-19 Response...... 169 Exhibit 93: Pilot Reports of COVID-19 Impact on WPC Processes, Procedures, or Policies, PY 5 Exhibit 95: WPC Pilots Reporting Enrollees in COVID-19 Target Population... 173 Exhibit 96: Monthly Enrollment and Total Quarterly New Enrollment in WPC, January 2017 to December 2021......c.cccsccssessssssesesssseseessssesessseseesssessssssesssesseessssseassessssssesusssseeusasseeaeseseasseansesseeeses 174 Exhibit 97: Characteristics of WPC Enrollees at Baseline Enrolled Before and During the COVID- 19 PandeMexanssnmnnmenreNa ONS 175 Exhibit 98: Rate of COVID Diagnosis per 1,000 Medi-Cal Member-Months for WPC Enrollees and their Controls from April 2020 to December 2021.......c.sssssesseesseesseesseesvesseeeseesseessessevesseaenensaeaes 176 UCLA Center for Health Policy Research ele) Health Economics and Evaluation Research Program Exhibit 99: Proportion of COVID-19-Related Health Services by Service Type among WPC Enrollees and their Controls with a COVID-19 Diagnosis .......c.cccesseesseeeeteeseseesteteeeeeeteeeeeeeee 177 Exhibit 100: Monthly Utilization of Primary Care and Specialty Care Services per 1,000 Member Months among WPC Enrollees and their Controls, 2019 Compared to 2020 and 2021............ 178 Exhibit 101: Monthly Utilization of Emergency Department Visits and Hospitalizations per 1,000 Member Months among WPC Enrollees and their Controls, 2019 Compared to 2020 and 2021 Exhibit 102: Proportion of Primary Care and Specialty Services that were Provided through Telehealth for WPC Enrollees, 2019 to 2021. . 180 Exhibit 103: Commonly Identified Challenges and Successes Related to the COVID-19 Pandemic arriong: WPGC Piléts,, PY5—PY!6 scccsecnsen anerncnmnen ce aremernmenmieannmenmivenamven eure 182 Exhibit 104: Demographics of WPC Enrollees Prior to WPC Enrollment.........:ceceeeeteeeeeeeeeeeee 184 Exhibit 105: Most Frequent Chronic Conditions Among WPC Enrollees, 24 Months Prior to WPC EMroll Ment ......ccccesecseseceeseseeeessseescseeeescseeescseseesssessesssesssseseeasscseeassesesassesssesseasassneassesssessseneaserenses 185 Exhibit 106: Selected Ambulatory Care Service Use per 1,000 Medi-Cal Months Among WPC Enrollees in Months Prior to WPC Enrollment .........cssssesesssssssesssestessseessssseenssessesssesessseeensseentes 186 Exhibit 107: Emergency Department (ED) Visits Followed by Discharge per 1,000 Medi-Cal Member Months Among WPC Enrollees in Months Prior to WPC Enrollment, Overall and by SPECITIg: CONITIONS).c+.c.rrsservoreemaxcnarnonnnnesenenvensoesnotuncnnonsncanousatinn soa esototoasstonecanncties envenesocenennunescennens: 187 Exhibit 108: Number of Hospitalization per 1,000 Medi-Cal Member Months Among WPC Enrollees in Months Prior to WPC Enrollment, Overall and by Specific Conditions ................ 188 Exhibit 109: Trends in Primary Care Services per 1,000 Beneficiaries per Year Before and During WPL, PY 2 - PY 6..cccccssccsscsescsssssseessececscsecscsessssesessnscsssessssasscsesessecsseasscsssessseasessesassasecsasessseacisenes 191 Exhibit 110: Trends in Specialty Services per 1,000 Beneficiaries Months Before and During WPL, PY 2 - PY 6... ccccccccscssescsesecscesssesecsesecscsececsesecsesecsesecscsasscseeesseeacsceesesesseessseseeessaeetseseeaeeaneeeees 192 Exhibit 111: Trends in Mental Health Services per 1,000 Beneficiaries per Year Before and During WPC, PY 2 - PY6 Sey) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 112: Trends in Mental Health Services per 1,000 Beneficiaries per Year Before and During WPC, PY 2 - PY 6, by Subpopulations Exhibit 113: Trends in Substance Use Disorder Services per 1,000 Beneficiaries per Year Before and During WPC, PY 2 - PY6 Exhibit 114: Trends in Substance Use Disorder Services per 1,000 Beneficiaries per Year Before and During WPC, PY 2 - PY 6, by SUbpopulations .......c ec ecesceeseeeeseeseeeteeseeeeeeseeesseeeraeeesaseeeaeeees 196 Exhibit 115: Trends in Follow-Up After Hospitalization for Mental Illness within 7 Days Before and During WPC for WPC Enrollees and the Control group, PY 2 - PY 6.....cscsssssssesssesseesesreseeee 197 Exhibit 116: Trends in Follow-Up After Hospitalization for Mental Illness within 30 Days Before and During WPC for WPC Enrollees and the Control group, PY 2 - PY 6 .198 Exhibit 117: Trends in Initiation of Alcohol and Other Drug Dependence Treatment Before and During WPC for WPC Enrollees and the Control Group, PY 2 - PY 6......:csccssessseeseeeeeeseeeeeeeeeeeee 199 Exhibit 118: Trends in Engagement of Alcohol and Other Drug Dependence Treatment Before and During WPC for WPC Enrollees and the Control Group, PY 2 - PY 6 ...cssessessssssesesseeseeneseeee 200 Exhibit 119: Pilot-Reported Universal and Variant Metrics That Indicate Better Care..............201 Exhibit 120: Percent of Enrollees Who Received a Comprehensive Care Plan Within 30 Days of Enfollinenit,: by Prog fait Veal sceccssccccasecsersucesennmaenenevreeeraninenaennnectenuumamrecnieeeE? 202 Exhibit 121: Percent of Enrollees Who Received a Comprehensive Care Plan Within 30 Days of the Anniversary of their Enrollment, by Program Year ..........ccsccsssessesssessesesesessseesssesnesesesneneeee 203 Exhibit 122: Percent of Adult Enrollees with a Diagnosis of Major Depressive Disorder That Received a Suicide Risk Assessment During the Visit in Which a New Diagnosis or Episode was Identified, bY PROSTaih XCar asec cee creer ween senna i ne er erenY 204 Exhibit 123: WPC Pilot Perceptions of Impact on Aspects of Better Care, PY 5... cesses 206 Exhibit 124: Trends in Ambulatory Care: Emergency Department Visits per 1,000 Beneficiaries per Year Before and During WPC, PY 2 - PY 6.0... cccccccsssesseeseeeseseseeeseessessecseasssacssecseaeseeeeeeeeseee 209 Exhibit 125: Trends in Ambulatory Care: Emergency Department Visits per 1,000 Beneficiaries per Year Before and During WPC, PY 2 - PY 6, by SUDPOPUIAtiONS.........eesseseeteeteteteeeteteteeee 210 UCLA Center for Health Policy Research ee Tyr) Health Economics and Evaluation Research Program Exhibit 126: Trends in Inpatient Utilization per 1,000 Beneficiaries per Year Before and During WPC, PY 2 - PY6 Exhibit 127: Trends in Long-Term Care Stays per 1,000 Beneficiaries per Year Before and During WPC, PY 2- PY6 Exhibit 128: Trends in Controlling High Blood Pressure Before and During WPC for WPC Enrollees and the Control Group, PY 2 - PY 6... ceccesssssseseeseneeeeeeseeeeesesecseseeecsesseaenetaseeteneeeeees 213 Exhibit 129: Trends in HbA1c Testing Rates Before and During WPC for WPC Enrollees and the CONS] GrOup, PY 2 PYG isersvevccxoscnoansensnonsscanovanvonvarsnonsennsonsnuenonnnse conus scene denned cexnsabonosensontnenennase: 214 Exhibit 130: Trends in All-Cause Readmission following an Acute Inpatient Admission, Before and During WPC for WPC Enrollees and the Control Group, PY 2 - PY 6 .215 Exhibit 131: Pilot-Reported Variant Metrics That Indicate Better Health.. .216 Exhibit 132: Number of Incarcerations per 1,000 WPC Member Months, by Program Year ....219 Exhibit 133: Percent of Enrollees Who Reported “Excellent” or “Very Good” Overall Health (OBH-O), by Year .ccccccssscsssssesssssssssssesssssssssssesesesssssessesessssueseseesasssssesseessssuvssesesssssesesesenssseseeeeeessssees 220 Exhibit 134: Percent of Enrollees Who Reported “Excellent” or “Very Good” Emotional Health (OBH-E), Dy VEAP... ssessesesesssesesssesesesesesesesesessesscscscecuessvesssesssesescsescseseacseseasscseaeasseseacaeaeaeaenenseeaeee 221 Exhibit 135: Percent of WPC Enrollees 18 to 59 years old with Controlled Blood Pressure, by Program YOAP oo... eecseseeseeseeneeseeseeseeeseeseeaeenassecneceeeeseesecsesaeeseeeeesseeseesecaaeeeeeeassesseeseseaseeeeaseaeeaeentee 222 Exhibit 136: Percent of WPC Enrollees 60 to 85 years old and Diabetic with Controlled Blood PRESSUFE BY RROBFANT VOOM sisesscesscovesscevavssesiseucesaszavesaceosagusvapauennsaaaaubanurasaiabonssarasva sts tonaanaaiaavannr’ 223 Exhibit 137: Percent of WPC Enrollees 60 to 85 years old and not Diabetic with Controlled Blood Pressure, by Programm Veal acereser ses neese een weer rere aren ere ener nenmee 223 Exhibit 138: Percent of Adult Enrollees with Diabetes Who Had Controlled HbA1c, by Program Exhibit 139: Percent of Enrollees Age 18 or Older with Major Depression or Dysthymia Who Reached Remission at 12 Months, by Program Y@ar..........s:ssscsssessesssessesesesssesrenssesescseeneseereneee 225 Exhibit 140: WPC Pilot Perceptions of Impact on Aspects of Better Health, PY 5 oo... eee 226 Sey) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 141: Trends in Total Estimated Medi-Cal Payments Before and During WPC, PY 2 - PY6 Exhibit 142: Trends in Total Estimated Medi-Cal Payments Before and During WPC, PY 2 - PY 6, by Subpopulations Exhibit 143: Trends in Estimated Medi-Cal Payments for Outpatient Services Before and During WRG PY 2S PVG iavsscscsusccsacuseranesussennsuneanevenpnean na eiascinveueaauemneens aveomnenenveetenevercectammerersest peat 231 Exhibit 144: Trends in Estimated Medi-Cal Payments for Outpatient Services Before and During WEE, PY¥'2.=PY'6, by SUDPOPUlAtlONS!..cor-scsovasvonsarsnonsennsonsnnenonsnsecenurssceandannedeenensavonosensonsseesonvene 232 Exhibit 145: Trends in Estimated Medi-Cal Payments for Outpatient Medications Before and During WPC, PY 2- PY6. . 233 Exhibit 146: Trends in Estimated Medi-Cal Payments for Outpatient Medications Before and During WPC, PY 2 - PY 6, by Subpopulations .........cceeeeeseeseeteessseeseneeeeseeeceeseeseseeeaeeeseeetatseeatees 234 Exhibit 147: Trends in Payments for Emergency Department Visit Before and During WPC, PY 2 - Exhibit 148: Trends in Estimated Emergency Department Payments Before and During WPC, PY 2 - PY 6, by SUBpOpUlations oo... ee eeeeseeseeteeseteeseseeeeseseeseseeseneeacsesecaesesseeceeseeesseeetseneeaseetetseeaeeee 236 Exhibit 149: Trends in Payments for Hospitalizations Before and During WPC, PY 2 - PY 6......237 Exhibit 150: Trends in Estimated Medi-Cal Payments for Long-Term Care Stays Before and During: WRC, PY:2 =: PY 6 viscsssisssssesussssessscnssnssenravesnesnoconmasvarasoveusousstaanenvassavessarenssavarsoresnasetiees 238 Exhibit 151: Trends in Estimated Long-Term Care Stays Before and During WPC, PY 2 - PY 6, by SUDPOPUlAtionS .......ccesseseseseseseseseessescsescsescsesesescecscecscecscacscesseasscusseseeeseeeseessestseseeeaseeeseeaseeateeees 239 Exhibit 152: Trends in Estimate Medi-Cal Payments for Residual Medi-Cal Before and During WEE PY 2 PV 6. core csrneues cxtananennenaunennaaaserenee eter nasa casauavnenneunoserapeneuatensvaynsteossheueseeinarnctantesenpanarvsen 240 Exhibit 153: Selected Examples of Outreach Approaches for Individuals Experiencing Homelessness in WPC........cccccsseseessseescseeesseseseesssessesssssssssseessecseasscsssassesssassesasassesesscsesassnseeeeereneee 244 Exhibit 154: Type of Housing Support Service(s), Provided by Lead Entity or WPC Partner Organization, Using WPC Funds or an Alternative Funding Source, PY 5 .......secseeeeeeeeseeeeeeeees 246 UCLA Center for Health Policy Research Peele) Health Economics and Evaluation Research Program Exhibit 155: Type of Direct Housing Services and Resources Provided by Lead Entity or WPC Partner Organization, Using WPC Funds or an Alternative Funding Source, PY 5 wo... .eeeeeeeee 247 Exhibit 156: Pilot Participation in Activities to Promote Community, Policy, and/or Systems Change Related to Homeless Assistance, PY 5 .....cssssssssssssssreesseesseessesssenssesssesseessessseasseseaenenenes 249 Exhibit 157: Participation of Lead Entity with Housing Agency in Select Collaboration Activities, Exhibit 158: Unduplicated Monthly and Cumulative Total WPC Enrollment among Enrollees Experiencing Homelessness, January 2017 to December 2021 .........cscssssessssessessesseeesessesesseneee 253 Exhibit 159: Total Unduplicated Enrollment in WPC by Pilot among Enrollees Experiencing Homelessness, December 2021.. .254 Exhibit 160: Percent of WPC Enrollees Experiencing Homelessness by Pilot, January 2017 to December 2021......cccscccssessssssesessesseessesseeesssesessssesssssesssssseesssssesssesssassessssssnsssseseeasseseesseensesssentee 255 Exhibit 161: Length of Enrollment in WPC Among Enrollees Experiencing Homelessness, January 2017 to December 2021 ......cecscccessssesseseseeseseseeseseseesssesssssseessssseesssesssassesssasssessasseaesssseesessneesereneee 256 Exhibit 162: Demographics of WPC Enrollee Experiencing HOMEleSSNESS ..........sssesesereseneeeteeee 257 Exhibit 163: Proportion of WPC Enrollees Experiencing Homelessness with Chronic Conditions anetasonietins steamer anunsrennaaN imbibe miteaconntrmNronRN ana eTsteacucerimiea canon Ieee eRtmnnTE ITER 258 Exhibit 164: Proportion of WPC Enrollees Experiencing Homelessness and Not Experiencing Homelessness That Received WPC Services, PY 2 to PY 6......ccsesesssesssesssessecsseeseeseseeeneeeeeee 259 Exhibit 165: Housing Metrics Selected by WPC Pilots .........cssecssessesssesssesseesseeseeeseaeseeesceeeeeeesees 260 Exhibit 166: Proportion of Enrollees Formerly Experiencing Homelessness in Permanent Housing Who Reached the Seventh-Month, by Program Year .........:cccccsseesesseseseeeeseeeeeeseeecseeetseteeseeeteenees 261 Exhibit 167: Proportion of Homeless Enrollees Who Received Housing Services After Being Referred for Housing Services, by Program Y@aP.........s:sssssssseseesssessesssesesessesssesseaesesresessseseseeeee 262 Exhibit 168: Proportion of Homeless Enrollees Who Received Supportive Housing after Being Referred, by Prograrn Yeal sississsisssiiscsnssrrnaeunwansanmnaninaninimnmnininaecet 263 Reyes) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 169: Trends in Ambulatory Care: Emergency Department Visits per 1,000 Beneficiaries per Year among WPC Enrollees Experiencing Homelessness Before and During WPC, PY 2 - PY6 Exhibit 170: Trends in Inpatient Utilization per 1,000 Beneficiaries per Year among WPC Enrollees Experiencing Homelessness Before and During WPC, PY 2 - PY 6.......sscssesessenesesrereee 266 Exhibit 171: Trends in Mental Health Services per 1,000 Beneficiaries per Year among WPC Enrollees Experiencing Homelessness Before and During WPC, PY 2 - PY 6......ceececessteeseteeeeeee 267 Exhibit 172: Trends in Substance Use Disorder Services per 1,000 Beneficiaries per Year among WPC Enrollees Experiencing Homelessness Before and During WPC, PY 2 - PY6. . 268 Exhibit 173: Trends in Follow-Up After Hospitalization for Mental Illness within 7 Days among Enrollees Experiencing Homelessness Before and During WPC for WPC Enrollees and the Control group, PY 2 - PY Go. cecsceeeccseessseeseeeeeeseeecseeecaeseeseseeeeseeessesecaeecsesenseseteeseeesseeetaetetseeataesees 269 Exhibit 174: Trends in Follow-Up After Hospitalization for Mental Illness within 30 Days among Enrollees Experiencing Homelessness Before and During WPC for WPC Enrollees and the Control group, PY 2 - PY6. .270 Exhibit 175: Trends in Initiation of Alcohol and Other Drug Dependence Treatment among WPC Enrollees Experiencing Homelessness Before and During WPC, PY 2 - PY 6.......ceceesseeseteeeeeee 271 Exhibit 176: Trends in Engagement of Alcohol and Other Drug Dependence Treatment among HHP Enrollees Experiencing Homelessness Before and During HHP by SPA, PY 2 - PY 6........... 272 Exhibit 177: Trends in All-Cause Readmission following an Acute Inpatient Admission, Before and During WPC for WPC Enrollees and the Control Group, PY 2 - PY 6 ......csessesssesssessseeseeees 273 Exhibit 178: Timeline of Key Dates and Activities for WPC Transition to CalAIM .......sseeeeeeeee 277 Exhibit 179: Lead Entity Participation in Transition Planning Meetings with DHCS, Medi-Cal Managed Care Plans, and Other WPC Partners, August 2020-May 2021 ........ccccscessesseeseeeee 278 Exhibit 180: Populations of Focus, Served through Enhanced Care Management and Whole Person Care, May 2022). ..ssssensevcassssonsessoncssnananesversinanconasvavaasvusvscsisusmassesmanavenssersssnnvecuansoursreey 282 Exhibit 181: Participation of WPC Pilots in Selected Community Supports by County, May 2022 UCLA Center for Health Policy Research ee Tyr) Health Economics and Evaluation Research Program Exhibit 182: Commonly Identified Challenges and Successes in Transition to CalAIM among WPC PilOtS, PY 6 vsscscccisccssesvsasnsenneseuevenssssinesiveviscnisevessviesrtscuceawacacontteuniacarioiatatabserutsavtsesaeecstenenNiNeN 287 Exhibit 183: Illustrative Examples of Plans to Sustain WPC Data Sharing Infrastructure under CAIAIM sessssssccsssvsssesscsssasvsavivasveavcasvacrsnsassnsuvavecsccrssiesesvaasistsstsasdaveasviausaaneasscaaseassassievesuiaasieeeeaN east 289 Exhibit 184: Beneficiary-Level Variables........cccssesscessessseeseesseessesssessseassessecssecseacaeeeneaeaeaeneeeeenees 301 Exhibit:185: Demographic NdiCAtOMS o..cscccnsseeceveceseveonssresevessnierenereaiuin aavermuaversnesies 302 Exhibit 186: WPC Metrics, Definitions, and Intended Direction ...........ccccscsesesseeeessesseseetsesees 303 Exhibit 187: Healthcare Utilization Indicators ........scsscssesseeseeseesesseseseseesecsesssessssesseseseessesesssenees 305 Exhibit 188: Variables Used to Select the Control Group ......sssesssesesesssesssenssesseesssensaeeeeeeneee 306 Exhibit 189: Comparison of Select Characteristics of WPC Enrollees with Two Years of Baseline Data and Matched Control Beneficiaries.........c.ccssssessessseessessesssesesssessessssesssestesssesieesseseesseenees 308 Exhibit 190: Description of Mutually Exclusive Categories of Service® 0... 312 Exhibit 191: Percentage of 2019 Total Estimated Payments by Category of Service for WPC Medi-Cal Claims cassscsscssssssssscsssvessocsssuovssissrsonsecassaesssvecsonuasvaivecussissuasscassvansivoctstoatbsensvaatsavcesseestests 314 Exhibit 192: Category of Service and Payment DeSCriptiOns..........:ccsesssesseseeeceteaeeeeteeeeteeees 315 Exhibit 93: Payrriéiit! Data! SOUTCES ncsescsesensverancamesnenesresenniamenr asian atemeventineatT 316 Exhibit 194: Comparison of Estimated Fee-for Service Payments and Paid Amounts for 2019 WPC Medi-Cal Claims.. -322 Exhibit 195: Comparison of Average Fee-for-Service and Managed Care Payments per Claim for 2019 WPC Medi-Cal Claiinis.isenisesossssvsenssscvssvsvsanrsesveavenssnsesenevscuvatentiasannsnsvetunstiunnesniveonesdseaveovnscbeve 322 Exhibit 196: Reporting for Variant Metric: Control Blood Pressure, Age 18-59.......:ccsseeseeee 327 Exhibit 197: Reporting for Variant Metric: Control Blood Pressure, Age 60-85, with Diabetes 329 Exhibit 198: Variant Metric: Control Blood Pressure, Age 60-85, without Diabetes ........0000. 331 Exhibit 199: Reporting for Variant Metric: Incarcerations per 1,000 Member Months ............ 334 Exhibit 200: Reporting for Variant Metric: Overall Beneficiary Health - Overall Health............ 337 Sele) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 201: Reporting for Variant Metric: Overall Beneficiary Health - Emotional Health....... 339 Exhibit 202: Reporting for Variant Metric: Comprehensive Diabetes Care... 342 Exhibit 203: Reporting for Variant Metric: Depression Remission at 12 Months ........:.seseeeeee 345 Exhibit 204: Reporting for Variant Metric: Major Depressive Disorder - Suicide Risk Assessment - 348 Exhibit 205: Reporting for Variant Metric: Permanent Housing 351. Exhibit 206: Reporting for Variant Metric: Housing Services . .354 Exhibit 207: Reporting for Variant Metric: Supportive Housing. .357 Exhibit 208: Reporting for Universal Metric: Comprehensive Care Plan - Within 30 Days of Enrollment Exhibit 209: Reporting for Universal Metric: Comprehensive Care Plan - Within 30 Days of Twelve-Month Anniversary of Enrollment .......ccccccccescssssesceeeseeseseeseseeeseseceeseeesseecseseeaseesseeeeeeee 362 Exhibit 210: Codebook Used for Preliminary Coding of Follow-up Interviews, PY 6.........:.:005 371 Exhibit 211: FFS and PMPM Categories, Associated Services, and Associated Annual Rates, 2017 MO OD reset secs settee hs cea teesee ea net eel area snare ator hs See ceneteeaca tte Schama et as 378 Exhibit 212: Primary Target Population by Pilot........ccceceececeteeseseeseseeeeseeeceeseeeeseesesetetaeeetetseeeeees 391 Exhibit 213: Enrollee Target Populations Reporting by WPC Pilot, PY 2 to PY 6.0... 392 Exhibit 214: Alameda WPC Pilot Target Populations.......0....cccccccesseeesceseseeeeseeeseseteeaeeeteeseeeeaee 394 Exhibit 215: Contra Costa WPC Pilot Target Populations ..........:ccsessesssessesseseeeeeeeieteteeseeeees 395 Exhibit 216: Kern WPC Pilot Target Populations ........... ce esesseesseescsessesesesecsseecseseseeseeasseseveeeeeees 396 Exhibit 217: Kings WPC Pilot Target Populations............ceecesessseeseeeeeseeeeseseeeeseeeraceeeaeeetenseeeeees 397 Exhibit 218: Los Angeles WPC Pilot Target Populations ......c cee eeeeeneeeeeeeeeeee 400 Exhibit 219: Marin WPC Pilot Target Populations 0... cece eeseeeeneeeseeeeeeeeeeaeeee 401 Exhibit 220: Mariposa WPC Pilot Target PopUulations.........ccsecsecssecssesssesssesseesesessecseeeseeeeeeeee 402 UCLA Center for Health Policy Research ee Tyr) Health Economics and Evaluation Research Program Exhibit 221: Mendocino WPC Pilot Target Populations. ........ccccceseeeesseeeeteeseteeseeeteeseeetseeeeeeee 403 Exhibit 222: Monterey WPC Pilot Target Populations... eeeeeeseeseeseteeeneeeeeeneee 404 Exhibit 223: Napa WPC Pilot Target Populations......cccscesessesssesssesssessseesscssssssscnsesnseensseeeeenees 405 Exhibit 224: Orange WPC Pilot Target Populations ........ssscsssesssesssesssesssesssessecsesensacacecaeeeneesneneee 406 Exhibit 225: Placer WPC Pilot Target Populations ..........cssceseesseeseereeseeetseneeeeseestseeeassesseseeneees 407 Exhibit 226: Riverside WPC Pilot Target Populations .........cccccccseeeeseeeeseeetseteeeeeeseseteesetetseeeeenee 408 Exhibit 227: Sacramento WPC Pilot Target Populations......cccsccsesesseseeeeeeeeeeeeeeeeee 409 Exhibit 228: San Benito WPC Pilot Target POpUlations........ccssecssesssesssesssesssesescsseseneneeeeeenees 411 Exhibit 229: San Bernardino WPC Pilot Target Populations .........ccsesessseesessseeeneeseeteeeeeeee 412 Exhibit 230: San Diego WPC Pilot Target Populations ...........ccecsesseseeeeseeetseeeeeeeseeeteeasteteeeeeenee 414 Exhibit 231: San Francisco WPC Pilot Target Populations ......ccceseeeeseeseeeeeeeeeeeeee 415 Exhibit 232: San Joaquin WPC Pilot Target Populations...........cccsccsssesessseeesessesesesresesseeeeseeneee 416 Exhibit 233: San Mateo WPC Pilot Target Populations...........:sscssesssesseseseessessesesesnesesseeseseeneee 417 Exhibit 234: Santa Clara WPC Pilot Target Populations........cccssessesssesssenseessecsseeseaseneteeteeeeeees 418 Exhibit 235: Santa Cruz WPC Pilot Target Populations .........cccseeeseseeseeseseteeeeseeseseteeaeterseseeeenee 419 Exhibit 236: Shasta WPC Pilot Target Populations ..........c.cccccccessseeseseeeeseeeeeeeeeeseeseseeeaeeeeseseeaeees 420 Exhibit 237: Solano WPC Pilot Target Populations ...........cccecsssessesesessesesesssscssesssesseesseeeseseeeeee 422 Exhibit 238: Sonoma WPC Pilot Target POpulations...........cccecssessesssesseseseeesessenssesreeesesesesreneee 423 Exhibit 239: Ventura WPC Pilot Target Populations «20.0... csessseeseseeeeseeeeseseeeeseestseeeeaseetseneneenee 424 Exhibit 240: Selected Illustrative Examples of WPC PDSAs Submitted by Category Type, PY 4-PY Gents enneminracnanneennerneninmastiarin airmen nenmn arian tne Tne aRN RANE 425 Exhibit 241: Demographics of WPC Enrollees by Target Population, Prior to WPC Enrollment 566 ere UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 242: Most Frequent Chronic Conditions Among WPC Enrollees by Target Population, 24 Months Prior to WPC Enrollment .. - 567 Exhibit 243: Selected Ambulatory Care Service Use per 1,000 Medi-Cal Months Among WPC Enrollees by Target Population, Semi-Annually Prior to WPC Enrollment.........cccsseeseeseeseeeee 568 Exhibit 244: Emergency Department Followed by Discharge Visits per 1,000 Medi-Cal Member Months Among WPC Enrollees, Semi-Annually Prior to WPC Enrollment, by Target Population Exhibit 245: Number of Hospitalizations per 1,000 Medi-Cal Months Among WPC Enrollees, Semi-Annually Prior to WPC Enrollment, by Target Population. Exhibit 246: Participation of WPC Pilots in Community Supports by County... UCLA Center for Health Policy Research Beier Health Economics and Evaluation Research Program i Executive Summary WPC Program Overview The Whole Person Care (WPC) program was implemented under the “Medi-Cal 2020,” a Section 1115 Medicaid Waiver from January 1, 2016 to December 31, 2021 and was focused on high- risk, high-utilizing enrollees with multiple service needs. A total of 25 Pilots, representing the majority of counties in California, implemented WPC and started enrollment in January 2017. The overarching goal of WPC was to improve health and wellbeing by coordinating care across physical health, behavioral health, and social service sectors. Pilots consisted of 27 Lead Entities (LEs) with expertise and resources to implement the program and form a public private partnership. Pilots were required to target one or more of the following six populations: (1) high utilizers of avoidable emergency department, hospitals, or nursing facilities (high utilizers); (2) individuals with two or more chronic physical conditions (chronic physical conditions); (3) individuals with severe mental illness and/or substance use disorders (SMI/SUD); (4) individuals experiencing homelessness (homeless); (5) individuals at-risk-of-homelessness; and (6) individuals recently released from institutions, including jail or prison (justice-involved). In the third quarter of 2020, a seventh target population was added to include individuals impacted by or at-risk of COVID-19. The total budget for WPC was $3 billion, with the approved 5-year budgets for participating Pilots ranging from $7,247,500 (Solano County) to $1,572,976,930 (Los Angeles County). Evaluation Methods The UCLA Center for Health Policy Research was selected to evaluate WPC and developed a conceptual framework and evaluation questions to conduct a rigorous, state-wide, mixed methods assessment of the program. UCLA used all available data for the evaluation, including Pilot applications, Pilot-reported universal and variant metrics, monthly enrollment and utilization reports, bi-annual narrative reports, and Medi-Cal enrollment and claims data. UCLA also conducted multiple surveys of LEs and involved partners, as well as follow-up interviews with LEs and frontline staff in PY 3 and PY 6. UCLA used the qualitative data sources to examine the infrastructure developed by Pilots for WPC, implementation processes, and services delivered. UCLA used Pilot-reported metrics and Medi-Cal data to determine whether WPC led to better care, better health, and lower costs. Analyses of Medi-Cal data included comparison of selected WPC metrics as well as utilization and cost measures before and after WPC implementation for WPC enrollees and a control group of Medi-Cal enrollees with similar characteristics. Whole Person Care Final Evaluation Report | Executive Summary Gea beee UCLA Center for Health Policy Research a Health Economics and Evaluation Research Program Results Structure of WPC Pilots WPC aimed to “increase integration among county agencies, health plans, providers, and other entities with the participating county” to effectively “serve high-risk, high-utilizing beneficiaries.” WPC also intended to “develop an infrastructure that would ensure local collaboration among the partners participating in WPC Pilots over the long term.” Evidence indicated that WPC Pilots developed infrastructure needed to implement the program and coordinate health, behavioral health, and social services provided. This included significant investment in promoting meaningful partner engagement and buy-in (e.g., frequent communication, active role in shared decision-making, consensus on roles and responsibilities). These conclusions are supported by the following evidence: e@ Pilots chose LEs with the leadership and administrative capacity to effectively implement WPC. These LEs included county health and health services agencies (15 of 27), healthcare systems (8), behavioral health departments (3), and a city municipality (1). e@ Pilots reported an average of 21 partners per Pilot and a collective total of 543 across all Pilots. More than half of partners (58%) were community-based organizations. Most community partners were health care providers (33%), or provided either housing support or other community based social services (37%). e LEs reported increased partner involvement between PY 3 and PY 5. Total number of partners increased during this time. In addition, in PY 3, LEs identified 47% of partners as actively involved in WPC, whereas by PY 5, 67% of partners across all Pilots were actively involved. e@ Most LEs experienced challenges with partner buy-in during the first few years of the Pilot. Consistent communication, consensus on strategic priorities, and in some Pilots, providing financial incentive for participation were identified as factors facilitating partner buy-in. e In PY 5, partners rated WPC (on a scale of 0: “not effective” to 10: “extremely effective”) as effective at improving the management of high risk and high utilizing populations (average rating of 7.5 of 10), improving integration of health and social services (7.4), and improving collaborative partnerships for program implementation (7.4). All of these ratings increased from the interim report. Executive Summary | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err] Health Economics and Evaluation Research Program Health Information Technology and Data Sharing Infrastructure WPC aimed to “improve data collection and sharing amongst partners to support ongoing case management, monitoring, and strategic program improvements in a sustainable fashion.” Evidence indicated that over time, WPC Pilots succeeded in developing innovative data sharing infrastructure needed to support cross-sector care coordination and facilitating data sharing with partners. These conclusions are supported by the following evidence: e@ By PY5, 20 of 25 Pilots had data sharing agreements in place with all key partners and the other five had agreements with at least some key partners. These agreements were new as a result of WPC (e.g., only 4 of 27 Pilots reported in PY 3). LEs most often had data sharing agreements in place with Medi-Cal managed care plans (21 of 25) followed by health care providers (20) and mental health treatment agencies (18). e@ Most Pilots (19 of 25) expanded, acquired, and/or developed a care management platform to facilitate tracking of important enrollee-level data. Outside of the care coordination team, access to enrollee-level data through the care management platform was most commonly granted to staff in county health (15 of 19) and mental health service agencies (14); 16 Pilots also provided staff with real-time notifications of events (e.g., ED visits). e@ Ininterviews and narrative reports, LEs described significant investment in developing data sharing capacity and ensuring buy-in from partners. In PY 6, 18 LEs reported utilizing financial incentives in contracts with partners to promote development of data sharing infrastructure (e.g., to increase functionality of existing or newly acquired case management platforms or ensure reporting of desired data elements). These incentives were considered effective (average rating of 7.5 out of 10) at achieving desired goals. e@ Throughout WPC, the three most common data sharing and reporting challenges included (1) lack of buy-in and/or readiness from partners and frontline staff, (2) inability to access certain data, and (3) inability to implement data sharing systems and/or integrate data as intended. e Pilots most often found successes with (1) sharing data across multiple systems, (2) developing new software platforms and/or data repositories, and (3) using data to inform decision making. e@ In PY 5, LEs reported relatively high perceived impact of WPC on improving data sharing between the LE and partners (average rating of 7.9 out of 10). WPC Enrollment Size, Patterns, and Trends WPC Pilots were required to identify eligible Medi-Cal beneficiaries using pre-defined inclusion criteria, enroll them in WPC, and engage enrollees in care. Evidence showed sustained growth and significant cumulative enrollment with limited churn among more vulnerable groups of enrollees. These successes were likely due to use of innovative and tailored approaches to gain Whole Person Care Final Evaluation Report | Executive Summary UCLA Center for Health Policy Research [pyran] ol-1gyA0 yas ‘ Health Economics and Evaluation Research Program trust and find eligible beneficiaries where they lived. These conclusions are supported by the following evidence: As of PY 6, Pilots perceived referrals from WPC partner agencies as more effective (average rating of 7.7 out of 10) than referrals from other (non-WPC partner) community-based agencies (6.5). Pilots also rated shelter, street, or other field-based (i.e., hospital/medical care delivery facility) outreach as highly effective (7.5), with the added benefit of allowing for warm-handoffs to WPC. Pilots most often utilized existing data to determine eligibility, including electronic medical records and other medical data (21 of 26) and information provided by WPC partners (e.g., SMI/SUD diagnosis, homelessness indicators; 21). Sustained enrollee engagement was an important focus of Pilots. Strategies included developing rapport and trust with enrollees, ensuring multiple points of contact, consistent care coordinator assignment, and utilizing staff, such as community health workers (CHWs) and peer support specialists with lived experience similar to that of the enrollee. Between January 2017 and December 2021, Pilots cumulatively enrolled 247,887 unique individuals with up to 100,968 enrollees at a time. Most enrollees either stayed continuously enrolled or were disenrolled once; only 17% of enrollees enrolled and disenrolled multiple times. Enrollment size varied significantly by Pilot and often reflected county population size. Los Angeles was the largest Pilot with 76,107 enrollees and there were six total Pilots with enrollment numbers over 10,000. SCWPCC had the smallest enrollment size with 143 enrollees. Ten Pilots had enrollment under 1,000. The average length of enrollment was 14.2 months. Shorter enrollment lengths were common, with 38% enrolled for less than 6 months and 11% enrolled for one month. Enrollment length varied significant by Pilot, from mean of 5.8 months in Shasta to 29.7 in Marin, likely reflecting differences in populations of focus and in program goals. Of the 200,734 disenrollments from WPC, the most commonly reported reasons for disenrollment were “Lack of Engagement” (26%), “WPC Services No Longer Needed” (23%), “Other” (21%), and “Not Eligible for Medi-Cal” (16%). An additional reason for disenrollment, “Graduated,” was not added until PY 3 and accounted for 6% of disenrollments. Pilot used different approaches to classifying enrollees in the target populations. The majority of enrollees were in the high utilizers (57%) and homeless (53%) target populations and fewest enrollees were in the COVID-19 (16%) and chronic physical conditions (10%) target populations. Enrollees classified in the COVID-19, chronic physical conditions, and SMI/SUD target populations had the longest average length of enrollment, ranging from 17.2 to 20.0 months. Executive Summary | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Beier Health Economics and Evaluation Research Program i WPC Services Offered and Delivered WPC Pilots aimed “increase coordination and appropriate access to care” and “increase access to housing and supportive services.” Analysis of data showed that Pilots offered more services than expected to address various social and health needs of enrollees and the intensity of services were often greater for highest need enrollees such as those with SMI/SUD or chronic physical conditions. These conclusions are supported by the following evidence: e Pilots designed service categories in bundles (per-member, per-month or PMPM) or individually (fee-for-service or FFS) depending on whether Pilots were paid through capitated payments or single payments for defined services, respectively. Pilots offered as many as 16 and as few as 1 PMPM bundles. They also offered as many as 21 and as few as 1 individual services (FFS). Some Pilots disaggregated services into numerous bundles and individual services (e.g., Alameda) and others relied on very few bundles (e.g., San Mateo, Solano). e@ Consistent with the goals of WPC, all Pilots offered outreach, care coordination, housing support, benefit assistance and transportation. The majority of Pilots also offered health education (92%), legal services (84%), employment assistance (76%), and medical respite (72%). Sobering centers and re-entry services were the least often offered (56% and 28% of Pilots, respectively). e Enrollees most often received care coordination services (89%), followed by benefit assistance (79%) and outreach (73%). Other common services included housing support (70%), legal services (68%), and transportation (63%). e About 14% of enrollees received sobering center care and 6% received medical respite care. These services offered alternatives to EDs, hospitals, or jails. Under WPC, sobering center care services could be offered to eligible populations not enrolled in the program and were provided to 15% of this group. e@ The proportion of each target population receiving specific services varied. For example, enrollees identified in the chronic physical conditions target population were the most likely to receive medical respite (28% compared to 6% of all enrollees). Similarly, those in the SMI/SUD target population were most likely to receive sobering center services (49% compared to 14% of all enrollees). The justice-involved target population was most likely to receive housing support services (89% compared to 71% of all enrollees). © Overall, nearly $3.6 billion was paid to WPC Pilots, ranging from $6.2 million (Solano) to $1.5 billion (Los Angeles) per Pilot. Annual payments increased from $361 million in PY 2 to $778 million in PY 5. e Payments for PMPM bundles and FFS made up 45% and 8%, respectively, of the total payments to WPC Pilots between PY 2 and PY 6. Twenty out of 25 Pilots were mainly paid for services through PMPM bundles. e Assessment of payments by target population was a reasonable proxy for the intensity of service use and showed higher intensity of services to the SMI/SUD target population. Whole Person Care Final Evaluation Report | Executive Summary erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program On average, Pilots were paid $13,541 for WPC services for SMI/SUD enrollees overall ($670 per month), which was higher than the average overall payment per enrollee of $6,272 ($397 per month). WPC Care Coordination WPC aimed to “increase coordination and appropriate access to care for the most vulnerable Medi-Cal beneficiaries.” Evidence suggests Pilots were successful in developing diverse and appropriate infrastructure (e.g., staffing, data sharing, standardized protocols) and effectively delivered of care coordination services (e.g., needs assessment, care plan, referrals) needed to support effective care coordination. These efforts were particularly innovative and notable in development of multidisciplinary care coordination teams with lived experience and delivery of services to enrollees where they lived. These conclusions are supported by the following evidence: e@ In PY5, 18 of 25 Pilots reported using community health workers, peer coaches, or other staff with lived experience relevant to enrollees to provide care coordination services. e@ Median caseload across all Pilots was approximately 20 to 30 enrollees per care coordinator. Pilots offered tiered caseloads to best meet enrollee need. e Twenty of 25 Pilots had standardized protocols for referring enrollees to medical, behavioral health, or social services. Standardized protocols helped minimize undesirable variation in delivery of care coordination services, while improving staff workflows and data reporting. e@ In PY 6, 18 of 26 Pilots indicated that they provided financial incentives to partner organizations for engagement in WPC activities and Pilots rated these incentives as effective (6.8 of 10, with 0 = not effective and 10 = extremely effective). Incentives to promote development of data sharing infrastructure within participating partner organizations and for Pilots to achieve set process targets were considered most effective. e In PY5, 21 of 25 Pilots indicated the most common type of contact between care coordinators and enrollees was in-person. e Pilots reported using active referral strategies, such as providing/arranging transportation to and from appointments (24 of 25), ensuring warm hand-offs to other providers (24), and follow-up with enrollees and/or service providers to monitor referral status (23). e@ Fourteen of 25 Pilots reported co-locating or otherwise embedding care coordinators within partner organizations. @ Across all reporting periods, as noted in narrative reports, the three most common care coordination challenges included (1) limited availability and/or accessibility of services being coordinated, (2) engagement of appropriate interdisciplinary partners, and (3) staffing issues. Pilots described efforts to address these challenges by (1) implementing Executive Summary | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program new or improved care coordination services, (2) using data systems to support care coordination activities, (3) working with partners in new ways that improved understanding of mutual goals for shared clients. WPC Quality Improvement, Program Monitoring, and Stakeholder Engagement WPC aimed to “achieve targeted quality and administrative improvement.” Pilots were required to engage in regular quality improvement activities and document their efforts. Evidence indicated substantial effort by Pilots in these quality improvement activities focusing on improving WPC implementation and improving specific outcomes/metrics. These conclusions are supported by the following evidence: @ Of those 2,133 PDSA reports submitted from PY 2 - PY 6, the most common categories submitted included ambulatory care PDSAs (19%), followed by care coordination PDSAs (18%), and inpatient utilization PDSAs (17%). e Since the interim report, DHCS and the contracted WPC Learning Collaborative teams continuously checked-in with the LEs through surveys, phone calls, virtual meetings, and email communications to better understand the issues that were of most interest and concern to help guide provided technical assistance. e Many Pilots attempted to integrate and elevate stakeholder perspectives into their Pilot. In PY 6 surveys, 18 of 26 Pilots felt they had allocated sufficient resources (i.e., time, staff, compensation) to capture key stakeholder input (e.g., frontline staff, enrollees, other community members) throughout their WPC Pilot. WPC and COVID-19 The COVID-19 pandemic started in early 2020, during the fourth year of WPC implementation and resulted in the program being extended for an additional year. UCLA investigated the impact of COVID-19 on WPC implementation, enrollment, and enrollees, as well as whether the impact of the pandemic was similar among enrollees and their matched controls. The findings indicated that Pilots were able to respond to the challenges presented by the pandemic quickly and minimize its impact on WPC enrollment and service use; the unanticipated value of WPC investments in system-wide integration in responding to emergencies such as COVID-19; and a similar rate of COVID-19 infections and service use for WPC enrollees and the control group. These conclusions are supported by the following evidence: e@ In PY 5, most Pilots (18 of 24) reported that using WPC staff greatly impacted their ability to respond to the pandemic due to the staff’s training and expertise developed through WPC. ®@ Specific WPC processes, procedures, or policies were impacted by COVID-19, including staffing policies and procedures (e.g., shifts to telework and protocols for use of personal protective equipment; 21), approaches for engagement of eligible beneficiaries Whole Person Care Final Evaluation Report | Executive Summary erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program or enrollees in WPC services (20), and care coordination processes (19). Pilots successfully adapted their programs to account for the evolving and changing pandemic environment and to continue service delivery to WPC enrollees. @ Monthly enrollment in WPC continued to grow throughout 2020, increasing from 76,015 in December 2019 to 95,866 in December 2020. There was a small increase to 96,416 in December 2021 or the end of WPC. Quarterly new enrollments were smaller as the end of the program neared, but enrollment continued throughout the pandemic. Only nine of the 25 Pilots elected to add the new COVID-19 target population. e@ UCLA estimated the prevalence of COVID-19 infections by identifying claims or encounters with a primary or secondary diagnosis of COVID-19 starting in April 2020. Overall, 10% of enrollees and 8% of controls used a service with a COVID-19 diagnosis and the monthly trends in COVID-19 diagnosis mirrored the countywide trends in COVID-19 cases for both groups. COVID-19 related service use was similar for WPC enrollees and controls, with 23% and 27% of COVID-19 related services being hospitalizations and 16% and 14% being emergency department (ED) visits for WPC enrollees and controls, respectively. e@ The proportion of primary care services and specialty care services that were provided through telehealth was less than 0.1% in 2019. During the pandemic, these proportions increased to as much as 21% and 13%, respectively. e Innarrative reports, the most frequently reported challenges regarding COVID-19 were related to (1) the transition to telehealth and Pilots’ inability to provide WPC services in- person, (2) limited staff capacity due to reassignment of WPC staff employed by county agencies to support broader community COVID-19 emergency responses, and (3) inability to connect enrollees to services (e.g., due to facility closures or reduced provider capacity). ® Despite challenges, Pilots found success with (1) expanded short term housing or shelter availability, (2) partnership support for WPC and COVID-19 response efforts, and (3) improved outreach and engagement. Enrollee Demographics, Health Status, and Prior Health Care Utilization WPC Pilots aimed to enroll the “most vulnerable Medi-Cal beneficiaries,” but had flexibility in choosing from seven populations of focus (e.g., high utilizers, individuals with chronic physical or behavioral health conditions, individuals experiencing homelessness). Data showed that all WPC Pilots successfully enrolled the most vulnerable Medi-Cal beneficiaries who were at risk of or high utilizers. These conclusions are supported by the following evidence: e WPC enrollees were most frequently aged 18-34 (32%), 35-49 (28%), or 50-64 (31%) years old; male (56%); Hispanic (28%), White (28%) or Black (26%); communicated primarily in English (86%), and were enrolled in Medi-Cal managed care prior to WPC (90%). Executive Summary | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 e@ WPC enrollees had high rates of mental health conditions such as depression (37%), anxiety (34%), schizophrenia and psychotic disorders (26%); substance use disorders, such as drug (32%) and alcohol use disorders (21%); and chronic conditions, such as hypertension (33%). e Examination of outpatient services, ED utilization, and inpatient hospitalizations showed an upward trend pre-WPC. From 19-24 months prior to WPC enrollment to 1-6 months prior to WPC enrollment, primary care visits, ED visits and hospitalizations increased from 229 to 244 services, 162 to 211 visits and 32 to 52 stays per 1,000 Medi-Cal member months, respectively. Better Care WPC aimed to use care coordination and WPC services to “increase appropriate access to care.” Evaluation findings provided support for this WPC goal and further insights on how patterns of care changed over time and for important sub-groups of high utilizer Medi-Cal beneficiaries (Exhibit 1). Exhibit 1: Care Related Difference-in-Difference Model Outcomes for WPC Enrollees, PY 2 to PY 6 Differences in trends for WPC enrollees vs. the control group (DD) Medically Intended or Complex or High- Anticipated Enrollees with Risk (MC/HR) direction All Enrollees SMI/SUD/HML Enrollees Primary Care Services per 1,000 Beneficiaries Decrease -330 -255, -535 Specialty Services per 1,000 Beneficiaries Increase 133 133 132 Mental Health Services per 1,000 Beneficiaries Decrease -813 -1,125 43 Substance Use Disorder Services per 1,000 Beneficiaries Increase 56 53 357 Follow-Up After Hospitalization for Mental Illness within 7 days* Increase 2.7% NR NR Follow-Up After Hospitalization for Mental Illness within 30 days* Increase Not Significant NR NR Initiation of Alcohol and Other Drug Treatment* Increase Not Significant NR NR Engagement of Alcohol and Other Drug Treatment* Increase 1.9% NR NR Whole Person Care Final Evaluation Report | Executive Summary UCLA Center for Health Policy Research [pyran] e\-1gyA0 yee ‘ Health Economics and Evaluation Research Program Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: Green indicates significant change in the intended direction. Red indicates significant change in the unintended direction. NR indicates that the analysis was not reported. SMI/SUD/HML is severe mental illness, substance use disorder or experiencing homelessness. *Indicates a WPC universal metric that all Pilots had to report on. Specifically, data showed that enrollees use of outpatient services increased in the first year of WPC. Comparing trends from before to during WPC, enrollees had a reduction in primary care, an increase in specialty care, a decline in mental health care, and an increase in substance use treatment for enrollees overall vs. the control group. Additional analyses showed a somewhat different pattern of change for enrollees with serious mental illness or substance use disorders or experiencing homelessness (SMI/SUD/HML) and enrollees that are medially complex or high (MC/HR). These patterns likely indicated overuse of primary care services prior to enrollment due to barriers in access to other needed services such as specialty care and substance use treatment. These barriers were likely addressed by care coordination that helped patients receive these more appropriate services in the right settings. Further evidence from analyses of WPC metrics and Pilot interviews and surveys supported delivery of better care under WPC. These conclusions are supported by the following evidence: e« For WPC enrollees, their use of outpatient services increased in the first year of WPC enrollment compared to baseline, indicating successful connection to needed to services, likely due to care coordination efforts. ¢ Primary care services utilization was increasing before WPC for both enrollees and controls by 727 and 668 services per 1,000 beneficiaries per year, respectively. During WPC, utilization declined for WPC enrollees by 208 services per 1,000 beneficiaries per year while they continued to increase, although at a slower rate, by 63 services per 1,000 beneficiaries per year for controls. This declining rate of utilization from before to during WPC was greater among WPC enrollees by 330 services. ¢ Specialty service utilization was increasing both before and during WPC for WPC enrollees and their controls, but utilization rates slowed during WPC. The decline from before to during WPC was smaller for WPC enrollees by 133 services per 1,000 beneficiaries per year compared to controls. ¢ Mental health and substance use services utilization was increasing before WPC for both WPC enrollees and their controls. For WPC enrollees, their use of these services increased at the start of WPC and then declined during the program. In comparison to controls, WPC enrollees had a larger declining rate from before to during WPC for mental health services (-813 services per 1,000 beneficiaries per year) and a smaller declining rate for substance use disorder services (56 services per 1,000 beneficiaries per year). ¢ When examining the impact of WPC on utilization trends of outpatient services for SMI/SUD/HML enrollees compared to MC/HR enrollees, UCLA found that enrollees with these conditions had less of a reduction in primary care services and a much larger reduction in mental health services (however overall rates of mental health services Executive Summary |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program were much higher for this group). In contrast, the use substance use disorder services declined for this group, potentially reflecting lower need for these services over time due to use of mental health services. MC/HR enrollees had a much larger declining rate in primary care compared to controls, which may indicate it was easier to transition their care to specialty services. These enrollees also had a larger increase in mental health and substance use services compared to controls, but this is likely due to these enrollees having newly diagnosed mental health and SUD during the program. The declining rates of mental health services among WPC enrollees compared to their controls was isolated to SMI/SUD/HML enrollees. MC/HR enrollees saw a small but significant increase in change of utilization trend compared to controls. The increasing rates of substance use disorder services compared to controls was observed only among the MC/HR enrollees. SMI/SUD/HML enrollees saw no significant change in utilization trends compared to controls. Trends in rates of follow-up care after a hospitalization within seven days increased during WPC for WPC enrollees and the change in trend from before to during WPC was greater for WPC enrollees compared to controls by 2.7%. There was no significant difference between enrollees and controls for follow-up within 30 days. While there was no significant impact of WPC on initiation of alcohol and other drug dependence treatment, the change in trends from before to during WPC of engagement in alcohol and other drug dependence treatment was 1.9% higher for WPC enrollees compared to controls. Pilots reported improvements in annual rates of enrollees that received a comprehensive care plan within 30 days of enrollment (12% to 54%) and within 30 days of the anniversary of their enrollment (43% to 72%). There was a small decline in PY 6 to 46% for those that enrolled in the last year of the program. Pilots reported rates of suicide risk assessments among enrollees with a diagnosis of major depressive disorder increased from 10% to 32%. For enrollees with high and complex needs, such as those targeted by WPC, connection to other services, such as specialty care, would likely increase as a result of ED and IP utilization decreasing. This is particularly the case with Pilots’ concentrated efforts to screen, refer, and engage enrollees in services to best meet their needs and the development of comprehensive care plans. Better Health WPC aimed to “reduce inappropriate emergency and inpatient utilization” and “improve health outcomes for the WPC population.” Evaluation findings provided support for this WPC goal and further yielded insights in how patterns of care changed over time and for important sub- groups of WPC enrollees (Exhibit 2). Importantly, data showed a reduction in ED visits and hospitalizations and an increase in long-term stays for enrollees overall vs. the control group. These patterns likely indicated that care coordination and Pilot efforts to reduce avoidable Whole Person Care Final Evaluation Report | Executive Summary erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program acute care and to divert patients from EDs and hospitals to more appropriate settings were effective. Exhibit 2: Health Related Difference-in-Difference Model Outcomes for WPC Enrollees, PY 2 to PY6 Differences in trends for WPC enrollees vs. the control group (DD) Medically Intended or Complex or High- Anticipated Enrollees with Risk (MC/HR) direction All Enrollees | SMI/SUD/HML Enrollees Emergency Department Visits per 1,000 Beneficiaries* Decrease -130 -173 +11 Inpatient Stays per 1,000 Beneficiaries* Decrease -45 -53 +21 Long-Term Care Stays per 1,000 Beneficiaries Increase 78 95 32 Controlling High Blood Pressure** Increase -0.6% NR NR Not HbA1c Testing Increase Significant NR NR Not All-Cause Readmission** Decrease Significant NR NR Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: Green indicates significant change in the intended direction. Red indicates significant change in the unintended direction. NR indicates that the analysis was not reported. SMI/SUD/HML is severe mental illness, substance use disorder or experiencing homelessness. *Indicates a WPC universal metric that all Pilots had to report on. ** Indicates a WPC variant metric that Pilots could select to report on. Additional analyses emphasized the concentration of avoidable ED visits and hospitalization among enrollees with SMI/SUD/HML and the likely effectiveness of care coordination in reducing them. Hospital reported challenges provided further insights in improving some health outcomes were difficult. These conclusions are supported by the following evidence: e After increasing before WPC, emergency department visits declined during WPC for both WPC enrollees and their controls. Compared to their controls, the declining rates of ED visits from before to during WPC was greater for WPC enrollees by 130 visits. This decline was mainly a result of enrollees with SMI/SUD/HML (173 fewer visits compared to controls). MC/HR enrollees also had a decline of 11 visits per year compared to their controls. e Hospitalizations were rising before WPC and declining during WPC for both WPC enrollees and their controls. Comparatively, the declining rate from before to during Executive Summary | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program WPC was greater for WPC enrollees by 45 stays per 1,000 beneficiaries per year. This decline compared to their controls was present for both SMI/SUD/HML and MC/HR enrollees, but more so for SMI/SUD/HML enrollees. Long-term care (mainly stays in skilled nursing facilities) utilization rates increased during WPC compared to before WPC and at a greater rate than controls by 78 stays per 1,000 members per year. The increasing rate was greater among SMI/SUD/HML enrollees than in MC/HR enrollees. Indicators of better health that some Pilots choose to report as a variant metric included controlled blood pressure, controlled diabetes, and all-cause readmission. UCLA recreated these metrics, when possible, for all WPC Pilots using Medi-Cal enrollment and claims data. Reported rates of controlled blood pressure went up both before and during WPC for both WPC enrollees and their controls. However, the controls had a slightly greater change in trend from before to during WPC by 0.6%. UCLA reported the percent of enrollees with diabetes that had an HbA1c test during the measurement year as an alternative to reporting rates of controlled diabetes, because the latter was infrequently reported in claims data. There was no significant difference in trends between WPC enrollees and their controls. The percent of acute inpatient stays that were followed up by unplanned acute readmissions increased prior to WPC and declined during WPC for both enrollees and controls. There was no significant difference in trends between WPC enrollees and their controls. Among the seven Pilots reporting incarceration rates, the number of incarcerations slightly increased from baseline to PY 2 (18 to 24 per 1,000 member months), but then declined through PY 6 to 6 per 1,000 member months. Seven Pilots reported on the rates of enrollees that reported “excellent” or “very good” overall health and emotional health. Rates of both overall and emotional health were greater than baseline during all program years and ended at their highest rates in PY 6 (28% and 27%, respectively). Eight Pilots reported on controlled high blood pressure for three groups (individuals age 18-59, individuals age 60-85 with diabetes, and individuals age 60-85 without diabetes). For all groups, the rates of blood pressure control peaked in PY 4 and then declined in PY 5 and PY 6. Even after these declines, the rates remained above those reported in the baseline. Twelve Pilots reported the percent of enrollees with diabetes who had controlled Hemoglobin A1c. Rates remained fairly flat throughout the program, increasing from 52% at baseline to 58% in PY 3 and declining to 54% in PY 6. Whole Person Care Final Evaluation Report | Executive Summary erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program « Among the 15 Pilots that reported depression remission at 12 months, the rates of remission were low throughout the program, ranging from 1% to 4%, but did increase from baseline. e WPC Pilots implemented interventions to redirect utilization from emergency departments (ED) and inpatient hospitalizations to more appropriate services and levels of care, including the use of mobile crisis teams, real-time notifications of enrollee ED visits, addressing social needs such as lack of shelter/housing, building trust, and providing education on navigation and appropriate utilization of health services. Lower Costs UCLA assessed seven measures of health care costs that corresponded to majority of utilization measures examined in Better Care and Better Health chapters. The evaluation findings provided support for reduction in overall costs, an estimated $99 per enrollee per year (Exhibit 3). The decline in overall costs was likely accomplished through a decline in outpatient services and hospitalizations compared to the control group. This was despite increases in prescription medication costs and other residual services and no decline in costs of ED visits and long-term care stays. Exhibit 3: Cost-Related Difference-in-Difference Model Outcomes for WPC Enrollees, PY 2 to PY 6 Differences in trends for WPC enrollees vs. the control group (DD) Medically Complex Anticipated Enrollees with | or High-Risk direction AllEnrollees | SMI/SUD/HML | (MC/HR) Enrollees Estimated Total Payments Decrease -$383 -$311 -$581 Estimated Payments for Outpatient Services Decrease -$96 -$63 -$185 Estimated Payments for Outpatient Medications Increase $58 $36 $119 Estimated Payments for ED Visits Resulting in Discharge Decrease -$18 -$32 S21 Estimated Payments for Hospitalizations Decrease -$310 -$360 -$172 Estimated Payments for Not Long-Term Care Stays Increase Significant S47 -$79 Estimated Payments for Residual Medi-Cal Services Increase $50 $63 $17 Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: Green indicates significant change in the intended direction. Red indicates significant change in the unintended direction. Payments are reported per beneficiary per year. ED is emergency department. SMI/SUD/HML is severe mental illness, substance use disorder or experiencing homelessness. EI Executive Summary | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program SRS TEST aay: Evidence further showed differences in categories of costs for SMI/SUD/HML and MC/HR enrollees. The patterns of change for the former enrollees may be because many of their ED visits were non-emergent and their hospitalizations were also avoidable. The patterns of change for the latter enrollees may be because of previously untreated and undiagnosed need and better management or their care. These conclusions are supported by the following evidence: e For WPC enrollees, total estimated Medi-Cal payments were increasing by $3,025 per beneficiary per year before WPC and then were decreasing by $955 per beneficiary per year during WPC (Exhibit 4). While similar trends were seen in the control group, the difference in the change yearly estimated payments from before to during declined by an additional $383 per beneficiary per year for WPC enrollees compared to controls (DD). This decline in costs was greater among WPC enrollees that were mainly medically complex and not experiencing homelessness ($581 decline). For WPC SMI/SUD/HML enrollees, the decline was $311 greater than their controls. Exhibit 4: Difference-in-Difference Findings Comparing Trends in Yearly Estimated Medi-Cal Payments per Beneficiary for WPC Enrollees and Controls WPC Enrollees Controls ES $3,025 EE $2,943 Overall $955 $34 \ J — -$4,160 -$3,777 { } DD: -$383* EN $3,504 $3,425 SMYSUOML 511) a oss : t Se $4,724 -$4,413 ( 4 Dp: -$311* ME $1,618 ME $2,108 ~$411 . MOAR -$502 il a ____—__ i Y J “$2,611 -$2,030 { . J) ! DD: -$581* [BB Yearly change Before wec Hi Yearly Change During WPC Whole Person Care Final Evaluation Report | Executive Summary [aq UCLA Center for Health Policy Research [pyran] e\-1gyA0 yee ‘ Health Economics and Evaluation Research Program Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: *Denotes ps0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference from before to during is: (Change During WPC — Change Before WPC). Difference-in-difference (DD) is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). SMI/SUD/HML is serious mental illness, substance use disorder or experiencing homelessness. MC/HR is medically complex or high-risk. While there was an initial increase in outpatient services during the first year of WPC, utilization of many outpatient services then declined throughout WPC as medical conditions were addressed or stabilized. The estimated payments for outpatient services declined significantly more during WPC compared to before WPC among enrollees compared to their controls by $96 per beneficiary per year. The estimated payments for outpatient medications from before to during WPC increased significantly more for WPC enrollees compared to controls by $58 per beneficiary per year. This change existed for both SMI/SUD/HML and MC/HR enrollees ($36 and $119 per beneficiary per year, respectively). An increase in outpatient medication costs is likely to follow as enrollees experienced improved access to outpatient services and their existing health conditions were better managed. Overall estimated payments for emergency department visits were increasing before WPC and then decreased during WPC, a significant decline of $18 per beneficiary per year among WPC enrollees compared to controls . For SMI/SUD/HML WPC enrollees, there was a significant decline of $32 per beneficiary per year. In contrast, there was an increase for MC/HR enrollees ($21). These findings align with changes observed in utilization. Estimated payments for hospitalizations increased before WPC by $752 per beneficiary per year and declined during WPC by $472. Aligning with the declining rates of utilization, the change in estimated payments from before to during WPC declined by an additional $310 per beneficiary per year for WPC enrollees compared to controls and these declines were observed for both SMI/SUD/HML and MC/HR enrollees. There was no significant difference in the change of estimated payment for long-term care between all enrollees and controls. However, when restricting to MC/HR enrollees, the trend declined by an additional $79 compared to controls. Appropriate coordination of care for individuals that were medically complex and without the complications of SMI/SUD or homelessness may have resulted in these individuals being able to maintain their health out in the community rather than needing long-term care. Residual estimated payments for WPC enrollees and controls were increasing before WPC, but then continued to increase for WPC enrollees while decreasing for controls. Compared to controls, the trend in estimated payments for residual services increased by an additional $50 for WPC enrollees. Executive Summary | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Homeless WPC Enrollee Services and Outcomes WPC targeted beneficiaries who were experiencing or at-risk of homelessness and aimed to “increase access to housing and supportive services.” Evaluation findings showed that Pilots succeeded in enrolling mostly beneficiaries who were experiencing homelessness; provided housing support services to them using innovative and effective approaches; and improved their outcomes. These conclusions are supported by the following evidence: In PY 5 surveys, 24 out of 25 Pilots reported providing one or more housing related services either through the Lead Entity or the WPC partnership network, at time using alternative funds to supplement WPC funds. Nearly all Pilots (23) promoted a "Housing First" approach in which provision of permanent housing was prioritized (i.e., persons experiencing homelessness were not required to address behavioral health problems or graduate from other service programs before accessing housing). Twenty LEs participated in a data-related activity with a housing agency as a part of WPC. All but five Pilots had housing navigators involved directly in care coordination with enrollees. Nearly all (22) LEs reported the use of housing specialists, many of whom had lived experience of homelessness or risk of homelessness to provide housing and supportive services for WPC enrollees. In PY 6 follow-up interviews and narrative reports, common challenges Pilots faced included: (1) a lack of affordable housing stock, (2) collecting data to measure housing outcomes, and (3) successfully linking enrollees to appropriate supportive services once housed. A major issue in addressing housing challenges for enrollees experiencing homelessness was lack of funding to directly provide housing and insufficient housing supply. Some Pilots leveraged other funding sources and worked with external partners to mitigate these challenges. COVID-19 emergency housing projects expanded short-term housing availability for many WPC enrollees and facilitated care coordination through co-located medical, behavioral, and social services. Half of WPC enrollees (50.2%) were identified as experiencing homelessness by the Pilots. By the end of the program, 124,414 enrollees experiencing homelessness had been in the program with up to 50,610 enrolled at any given time and they had an average enrollment length of 15 months. There was variation in the number of enrollees experiencing homelessness by Pilot. Los Angeles has the most enrollees experiencing homelessness (56,413), followed by San Francisco (22,749) and Orange (13,861). The majority of enrollees experiencing homelessness were male (64%) and 18 to 64 years old (28% 18 to 34, 30% 35-49, and 34% were 50-64 years old). They were most Whole Person Care Final Evaluation Report | Executive Summary [Ej UCLA Center for Health Policy Research [pyran] e\-1gyA0 yee ‘ Health Economics and Evaluation Research Program often White (28%), Black (28%), or Hispanic (25%) and primarily communicated in English (92%). @ Behavioral health conditions were common in this population, with over one-third of these enrollees having depression, drug use disorders, depressive disorders, or anxiety disorders. Over one-quarter had schizophrenia and other psychotic disorders, bipolar disorder, tobacco use, or alcohol use disorders. @ UCLA analysis of WPC service utilization showed that enrollees experiencing homelessness more frequently received re-entry services and medical respite and less frequently received employment assistance and health education. The average amount paid to Pilots for WPC services for enrollees experiencing homelessness was $8,481 compared to $3,798 for those not experiencing homelessness. e Based on Pilot reporting, high rates of permanent housing, defined as being permanently housed for seven months after being housed for six months, were maintained throughout the program (94%-99%). e Pilots reported the rates of enrollees receiving housing services and supportive housing after being referred for those services. Housing service rates increased from baseline through PY 5 (47% to 78%) before declining in PY 6 (61%). Supportive housing rates declined after baseline (42%) to a low of 4% in PY 6. Supportive housing rates were highly influenced by one large Pilot with low rates. e Enrollees experiencing homelessness had declining trends in both emergency department visits and hospitalizations from before to during WPC that were significantly greater than their controls. e Both mental health and substance use disorders service use increased in the first year of WPC compared to baseline, but then declined during WPC. For mental health services, the declining trend in utilization was greater for the WPC enrollees. For substance use disorder services the declining rate was not significantly different from controls. e@ There was no significant difference in the change in trends from before to during WPC for follow-up after hospitalization at 7 days or 30 days or all-cause readmission rates for WPC enrollees experiencing homelessness compared to controls. @ While there was no significant change in trends for initiation of alcohol and other drug dependence treatment for WPC enrollees experiencing homelessness compared to controls, there was a significantly slower decline in engagement of treatment. EI Executive Summary | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err] Health Economics and Evaluation Research Program WPC Transition to CalAIM The sustainability of WPC was ensured by inclusion of Enhanced Care Management (ECM) and Community Support (CS) services under Medi-Cal and similarities between the WPC target populations with the CalAIM “populations of focus.” DHCS provided significant meeting facilitation and technical support during PY 5 to address transition challenges. These efforts led to participation of all WPC Pilots, either the Lead Entities or their partners, in CalAIM as ECM or CS providers. This transition insured that the major goals of WPC including promoting development of local public-private partnerships that were supported by data sharing infrastructure in order to provide care coordination to Medicaid beneficiaries who were high utilizers of care were sustained. These conclusions are supported by the following evidence: e@ DHCS provided technical assistance and support to LEs, and all LEs participated in planning meetings about the transition and sustainability of key components of WPC. The CalAIM planning meetings with DHCS helped ensure appropriate handoffs and care continuity for WPC enrollees. e As of May 2022, based on administrative data from DHCS, 18 WPC LEs were operating as ECM providers. In an additional five counties, the LE was not an ECM provider, but WPC partner(s) were. Only two Pilots and their partners did not participate in ECM (Small County Collaborative counties and Solano). e@ ECM included WPC target populations including individuals experiencing homelessness (23 of 23 counties), adults with SMI/SUD (23), high utilizers (17), and justice-involved (14). e All WPC-participating counties, except Placer, began serving new populations of focus under ECM, with the biggest increases seen in the percentage of counties serving adults with SMI/SUD (from 35% in WPC to 100% in ECM) and adults transitioning from incarceration (from 17% to 61% in ECM). e@ The most common CS services provided by LEs were housing tenancy and sustaining services (8 of 23), followed by housing transition navigation services (7) and housing deposits (7). e@ Innarrative reports, the most frequently mentioned challenge by Pilots was that the scope of services and eligibility requirements for ECM differed from WPC (14 of 23). e Eighteen Pilots noted success in regular planning meetings and workgroups, which brought participating partners together to discuss the necessary next steps in the transition to CalAIM. @ When asked about their commitment to sustaining key goals of WPC, all Pilots expressed commitment to increased coordination of care and access to WPC-like services. Whole Person Care Final Evaluation Report | Executive Summary [REY erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program e@ Transition of WPC was further aided by the DHCS WPC Services and Transition to Managed Care Mitigation Initiative”. The initiative provided direct funding to specific former WPC Pilot to pay for existing WPC services that mapped to ECM and CS services until they transitioned to CalAIM. Ten Lead Entities were approved for a total of $137 million to sustain WPC services until 2024. Implications The evaluation findings described a major and expansive effort by California Department of Health Care Services to address the needs of the most vulnerable Medi-Cal beneficiaries who were at risk of or high utilizers of acute services in emergency departments and hospitals. The WPC approach to care coordination and provision of housing and other support services were sustained under CalAIM with creation of two new Medi-Cal services called Enhanced Care Management (ECM) and Community Supports (CS) and participation of LEs or their partners in delivery of those services. The WPC implementation approach and best practices are helpful for ongoing implementation of ECM and CS and other states contemplating similar interventions. The findings of the changes in patterns of care implied that similar outcomes may be expected with similar interventions. The differential impact of provision of WPC services on enrollees with variations in complexity of their conditions further implied the importance of a clearer understanding of the beneficiary needs and tailoring interventions to match those needs. These findings also implied the importance of better understanding of what outcomes and benefits can be expected when providing WPC or similar services. Ea Executive Summary | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ele e) Health Economics and Evaluation Research Program i Chapter 1: Introduction WPC Program The California Department of Health Care Services (DHCS) implemented a Section 1115 Medicaid Waiver called “Medi-Cal 2020” that started on January 1, 2016 and was scheduled to end on December 31, 2020. Under this Waiver, DHCS implemented the Whole Person Care (WPC) program to address the challenges in Medi-Cal associated with high-risk, high-utilizing enrollees who have complex care needs. In December 2020, largely due to the impacts of COVID-19, DCHS received approval from the Centers for Medicare & Medicaid Services (CMS) to extend the waiver for one year, through December 31, 2021. WPC Goals The overarching goal of WPC was to improve enrollee health and wellbeing by coordinating needed health, behavioral health, and social services. The program was expected to be patient- centered and lead to efficient and effective use of resources. In the Special Terms and Conditions of the waiver, WPC goals were specified as: 1. Increase integration among county agencies, health plans, providers, and other entities with the participating county that serve high-risk, high-utilizing beneficiaries and develop an infrastructure that will ensure local collaboration among the partners participating in WPC Pilots over the long term; 2. Increase coordination and appropriate access to care for the most vulnerable Medi-Cal beneficiaries; 3. Reduce inappropriate emergency and inpatient utilization; 4. Improve data collection and sharing amongst partners to support ongoing case management, monitoring, and strategic program improvements in a sustainable fashion; 5. Achieve targeted quality and administrative improvement; 6. Increase access to housing and supportive services; and 7. Improve health outcomes for the WPC population. WPC was implemented by 25 Pilots representing the majority of counties and one city in California. Under WPC, Pilots systematically identified target populations, shared data, coordinated care, and evaluated improvements in health of their enrolled population. Pilots consisted of partnerships of public and private organizations, led by a single Lead Entity (LE) responsible for program implementation and submission of various reports to DHCS. Pilots were primarily led by county agencies, and included at least one Medicaid managed care plan, Whole Person Care Final Evaluation Report | Introduction December 2022 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program one health services agency, one specialty mental health agency, one other type of public agency, and at least two community partners. In their applications, Pilots described in extensive detail how they would establish the infrastructure needed for WPC, which eligible populations they were to serve, what bundles of services they would provide and at what level of reimbursement, and whether they would be responsible for pay-for-outcomes (P40) for specific metrics. DHCS solicited two rounds of WPC Pilot applications. The first group of eighteen Pilots were awarded in November 2016 and the second group of seven Pilots were awarded in June 2017 (Exhibit 5). Exhibit 5: Timeline of Key Whole Person Care Activities January 3, 2017 snus Round 1- Legacy: Implementation Round 1- Expansion: Approved Group 1- Expansion: Implementation Round 2. New: Implementation July 1, 2036 March 1, 2017 1 Round 2 Round Applications Due Applications Due January 3, 2016 December 31,2020 December 31, 2021 “Medi-Cal 2020" Waiver Waiver Waiver Starts Original End Extension End WC Extension || sitet serene 1/2018 ~ 12/2018 y2019~12/2019 | 3/2020~12/2020 1/2021 - 12/2021 | ot Py3 PY4 Pys Pv6 GO pase ic January 2017 December 2021 WPC Enrollment Starts Weceackoon al Last Date of Data for Final Report ES Introduction | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program Pilots in the first round could submit an application to expand their program in the second round. A total of 25 Pilots ultimately implemented WPC, including one Pilot that consisted of three small, rural counties. Collectively, these Pilots provided WPC services to a large geographic area of California (Exhibit 6). Exhibit 6: Map of Participating Lead Entities and Counties in California Te San Francisco. Seeker) Sonoma San Bernardino Yee] 5 Legend Round 1 Pilots BBE Found 1 Pilots that Expanded HBB ground 2 Pilots Source: Whole Person Care Pilot Applications (n=25). Note: There were 25 WPC Pilots which consisted of 27 unique Lead Entities. San Benito, Mariposa, and Plumas Counties together formed the Small County Whole Person Care Collaborative (SCWPCC). Plumas left SCWPCC in September 2018. The remaining two SCWPCC counties and Solano did not participate in the PY 6 (2021) extension year. Whole Person Care Final Evaluation Report | Introduction [pyran] e\-1gyA0 yee WPC Lead Entities UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Under WPC, LEs could be (1) a county; (2) a city and county; (3) a health or hospital authority; (4) a designated public hospital; (5) a district/municipal public hospital; (6) a federally recognized tribe; (7) a tribal health program under a Public Law 93-638 contract with the federal Indian Health Services; or (8) a consortium of any of the above. The LE, type of organization, and the abbreviated Pilot name used throughout this report are displayed in Exhibit 7. Plumas, Mariposa, and San Benito counties were considered a single Pilot and participated as part of the Small County Whole Person Care Collaborative (SCWPCC). Plumas stopped implementation in September 2018. Solano and San Benito and Mariposa did not participate in the WPC extension year and stopped implementation in December 2020. Exhibit 7: WPC Pilots and Participating Lead Entities WPC Pilot Lead Entity Type of Lead Entity Abbreviated Pilot Name Alameda County Health Care Services Agency Public health/health services agency Alameda Contra Costa Health Services Healthcare system Contra Costa Services Agency agency Kern Medical Center Healthcare system Kern Kings County Human Services Agency Public health/health services Kings agency Los Angeles County Department of Health Healthcare system Los Angeles Services County of Marin Department Health and Human | Public health/health services Marin Services agency Mendocino County Health and Human Services Public health/health services Mendocino Agency agency Monterey County Health Department Public health/health services Monterey agency Napa County Health and Human Services Public health/health services Napa Agency agency County of Orange, Health Care Agency Public health/health services Orange agency Placer County Health and Human Services Public health/health services Placer agency Riverside University Health System — Behavioral | Behavioral health department | Riverside Health City of Sacramento City government Sacramento Arrowhead Regional Medical Center Healthcare system San Bernardino County of San Diego, Health and Human Public health/health services San Diego San Francisco Department of Public Health Healthcare system San Francisco agency San Joaquin County Health Care Services Agency | Public health/health services San Joaquin agency San Mateo County Health System Healthcare system San Mateo Santa Clara Valley Health and Hospital System Healthcare system Santa Clara County of Santa Cruz, Health Services Agency Public health/health services Santa Cruz E09) Introduction | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program ISSIR AP: WPC Pilot Lead Entity Type of Lead Entity Abbreviated Pilot Name Shasta County Health and Human Services Public health/health services Shasta Agency agency Plumas County Behavioral Health Department * | Behavioral health department | SCWPCC San Benito County Health and Human Services Public health/health services SCWPCC Agency * agency Mariposa County Human Services Department * | Public health/health services SCWPCC agency Solano County Health and Social Services * Public health/health services Solano agency County of Sonoma-Department of Health Sonoma Services Behavioral Health Division Behavioral health department Ventura County Health Care Agency Healthcare system Ventura Source: Whole Person Care Pilot Applications (n=25). Note: There were 25 WPC Pilots which consisted of 27 unique Lead Entities. Three WPC LEs (Mariposa, Plumas, and San Benito) formed the Small County Whole Person Care Collaborative (SCWPCC) and submitted application materials together in order to reduce administrative burden. Plumas left SCWPCC in September 2018. The remaining two SCWPCC counties (San Benito and Mariposa) and Solano did not participate in the 2021 extension year. Target Populations, Services, and Reporting WPC Pilots were required to identify and enroll eligible Medi-Cal enrollees in their geographic area. Pilots were allowed to identify others that were eligible for WPC but not enrolled in Medi- Cal, assist them to enroll in Medi-Cal, and subsequently enroll them in WPC. In determining WPC eligibility, WPC Pilot were required to select target populations from one or more of the following six groups identified by DHCS: (1) high utilizers of avoidable emergency department, hospitals, or nursing facilities (high utilizers); (2) individuals with two or more chronic physical conditions; (3) individuals with severe mental illness and/or substance use disorders (SMI/SUD); (4) individuals experiencing homelessness (homeless); (5) individuals at-risk-of-homelessness; and (6) individuals recently released from institutions, including jail or prison (justice involved). In the third quarter of 2020 DHCS added a seventh target population that included individuals impacted by or at-risk of COVID-19, which could be retrospectively applied to individuals going back to the start of 2020. In their applications, WPC Pilots were required to define individual services or bundles of services that would be provided to enrolled populations. Pilots were required to provide care coordination and housing support, but otherwise had discretion in the types and intensity of services offered. Services varied significantly across Pilots, with some Pilots choosing to bundle and deliver a broad array of services to all enrollees, and others creating bundles with fewer services that could be mixed and matched based on specific enrollee needs. Certain services such as outreach, sobering centers, and medical respite were typically not bundled and only provided on an individual basis. Whole Person Care Final Evaluation Report | Introduction Ea UCLA Center for Health Policy Research [pret] o\-1gyAs yee . ‘ Health Economics and Evaluation Research Program All WPC Pilots were required to report on individual enrollment and utilization or WPC services on a quarterly basis, as well as semi-annually report on five universal, and a minimum of four out of 10 variant metrics (Exhibit 8). Exhibit 8: WPC Universal and Variant Metrics Universal Metrics Variant Metrics « Ambulatory Care - Emergency Department Visits Health e — Inpatient Utilization - General Hospital/Acute Care | * 30-day All Cause Readmissions « Follow-up After Hospitalization for Mental Illness Decrease Jail Recidivism ¢ Initiation and Engagement of Alcohol and Other * Overall Beneficiary Health Drug Dependence Treatment * Controlling Blood Pressure e Proportion of participating beneficiaries with a *® —HbAtc Poor Control comprehensive care plan « Depression Remission * — Suicide Risk Assessment Housing « Permanent Housing « Housing Services « Supportive Housing Notes: WPC Pilots were required to report semi-annually on the four universal metrics and had to choose a minimum of four of 10 variant metrics. Permanent housing = percent of homeless who are permanently housed for greater than 6 months; Housing services = percent of homeless receiving housing services in PY that were referred for housing services; Supportive housing = percent of homeless referred for supportive housing who receive supportive housing. WPC Funding and Pilot Payment Methodology The total budget for WPC was $3 billion. This included $1.5 billion from participating Pilots to implement WPC and $1.5 billion in matching funds from the Medicaid program. Pilots submitted their requested budgets in their applications and provided a rationale and additional information on the broad categories for which funds were to be used. The categories included in the budget requests are described in Exhibit 9. Exhibit 9: Whole Person Care Budget Categories Category Name Category Description Examples Administrative Infrastructure Administrative funding needed to develop and implement the WPC Pilot Administrative staffing, information technology infrastructure Delivery Infrastructure Non-administrative funding with costs allocated to the WPC Pilot Advanced Medical Homes, Mobile Street Teams, Community Resource Databases Incentive Payments Funding of items intended as incentive payments for timely achievement of deliverables by downstream providers Service Integration Team Contractors, Incentive payments for reporting outpatient services Introduction |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program SIRES AP: Category Name Category Description Examples Bundled PMPM Services Funding for more than once service or activity to WPC enrollees Comprehensive Complex Care Management and Housing Support Services Fee for Service Funding for single per encounter payment for a discrete WPC service Sobering Center, Service Integration Team, Field-based Outreach Activity Pay for Metric Reporting Funding planned for collecting and reporting on pilot metrics Number of emergency department visits, Suicide risk assessments Pay for Metric Outcomes Funding depending on outcome achievement with set goals used to determine payments Reduction in the number of emergency department visits, Increase in the percentage of follow-up after hospitalization Source: DHCS’ Whole Person Care Pilot — Budget Instructions. WPC Pilots were reimbursed for delivery of services within the PMPM bundles or FFS budget categories. PMPM bundles comprised of one or more services delivered at a set price per month to the WPC enrollee, while FFS items were single per-encounter payments for a discrete service. Pilots were able to receive additional financial incentives under three other budget categories, including pay for reporting (P4R), pay-for-outcome (P40), or incentive payments to partners. In PY 1, WPC Pilots were to receive infrastructure payments following submitting applications and reporting baseline data. In PY 2 and later years, Pilots were eligible for PMPM and FFS reimbursement, P4R, P40, and incentive payments. Pilots submitted invoices every six months detailing their activities and progress. Whole Person Care Final Evaluation Report | Introduction Try UCLA Center for Health Policy Research Health Economics and Evaluation Research Program UCLA Evaluation The UCLA Center for Health Policy Research (UCLA) was selected by DHCS to evaluate WPC from 2016 to 2020. Following the approved extension of WPC to 2021, the UCLA evaluation was also extended by one year. The evaluation was designed to assess whether WPC achieved its overarching goals. The evaluation broadly examined: if WPC Pilots successfully implemented their planned strategies and improved care delivery; if WPC resulted in better care and better health; and if better care and health resulted in lower costs through reductions in avoidable utilization. Conceptual Framework The original conceptual framework for the WPC evaluation approved by DHCS and CMS highlights how the program was expected to develop the needed infrastructure, improve service delivery (better care) and health outcomes (better health), and enhance sustainability of infrastructure improvements and program interventions and reduce costs through reductions in avoidable utilization (Exhibit 10). YS Introduction | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research " a December 2022 Health Economics and Evaluation Research Program Exhibit 10: Whole Person Care Conceptual Framework Infr cture === Crate = Coca > eer} Sure iag Increased integration \ x Reduced all-cause — elmproved data between county agencies, and avoidable ED *Reduced costs collection health plans, providers, and visits for participating *lmproved data social and other service Reduced all-cause beneficiaries sharing entities re and mental health Reduced Medi- *Expand capacity for *Increased coordination hospitalizations, Cal expenditures delivery of among providers length of stay ,and *Planned for coordinated and *increased care plan readmissions continued integrated care development *Reduced nursing integration of *Develop methods *Achieve targeted quality and facility admissions, county agencies, to assess soc ial administrative improvement readmissions, and health plans, determinants of benchmarks length of stay providers, and health «Increased access to primary eimproved self- social and other Expansion or and specialty care services assessed health service entities development of *Increased access to mental *improved blood Embedded care new programs health and substance use pressure control coordination treatment services : protocols in a . elmproved diabetes standards of Increased access to housing HbA1c control " and supportive services eReduced depression practice Increased primary and remission specialty care, mental health, *Decreased jail substance use and social recidivism service use and follow up post ED visits, hospitalizations, ——————— nursing home stays rs eS Source: UCLA Whole Person Care Evaluation Design, 2017. Notes: ED is emergency department and HbA1c is hemoglobin A1c. Whole Person Care Final Evaluation Report | Introduction erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Evaluation Questions The UCLA evaluation questions are displayed in Exhibit 11. The findings associated with each question are distributed throughout the report as shown in the exhibit. The evaluation questions were divided into overarching questions that described the program broadly, followed by specific questions that were aligned with elements of the conceptual framework. Exhibit 11: WPC Evaluation Questions and Location of Associated Findings Research Question Location in Final Report Overarching Questions information sharing amongst local entities to support identification of target populations, ongoing case management, monitoring, and strategic program improvements in a sustainable fashion; and B) achieve the approved application deliverables relating to data collection and information sharing? 1, What are the demographics of WPC enrollees? What services did WPC Enrollment Processes, Size they receive? and Patterns; WPC Services Offered and Delivered Demographics, Health Status, and Prior Health Care Utilization 2. What key factors aided or hindered the success of specific strategies Conclusions in implementing or achieving the intended outcomes, and what measures are WPC Pilots taking to address these barriers? 3. What are the structural differences of the various WPC Pilots and Structure of WPC Pilots how are differential WPC Pilot outcomes related to structural differences? Infrastructure 4. To what extent did the WPC Pilot: A) develop collaborative Structure of WPC Pilots leadership, infrastructure, and systematic coordination among public and private WPC Pilot partners, including county agencies, health plans, providers, and other partners that serve high-risk, high-utilizing Medi-Cal beneficiaries; and B) achieve the approved application deliverables relating to collaboration, infrastructure, and coordination? 5. To what extent did the Pilot: A) improve data collection and Chapter 3: Health Information Technology and Data Sharing InfrastructureHealth Information Technology and Data Sharing Infrastructure Better Care supportive services and improve housing stability? 6. To what extent did the Pilot: A) improve comprehensive care WPC Care Coordination coordination, including in-real-time coordination, across participating entities; and B) achieve the approved application deliverables relating to care coordination? 7. To what extent did the Pilot: A) increase appropriate access to care Better Care; WPC Services Offered and social services; and B) achieve approved application deliverables | and Delivered relating to WPC service delivery? 8. To what extent did the Pilot increase access to housing and Homeless WPC Enrollee Services and Outcomes Better Health 9. To what extent did the Pilot: A) improve beneficiary care and health outcomes, including reduction of avoidable utilization of emergency and inpatient services; and B) improve outcomes such as controlled blood pressure and Hemoglobin Alc (HbA1c)? Better Health Introduction |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ely Health Economics and Evaluation Research Program Research Question Location in Final Report Lower Costs and Sustainability 10. To what extent did WPC Pilots reduce costs of care for WPC enrollees compared to the control group and were total Medi-Cal expenditures reduced during the WPC program? Lower Cost 11. What lasting collaboration between partners and care coordination protocols will continue after the WPC program? In addition, how will counties ensure that improvements achieved by the Pilots will be sustained after WPC program funding is exhausted? WPC Transition to CalAIM Source: UCLA Whole Person Care Evaluation Design, 2017. Data Sources UCLA used multiple qualitative and quantitative data sources for the evaluation and expanded data collection efforts due to the COVID-19 pandemic and the extension of WPC in 2021. Data sources are summarized in Exhibit 12 and described in further detail below. When available, UCLA presents data points across multiple time periods of program implementation. Exhibit 12: Overview of WPC Evaluation Data Sources Data Source [Time Period [Pilots included Reports to DHCS WPC Pilot Applications 2016 All 25 Pilots including 3 LEs from SCWPCC. WPC Mid-Year and Annual Narrative Reports Bi-annual, 2017- 2021 Narrative Report Attachments, Including Plan-Do-Study-Act Reports Bi-annual, 2017- 2021 Annual Universal and Variant Metrics Reports Baseline-2021 WPC Enrollment and Utilization Reports Quarterly, 2017- 2021 Annual WPC Invoices 2016-2021 All 25 Pilots through PY 5. Sonoma and SCWPCC did not participate in PY 6. UCLA Surveys PY 3 Lead Entity (LE) Survey June-September 2018 All 25 Pilots including 3 LEs from SCWPCC. PY 3 Partner Survey June-September 2018 227 partner organizations from 24 Pilots; Sonoma partners did not participate due to delayed implementation and Plumas (from SCWPCC) exited Pilot in September 2018. PY 5 COVID-19 Impact Survey Rapid response; April 2020 24 Pilots including 2 LEs from SCWPCC; Napa did not respond. PY 5 LE Survey June-August 2020 All 24 Pilots including 2 LEs from SCWPCC; Napa did not respond. PY 5 Partner Survey June-August 2020 166 partner organizations from 24 Pilots; partners from Napa did not participate. Whole Person Care Final Evaluation Report | Introduction Ea err UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Data Source Time Period Pilots Included PY 6 LE Survey May-June 2021 All 25 Pilots including 2 LEs from SCWPCC; Solano and SCWPCC did not participate in PY 6 and were asked to complete with perspective through PY 5. UCLA Interviews PY 3 Follow-up Interviews with LEs September 2018- All 25 Pilots including 3 LEs from SCWPCC; and Frontline Staff March 2019 Plumas participated in follow-up after exiting the Pilot. PY 6 Follow-up Interviews with LEs June-September All 25 Pilots including 2 LEs from SCWPCC. and Frontline Staff 2021 Solano and SCWPCC did not participate in PY 6 and answered with perspective through PY 5. Medi-Cal Data Enrollment, Encounter, and Claims 2015-2021 At least two years of baseline for WPC enrollees and a group of potential controls that met specific criteria. Qualitative Data WPC applications included Pilots identification of the target population; a description of the WPC Pilot structure, partnerships for implementation, and the needs of the target population; services that would be provided and interventions applied; and the associated funding request. In PY 3, UCLA fielded a web-based interim survey to LE leadership. Questions assessed health information technology infrastructure, specific activities related to project implementation, ratings of level of effort, staffing and workforce development, participation in quality improvement activities, and challenges and solutions. Additionally, during this time, UCLA fielded an interim survey to key partners that was completed by 227 partner representatives from 24 WPC Pilots. Sonoma partners did not participate due to delayed implementation and Plumas was not included because they stopped implementation in September 2018. Questions assessed partners’ motivation to participate, collaboration with the LE, and perceived impact of the WPC program. In early PY 5, UCLA administered web-based COVID-19 impact surveys to WPC Pilots, of which Napa did not participate. Questions assessed the impact of COVID-19 on key WPC processes, policies, and procedures and how WPC infrastructure and processes facilitated COVID-19 response. In mid-PY 5, UCLA fielded a web-based survey to LE leadership to WPC LEs, of which Napa did not participate. Questions assessed more detailed data on data sharing infrastructure and resources, care coordination processes and supports, housing related services, integration of health and social services, perceived impact of WPC, and sustainability. In PY 6, UCLA fielded an additional survey to LE leadership in all WPC Pilots during the waiver extension year. Questions assessed additional information on WPC implementation, changes to Ei Introduction |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ely Health Economics and Evaluation Research Program WPC since the PY 5 survey, and updates on sustainability planning and progress on transition to Cal-AIM. The PY 3 LE and partner surveys were followed by in-person or telephone follow-up interviews with all WPC LEs. Additional in-depth key informant interviews conducted via Zoom with all operating Pilots occurred in PY 6. Both rounds of interviews were conducted with: (1) key leadership and management, such as project managers, administrators, and directors of the WPC program and (2) frontline staff, such as care coordinators, public health nurses, and social workers. The key informant interview protocol contained a set of standardized questions asked of each WPC Pilot, as well as follow-up questions specific to the WPC Pilot’s individual survey responses, to obtain clarification and additional detail on various aspects of project implementation. Interviews were systematically coded in NVivo to determine key themes across WPC Pilots. Narrative reports were submitted to DHCS bi-annually (beginning with PY 2 Mid-Year and ending with PY 6 Annual). These data included a summary of program achievements and challenges in care coordination, data and information sharing, and data reporting; as well as context around sustainability efforts. Pilots submitted PDSA reports along with their semi- annual reports, which outlined specific quality improvement projects and provided a description of change-management plans and processes to achieve specific Pilot goals related to care coordination, data sharing, and metrics. Quantitative Data UCLA used baseline and annual Universal and Variant Metric Reports to examine Pilot-reported metrics. The baseline report included data from PY 1 when possible and PY 2 when data could not be retroactively collected. These data -included all universal metrics and the subset of Pilot- selected variant metrics. Due to limitations in data sharing or enrollment, some Pilots did not include pre-selected metrics in all annual reports. The Quarterly Enrollment and Utilization Reports included monthly data including the names of WPC enrollees, their date of enrollment, target population(s), homelessness status, and their date and reason for disenrollment when applicable. Additionally, there reports included individual-level WPC service utilization data. For each month, Pilots reported the PMPM bundle and the number of FFS services provided as applicable. Annual WPC Invoices included a breakdown of approved budgets and expenditures for each Pilot by the seven budget categories. The invoices included specific details for each budget category, which showed the components of the approved budgets the Pilots were able to Whole Person Care Final Evaluation Report | Introduction err UCLA Center for Health Policy Research Health Economics and Evaluation Research Program successfully claim. Additionally, the annual invoices contained the cost of each PMPM and FFS categories each year. Medi-Cal enrollment, encounter and claims data for this report were received by UCLA in April 2022 and included data from January 2015 to December 2021. All data from WPC enrollees were received along with data from a pool of potential controls. UCLA additionally received an updated pull of the Medi-Cal data in July 2022. These data included further matured claims from 2021 along with complete data for any WPC enrollees identified after the April 2022 data pull. Analytic Methods UCLA analyzed all data using appropriate qualitative and quantitative methods. The qualitative methods included extracting relevant information from applications, coding and developing themes from the narrative reports and follow-up interviews in NVivo, and reporting descriptive data from survey results. A detailed explanation of the qualitative analyses is available in Appendices C, D, E, F, and G. The quantitative methods included calculating average weighted Pilot-reported metrics and conducting a descriptive assessment of WPC enrollment and enrollment patterns, WPC enrollee characteristics, and WPC enrollee health status. WPC invoice data and individual-level WPC service utilization were combined to create a descriptive assessment of the proportion of enrollees offered WPC services. Using the Medi-Cal data, a control group was constructed using a propensity score methodology and the resulting control group was used in difference-in- difference (DD) analyses of both WPC metrics and UCLA-created metrics. A detailed explanation of the Pilot-reported metrics and the DD analyses are available in Appendices A and B. EI Introduction |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Chapter 2: Structure of WPC Pilots The two primary goals of WPC were to “increase integration among county agencies, health plans, providers, and other entities within the county that serve high-risk and high-utilizing beneficiaries” and “develop an infrastructure that would ensure local collaboration among the entities participating in the WPC Pilots over the long term.” This chapter provides an overview of the organizational structure and partnership networks that established the foundation for achieving these program goals. This chapter addresses the first part of the following UCLA evaluation question: “what were the structural differences of the various Pilots and how were differential Pilot outcomes related to structural differences?” The 25 WPC Pilots were led by 27 Lead Entities (LEs). LEs served as the primary administrative and governing body throughout the duration of WPC. UCLA explored the following evaluation questions in depth in the interim report: “to what extent did the Pilot (a) develop collaborative leadership, infrastructure, and systematic coordination among public and private WPC Pilot entities, including county agencies, health plans, and providers, and other entities within the participating county or counties that serve high-risk, high-utilizing beneficiaries; and (b) achieve the approved application deliverables relating to collaboration, infrastructure, and coordination?” This chapter provides new information on Pilot networks and partner perceptions as of PY 6 (2021). Data sources for this chapter included 25 WPC Pilot applications (including a single application from three Pilots), PY 3 (2018) and PY 5 (2020) LE and partner surveys, and PY 3 and PY 6 follow-up interviews with leadership and frontline staff of all 25 Pilots. Additional qualitative data around challenges and solutions were provided in 25 WPC mid-year and annual narrative reports. For additional detail on data sources and methodology please see Appendices C, D, E, and F. Whole Person Care Final Evaluation Report | Structure of WPC Pilots err UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Organizational Structure The interim report included a description of the types of Pilot Lead Entities (LEs), indicating that the majority (15) were public health or health services agencies, followed by eight healthcare systems, three behavioral health departments, and one city municipality. In September 2018, Plumas left the Small County Whole Person Care Collaborative (SCWPCC) LE, citing limited resources/capacity and staffing issues in UCLA follow-up interviews. The remaining counties, San Benito and Mariposa, ended participation in WPC for the PY 6 extension year, citing limited administrative capacity, particularly considering the COVID-19 pandemic. Throughout the final evaluation report, Plumas is included in data collection and reporting prior to September 2018, and San Benito, Mariposa, and Solano are included in data collection and reporting prior to January 2021. In PY 3 follow-up interviews, Pilots described that the choice of LE was based on which organization was best equipped to provide overall administrative and strategic guidance. For example, Plumas County Behavioral Health Department was described as the logical choice for the LE because of the program’s emphasis on facilitating enrollee access to behavioral health services. Similarly, the San Francisco Department of Public Health was selected as the LE due to its prior experience working with the target population (homeless individuals) and engagement in prior initiatives aligned with WPC goals, such as their Street Medicine program. Finally, Contra Costa County Health Services was identified as the LE because it was an “umbrella agency” for the county’s behavioral health services, public health, emergency medical services, and health plan. “| would ... say that where we placed our Whole Person Care Pilot made a huge impact, like having it based in public health inside the integrated health system at Contra Costa, | mean, it's a unique model for that county-run health system. But it's really like we put this in the heart of the system of the group that is in the community and is also in the health centers and has those existing relationships.” -Contra Costa Structure of WPC Pilots | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program In addition to the six target populations identified by DHCS at the start of WPC, a new COVID-19 target population was added in PY 5 that included “those at risk of contracting COVID-19, those who have contracted COVID-19, and those recovering from COVID-19.” As in the past, Pilots had discretion to identify enrollees in more than one target population. Target Populations Exhibit 13 highlights the primary target population(s) by Pilot. The primary target population is defined as the key demographic of focus that WPC Pilots designed their services, infrastructure, and processes around. Many Pilots had more than one primary target population (17 of 27). Contra Costa, San Bernardino, San Mateo, Santa Clara, Shasta, and Ventura focused only on high utilizers, which was the most inclusive and broad category. In PY 3 and PY 6 follow-up interviews, Pilots described their rationale for selection of specific target populations and some Pilots reported broad and inclusive definitions to provide more flexibility in program implementation and to ensure they could meet projected enrollment goals. Other Pilots developed more restrictive inclusion criteria with the intent of focusing services on specific populations. For instance, Riverside exclusively targeted justice-involved, while San Francisco exclusively targeted individuals experiencing homelessness. “Ours has primarily, from the beginning, focused on a high utilizing population, and | felt like that was almost the broadest net to capture potential participants in it because as part of serving a high utilizing population, we do pull in people who are homeless, people who are recently incarcerated, people with behavioral health concerns, et cetera, so all of the other kind of allowable target populations.” -Ventura “Very early on, we decided that the target population we wanted to serve would be individuals experiencing homelessness. There's been a lot of focus in our community and by our policymakers on people experiencing homelessness .. [put] We have a history of ... difficulty engaging with people experiencing homelessness in some of our other Health and Human Services programs... We weren't sure how much success we [were] going to have, whether we were going to be able to enroll enough people experiencing homelessness ..., and so we left it [inclusion criteria] broad.” -Placer Whole Person Care Final Evaluation Report | Structure of WPC Pilots ery UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 13: Selection of Primary Target Population by WPC Pilot, PY 6 Serious Total Mental Number of IlIness/ Target Chronic Substance At-risk-of- Population High Physical Use Homeless- | Justice- Selected by WPC Pilot Utilizers_| Conditions | Disorder Homeless | ness Involved Each Pilot Alameda x x 2 Contra Costa xX 1 Kern x x x x 4 Kings xX x 2 Los Angeles x x x x x x 6 Marin x x 3 Mendocino x 1 Monterey x 1 Napa x x 2 Orange x x 2 Placer x x x x x x 6 Riverside x 1 Sacramento x x 2 San Bernardino 1 San Diego xX xX x 3 San Francisco xX 1 San Joaquin x x x x 4 San Mateo 1 Santa Clara 1 Santa Cruz x x 2 Shasta 1 Solano x x 2 Sonoma x x x 3 Ventura Xx 1 San Benito (SCWPCC) xX x x 3 Mariposa (SCWPCC) xX x 2 Plumas (SCWPCC) x x 2 Total that Selected Each Target Population 17 4 12 15 9 4 Source: Initially provided in PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; verified in Pilot specific case studies in February-April 2022. Note: SCWPCC is the Small County Whole Person Care Collaborative. Structure of WPC Pilots | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 PY 6 LE surveys highlighted variations in the inclusion and exclusion criteria used by Pilots for attribution of enrollees to target population(s) in their enrollment and utilization reports (Exhibit 14). Pilots used a wide variety of data sources (e.g., standardized screening/assessment tools, electronic medical records, homeless management and information systems) to classify enrollees into one or multiple target populations (see Chapter 4: WPC Enrollment Processes, Size, and Patterns for additional details). Exhibit 14: Examples of Criteria Used by WPC Pilots to Assign Enrollees to Primary Target Populations Primary Target Population WPC Pilot | Target Population Criteria High Utilizers Shasta Adults ages 18 to 64 with two or more ED visits or hospitalizations in the last three months and were homeless or at-risk of homelessness, based on HUD criteria (i.e., people living in a place not meant for human habitation, in emergency shelter, in transitional housing, or exiting an institution where they temporarily resided). Potential enrollees also needed to fulfill one or more of the following criteria: « SMI diagnosis e SUD diagnosis ® —Undiagnosed/undisclosed opioid addiction Kern Top 15% of Medi-Cal beneficiaries by utilization according to predictive risk model including emergency department, inpatient, length of stay, outpatient, primary care visits, behavioral health visits, alcohol and drug visits, history of detention, psychiatric emergency, homeless coordinated entry, foster care, specific prescription drug classes, and chronic conditions. Chronic Physical Kings Individuals with a chronic health condition of diabetes or high blood pressure. Conditions Los Individuals hospitalized and being discharged from a partner medical center Angeles who were not going to a skilled nursing facility, with two or more admissions (medical or psychiatric) within the last 12 months and at least one of the following: 1) initiation of insulin or anticoagulation during the recent admission, and/or 2) taking greater than six medications daily. Serious Mental Los Individuals with a substance use disorder and at least one of the following: 1) IlIness/Substance | Angeles three or more ED visits related to SUD within the past year; 2) two or more Use Disorder inpatient admissions for physical and/or mental health conditions; 3) three or more sobering center visits within the past year; 4) more than two residential SUD treatment admissions within the past year; 5) history of two or more incarcerations with drug use; 6) drug court referral; and/or 7) history of overdose in the past two years. Mariposa Individuals with a behavioral health condition (mental health, substance abuse (SCWPCC) or co-occurring diagnosis) and one or more of the following: 1) repeated incidents of ED use, hospital admissions, or nursing facility placement; 2) two or more chronic conditions; 3) homeless or at-risk-of-homelessness (based on HUD criteria); and/or 4) recently released from institutions (e.g., hospital, county jail, institutions for mental diseases, skilled nursing facility, etc.) or connection to the criminal justice system. Homeless Monterey HUD definition of homelessness (i.e., people living in a place not meant for human habitation, in emergency shelter, in transitional housing, or exiting an institution where they temporarily resided). Whole Person Care Final Evaluation Report | Structure of WPC Pilots [pret e\-1gyAe yee UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Primary Target Population WPC Pilot | Target Population Criteria San Diego Identified through the homeless management and information system or those who had recently accessed homeless services. At risk of San Diego At-risk for homelessness if in an institutional setting, such as jail, a psychiatric homelessness hospital or other mental health facility, or a substance use residential or detoxification program; as well as those in skilled nursing facilities who did not have stable housing at discharge. Sonoma Individuals who were to be unsheltered within two weeks; verification via eviction notice. Justice-Involved Riverside Probationers with the following criteria were targeted: on probation or parole; released from jail/prison in past year; to be released from jail in the following 90 days; at-risk of or experiencing homelessness; had a behavioral health diagnosis; had a physical health diagnosis. COVID-19 Contra Data from homeless management information system informs; criteria Costa included individuals staying at and/or receiving services at FEMA funded sites related to COVID-19 (e.g., Project Roomkey hotels). Monterey Proof of CDC identified high risk factors; medical summary from primary care provider or ED; self-certification form. Source: PY 6 Lead Entity Survey (n=26), May-June 2021, and PY 6 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), June-September 2021. Notes: ED is emergency department. HUD is the Department of Housing and Urban Development. SMI is serious mental illness. SUD is substance use disorder. SCWPCC is the Small County Whole Person Care Collaborative. FEMA is Federal Emergency Management Agency. CDC is Center for Disease Control. Ei Structure of WPC Pilots | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ely) Health Economics and Evaluation Research Program WPC Pilots were required to “increase integration among county agencies, health plans, and Partnerships providers, and other entities within the participating county or counties that serve high-risk, high-utilizing beneficiaries and develop an infrastructure that will ensure local collaboration among the entities participating in the WPC Pilots over the long term.” WPC Pilots were permitted to partner with as many organizations as they wished but were required to include at least one Medi-Cal managed care health plan, one county health services agency, one county specialty mental health agency, one county public agency, and two community partners. The interim report described aspects of Pilot-level decision-making related to earlier stages of the WPC Pilot. Partnerships were classified as internal or external, depending on their relation to the LE. Internal partners were entities that worked under the same umbrella agency as the LE, such as the county hospital or county mental health department, and comprised 17% of partners as of PY 3 surveys. External partners, like health plans, community clinics, and housing service providers, comprised 83% of partners among WPC Pilots in PY 3 surveys. Distribution of internal and external partners varied considerably by Pilot, depending on county resources and structure. The interim report also described partner engagement in WPC development and implementation, and identified impacts of WPC on relationships between partnering agencies. Partner Types Pilots organized their partner organizations into pre-specified categories, determined by DHCS. As of PY 5, Pilots reported a total of 21 partners on average (18 in PY 3), ranging from a minimum of eight partners to a maximum of 50. Overall, Pilots reported 543 total partners (478 in PY 3; Exhibit 15). Across all Pilots, 58% of all partner organizations were community partners (e.g., non-county agencies including private service providers, community-based organizations, non-profits); 23% were county public agencies (e.g., social services, housing); 9% were Medi-Cal managed care plans; 5% were county specialty mental health services agencies; and 5% were county health agencies. The partner type composition was similar to that presented in the interim (PY 3), with variation at the Pilot level. Whole Person Care Final Evaluation Report | Structure of WPC Pilots UCLA Center for Health Policy Research Health Economics and Evaluation Research Program [pyre el-1gyAe ye Exhibit 15: DHCS Pre-Specified Partner Type by Lead Entity, PY 5 Alameda | 63% (5% 29% w Contra Costa | 50% | 8% Kern 40% (EB 40% Kings 38% (Sx 38% rE Los Angeles | 84% ee + Marin 82% 3% 13% zy Mendocino 50% EB 13% aa Monterey 65% (ee) 12% Napa 50% ey sx Per Orange — 88% Placer 67% ey) oy Riverside 43% 4%) 36% Sacramento | 63% BEY 10% San Bernardino 56% Ea) 22% San Diego | 45% (Bs 15% San Francisco | 33% EN 33% San Joaquin | 44% (Bx) 36% San Mateo 88% Santa Clara 63% (B% 21% SantaCruz "61% (ee) 28% Shasta | 73% 17) 7» EAE Solano | 55% (SA) 18% 99 Sonoma | 76% eR) 6x Ventura 59% Th 35% San Benito (SCWPCC) — 50% oe) 20% Mariposa (SCWPCC) 27% se 27% Community partner Medi-Cal Managed Care Plan County public agency Specialty mental health agency m™ County health agency Source: PY 5 Updated Partnership Lists, January-March 2020. Note: WPC Pilots were permitted to partner with as many organizations as they wished but were required to include at least one Medi-Cal managed care health plan, one county health services agency, one specialty mental health agency, one county public agency (e.g., social services, housing), and two community partners (i.e., non- county agencies including private service providers, community-based organizations, non-profits). Ei Structure of WPC Pilots | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research December 2022 Health Economics and Evaluation Research Program Pilots indicated that some community partners, such as Bay Area Community Services were in several counties (Solano, Alameda, and San Mateo). Examples of specific partner organizations and their role in the WPC Pilot are provided in Exhibit 16. Exhibit 16: Selected Examples of Specific WPC Partners by DHCS Pre-Specified Partner Type and their Role within the WPC Pilot, PY 5 Partner Type Partner Name and Pilot Role in Pilot County Public Marin Housing Authority (Marin) Provided housing and homelessness services, including housing navigation and waiver application support. enc «aval 5 Facilitated enrollee warm hand-offs to divert Agency Riverside County Probation Department " . 3 " aed incarceration or to support reentering (Riverside) 4 community. < Provided daily data feeds to the LE to facilitate CalOptima (Orange) ; Sartore * identification of eligible enrollees. Medl:Cal Integrated into local health information Managed Health Plan of San Mateo (San Mateo) eI exchange to share data for WPC. Care Plan Alameda Alliance for Health (Alameda) Facilitated care coordination services. Oversaw and subcontracted with community- Redwood Quality Management Company based behavioral health services in the county. : Later, responsible for employing and (Mendocino) . ae Specie supervising wellness coaches providing care pecary coordination under WPC. Mental 5 F Haste Contracted with LE to provide care Reena County Behavioral Health Services (Orange) coordination in conjunction with broader WPC Beney team. Ventura County Behavioral Health . Provided substance use treatment to Department, Alcohol and Drug Programs ee individuals over 18 years old. (Ventura) Emergency Medical Services (Contra Costa) Improved emergency department enrollee County discharge processes and workflows. Health Solano County Family Health Services Facilitated referrals and enrollee access to Services (Solano) services, Agenc Facilitated data sharing and access to needed Beney Placer County Public Health (Placer) " ata sharing t services for enrollees. Bay Area Community Services (Multiple) Provided social services and operated the largest homelessness program in the Bay Area. Provided multi-lingual comprehensive health La Clinica de la Raza (Multiple) care services in several counties in the Bay Area. i ili 1 havioral Ith Community | ¢, ont street (Santa Cruz) Facilitated enrollee access to behavioral healt Partner services. Sacramento Self Help Housing (Sacramento) Provided housing and supportive services, including tenancy support, long-term housing, emergency shelter, and outreach. Positive Directions (San Francisco) Facilitated enrollee access to behavioral health care. Whole Person Care Final Evaluation Report | Structure of WPC Pilots Ea err UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Partner Type | Partner Name and Pilot Role in Pilot Facilitated emergency department follow-up Sutter Health (Placer) visits and dissemination of real time alerts on enrollees. Facilitated outreach and access to housing Brilliant Corners (San Mateo) support for enrollees experiencing homelessness. Source: Whole Person Care Pilot Applications (n=25), 2016; PY 5 Updated Partnership Lists, January-March 2020; PY 6 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=26), June-September 2021. Note: WPC Pilots were permitted to partner with as many organizations as they wished but were required to include at least one Medi-Cal managed care health plan, one county health services agency, one specialty mental health agency, one county public agency (e.g., social services, housing), and two community partners (i.e., non- county agencies including private service providers, community-based organizations, non-profits). UCLA further classified community partner organizations into one of eight service-specific classifications to further illustrate type of services provided. Exhibit 17 shows the distribution of different types of community partners as classified by UCLA. Exhibit 17: WPC Community Partners by UCLA Service-Specific Classification, PY 5 Other, 6% Substance use treatment organizations, Justice involved 5% ‘ organizations and legal support, 6% Behavioral and mental health service organizations, 6% ~~ ? 8 Health care providers, 33% Advocacy organizations and foundations, 7% Social services, 19% Source: PY 5 Updated Partnership Lists, January-March 2020. Notes: Across all Pilots, 58% of partner organizations were community partners (non-county agencies including private service providers, community-based organizations, non-profits). UCLA classified community partner organizations into one of eight service/offering specific classifications. viel Structure of WPC Pilots | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research December 2022 Health Economics and Evaluation Research Program Exhibit 18 provides select examples of types of community partners by service-specific classification. Exhibit 18: Selected Examples of Types of Community Partners by Service-Specific Classification, PYS Community Partner Type Examples Description Health care providers La Clinica de la Raza St. Jude Medical Center LifeLong Medical Care Organizations ranging from community health clinics, regional medical centers, wellness centers, and hospital networks Social services St. Vincent de Paul Society Institute on Aging Second Harvest of Silicon Valley Organizations ranging from 211, food and nutrition services, and adult and aging services Housing and homeless support services People Assisting the Homeless (PATH) Abode Services The Gathering Inn Organizations including shelters, housing navigation, and comprehensive services related to “housing first” principles or becoming “document ready” Advocacy organizations and foundations Marin Community Foundation Los Angeles Advancement Project Organizations promoting community well- being through a wide variety of initiatives Behavioral and mental health service organizations Alcott Center for Mental Health Sierra Mental Wellness Group Organizations providing behavioral health or mental health services, typically for mild to moderate cases Justice-involved organizations and legal support California Rural Legal Assistance California State San Bernardino Reentry Initiative Organizations helping with the transition from jail/prison to the community or providing legal services Substance use treatment organizations Alcott Center for Mental Health Sierra Mental Wellness Group Organizations providing community-based treatment for SUD Other California Long Term Care Education Center Marin County Free Library Community partners that do not fall into other existing categories Whole Person Care Final Evaluation Report | Structure of WPC Pilots err UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Partners’ Level of Involvement For the interim report, LEs had categorized each partner’s level of engagement with WPC by indicating if partners had: (1) limited involvement (e.g., only served as service provider or referral source and not involved in planning or decision-making related to WPC); (2) some involvement (e.g., in data sharing or stakeholder meetings), and (3) active involvement (e.g., in WPC planning and implementation). LEs provided an updated categorization in PY 5. In PY 5, LEs indicated that partner involvement increased between PY 3 and PY 5 (Exhibit 19). In PY 3, 47% of partners across all Pilots were actively involved, 32% had some involvement, and 22% had limited involvement with WPC. Whereas in PY 5, 67% of partners across all Pilots were actively involved, 27% had some involvement, and 6% had limited involvement with WPC. Exhibit 19: Level of Partner Engagement in WPC across all Pilots, as Determined by the Lead Entity, PY 3 and PY 5S PY3 = 22% PYS 6% 27% Le Limited involvement m™Someinvolvement ml Active involvement Source: PY 3 Partnership Lists, January-March 2018; PY 5 Updated Partnership Lists, January-March 2020. The level of partner involvement varied across Pilots. Exhibit 20 shows the specific breakdown of partner involvement by Pilot. Overall, the level of involvement increased across partners from PY 3 to PY 5; in PY 5, 93% of partners were reported as having some or active involvement with WPC Pilots compared to 79% prior to PY 3. All Kings’, Monterey’s, and Orange’s partners (100%) were identified as actively involved. All but five pilots (Alameda, San Mateo, Ventura, Santa Cruz, Mendocino) rated more than half of partners as actively involved. Structure of WPC Pilots | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program BIRSer IESE Exhibit 20: WPC Lead Entity Designation of Level of Partner Engagement in WPC, PY 5 All Pilots Alameda Contra Costa Kern Kings Los Angeles Marin | Napa 8% aa San Bernardino Monterey San Diego San Francisco San Joaquin San Mateo Santa Clara Santa Cruz Shasta Solano Sonoma Ventura San Benito (SCWPCC) Mariposa (SCWPCC) Limited involvement ™Someinvolvement Active involvement Source: PY 5 Updated Partnership Lists, January-March 2020. Whole Person Care Final Evaluation Report | Structure of WPC Pilots eer UCLA Center for Health Policy Research Health Economics and Evaluation Research Program From PY 3 to PY 5, partners’ level of involvement in WPC increased by partner type (Exhibit 21). The increase was greatest from 39% to 64% for community partners having active involvement. Exhibit 21: Level of WPC Partner Engagement by DHCS Pre-Specified Partner Type, PY 3 and PY 5 py3 | Ea 30% Ey partner PYS PY3 § Specialty agency PYS 19% Ese PY3 33% ry agency pys —E a Ey PY3 PYS services agency PY3: Plan Medi-Cal Managed Care County health County public mental health Community PYS Limited involvement &Someinvolvement mActive involvement Source: PY 3 Partnership Lists, January-March 2018; PY 5 Updated Partnership Lists, January-March 2020. VCS) Structure of WPC Pilots | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research exter UI08) Health Economics and Evaluation Research Program In PY 3 and PY 5, involvement also increased by UCLA service classification (Exhibit 22). Partner types with the most increase to active involvement were substance use treatment organizations with 25% in PY 3 to 59% in PY 5, advocacy organizations and foundations (26% to 60%, respectively), and housing and homeless support services (43% to 74%, respectively). Exhibit 22: Level of Community Partner Engagement by UCLA Service-Specific Classification, PY 3and PY5 PY3 re g || ee PYS ia % Health care providers g PY3 Ba 37% sg 85 ail a PYS aa 35% By 2 2 a 0% & Beg PY3 17% Ea 2Paag gece 3235 pys & 19% 74% = =s @ 3S 6 Py3 | 32% 26% Seuk Sess 2h 2 PYS Ea a 5 2 a ts 5 Se.202 PY3 EE 18% SGELS 7 ZBESeN 2oeae SoP£SS pys 42% 26% a& = o a 65 7 Fi eBeBg mvs EE 54% SSe@ Bors 2 $eces SESSA pys Ea 61% ° 2 8 ES py3 ry EE ra eon | BgER Bige 3 £ & pys a Fa Ea o . PY3 il 33% EEDa G 2 6 PYS [ty 30% Ga Limited involvement &Someinvolvement mActive involvement Source: PY 3 Partnership Lists, January-March 2018; PY 5 Updated Partnership Lists, January-March 2020. Whole Person Care Final Evaluation Report | Structure of WPC Pilots [pyre el-1gyAe ye UCLA Center for Health Policy Research Health Economics and Evaluation Research Program In PY 6 follow-up interviews and mid-year and annual narrative reports, Pilots noted that these partnership gains required effort, and identified some inherent challenges in building fruitful relationships, such as partner staffing turnover and limited partner interest and buy-in. Most LEs experienced challenges with partner buy-in during the first few years of the Pilot, with relative ease of collaboration in PY 5 and PY 6. Specific examples of initial challenges and solutions related to partnerships buy-in are described in Exhibit 23. Exhibit 23: Selected Examples of Challenges and Solutions to WPC Partner Buy-in Challenges WPC Pilot Selected Examples Data sharing Alameda Initially, Alameda’s partners expressed skepticism about data sharing due to concerns around protecting enrollees’ privacy. Alameda demonstrated the need of data sharing to effectively coordinate care and built trust with partners through clear protections of enrollee data. Orange Integration of behavioral health system data was a challenge and inhibited understanding of which services enrollees were accessing. Persistent partner engagement and demonstration of the utility of shared data supported eventual buy-in by partners in Orange. Marin Marin experienced difficulty with partner uptake of their case management platform due to multiple competing or existing data systems. They developed data exchanges between various systems and found financial incentives supported uptake. Communication San Bernardino Partner engagement was a challenge in San Bernardino due to high staff turnover within partner organizations. San Bernardino utilized regular meetings and constant communication through a variety of modalities to ensure consistent messaging and understanding. Sonoma Sonoma emphasized establishing engagement with federally qualified health centers was an ongoing process. It took roughly six months to establish relationships strong enough to establish workflows and referral pathways, and these relationships required consistent attention. Los Angeles Los Angeles recognized communicating WPC goals and service opportunities with external partners (e.g., hospitals, community organizations) would have been better supported by emphasizing internal communications with County health systems partners early on. Partner goals and roles Mendocino Mendocino stated it was necessary to have a greater understanding of partner goals and capabilities to encourage meaningful engagement and understand partner roles within WPC. Placer Partner delivery on WPC housing principles was a challenge. Placer utilized direct communication with partners to gauge capacity and confirm alignment with WPC strategies related to permanent supportive housing. Kings Kings emphasized leveraging data storytelling to demonstrate the impacts of WPC on their county to increase buy-in from county governance. By convening various organizations, they reduced service duplication. Source: PY 6 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=26), June-September 2021. Structure of WPC Pilots | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 “| would say small, incremental, but important change is how | would characterize it. Have we seen a revolution? No. But have we seen small, steady progress where people understand across the divisions that this client population needs a special level of care that involves all of us as team members? Yes, we have seen that recognition grow and we've seen people actually more willing to participate. And not only that, actually now seeking out opportunities for partnering.” -San Mateo In PY 6 follow-up interviews, Pilots also described successes in increasing partner engagement and buy-in (e.g., frequent communication, active role in shared decision-making, consensus on roles and responsibilities). It was important for Pilots to “meet partners where they were at” and to develop compromises when partner agencies faced competing priorities. Specific examples of partnership buy-in and engagement successes are described in Exhibit 24. Exhibit 24: Selected Examples of Partnership Buy-in Successes Among WPC Pilots WPC Pilot Selected Examples San Diego Continued discussions with partners around HIPAA and updating MOUs as needed increased transparency and clarity among partners sharing data. Kern Increased collaboration between partner county agencies, health plans, and community-based organizations occurred in Kern due to the impact of WPC. As a result of the improved engagement, Kern identified additional programs that can be leveraged to identify solutions and compromises for partners. Kings The leadership of King’s steering committee improved engagement among county agencies, health plans, and other partner organizations; partners’ roles increased and decision-making improved as a result. Riverside Integrating WPC screening nurses in probation offices improved engagement among probation and housing partners significantly. Having the nurse stationed at the probation office facilitated communication and relationship building with cross-sector partner organizations. Santa Cruz Santa Cruz went on a “road show” to meet with partner agencies to gain a better understanding of their programs and services to WPC enrollees. This resulted in increased buy- in from partners by opening communication channels and additional opportunities to collaborate. Los Angeles Los Angeles worked with partners in hospitals and community programs to have “WPC champions” in service-delivery settings to increase care integration and spread the word about WPC services. San Joaquin San Joaquin established a bi-weekly operations meeting with partner agencies in order to build shared understanding of partner agency roles, responsibilities, and objectives in order to reduce duplication of services and getting involved in others’ responsibilities. Sonoma The WPC team met with the multidisciplinary team on a weekly basis to discuss care coordination amongst the Sonoma County safety net agencies. During these meetings, case managers and care team members from the various agencies discussed the enrollees who were seeking services and discuss strategies in this intimate setting to expedite care for the clients. The care team helped locate clients, identify potential referral or service opportunities, upcoming appointments or deadlines, and other opportunities based on the clients’ needs. This Whole Person Care Final Evaluation Report | Structure of WPC Pilots UCLA Center for Health Policy Research D le grAeyyd < ‘ peti Health Economics and Evaluation Research Program WPC Pilot Selected Examples group was extremely successful getting clients in supportive housing, on general assistance programs, supporting upcoming court dates, and getting clients into treatment. Marin Marin General Hospital invited the homeless service providers to monthly meetings with their behavioral health, care coordination, and social work unit supervisors to improve communication and ultimately, successful discharges for these enrollees. Monterey Monterey implemented monthly meetings with core partners that helped to build understanding between partners’ various scopes of work, enhance communications, and streamline workflow. San Diego During internal coordination meetings, San Diego LE continually led discussions on data projects and transition planning for the Pilot to Cal-AIM. Discussions resulted in data mining ahead of transitions to services specific to serious mental illness, allowing for greater buy-in and participation from behavioral health leadership through the transition coordination period. Source: PY 6 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=26), June-September 2021. Notes: HIPAA is Health Insurance Portability and Accountability Act. MOU is Memorandum of Understanding. Structure of WPC Pilots | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program SISSIES Aa: Perceived Impact of WPC on Cross-Sector Collaboration and Integration of Care From PY 3 and PY 5, LEs (75% to 97%) and partners (65% to 72%) reported higher levels of collaboration with each other (Exhibit 25). When asked about specific interactions, LEs reported increases in joint advocacy and planning (65%), referrals (58%), communication about clients (49%), and data sharing (52%) during WPC. Partners reported increases in similar activities as LEs. Exhibit 25: Type of Interaction with Partners among WPC Lead Entities and Partners, Percentages Before WPC, PY 3, and PY5 Lead Entities Partners Any interaction with 75 Interaction with other 65 98 70. Partners WPC Partners Joint Advocacy or 49 Joint Advocacy or Be Planning — Planning Client/P: Referral: as Client/Patient Referral: 6 jient/Patient Referrals 70 ient/Patient Referrals Communication About 36 ~ Communication About Be Client Needs or Care Client Needs or Care Joint Service Deli 8 46 J St Del ia loint Service Delivery joint Service Deliver 14 y 22 17 Data Sharing oe Data Sharing 21 ns Before WPC ™PY3 MPYS Before WPC PY3 MPYS Sources: PY 3 Lead Entity (LE) Survey (n=27), June-September 2018; PY 3 Partner Survey (n=227), June-September 2018; PY 5 Lead Entity Survey (n=25), June-August 2020; PY 5 Partner Survey (n=166), June-August 2020. Notes: Numbers are displayed as percentages. PY 3 partner survey (2018) included partners actively involved or with some involvement and excluded partners with limited involvement. Data Sharing rating derived from question "Please indicate the ways in which your LE CURRENTLY interacts with each of the following WPC partners. Please select all that apply: Administration, Data sharing (e.g., for client/patient care, needs assessment)". Rating not available for WPC Partners in PY 5. In PY 6 follow-up interviews, Pilots reported that WPC provided an important opportunity to develop and/or enhance working relationships with partners. Improved communication and stronger relationships with partners following WPC were often attributed to time spent better understanding how their respective organizations worked, and Pilot investment in data sharing and care coordination. Whole Person Care Final Evaluation Report | Structure of WPC Pilots err UCLA Center for Health Policy Research Health Economics and Evaluation Research Program In the PY 3 and PY 5 partner surveys, partners rated how effective the WPC program was at achieving goals from 0 (not effective) to 10 (extremely effective). Ratings increased between PY 3 and PY S, indicating increased effectiveness of reaching WPC goals (Exhibit 26). On average, partners rated relatively high effectiveness of WPC managing the care of high-risk, high-utilizing populations (7.5) and in improving the coordination of health and social services and collaborative partnerships for program implementation (7.4). Exhibit 26: Partners’ Average Perceived Effectiveness of WPC in Achieving Goals, PY 3 and PY 5 Improving management of care of high risk and high 7. utilizing populations es 7. 5 Improved integration of health and social services a 7 is Improving collaborative partnerships for program 74 implementation EEE 71 PYS mPY3 Sources: PY 3 Partner Survey (n=227), June-September 2018; PY 5 Partner Survey (n=166), June-August 2020. Notes: In response to the question "On a scale from 0 to 10, where 0 = Not effective and 10 = Extremely effective, please indicate the overall WPC Pilot’s effectiveness at achieving the following goals. If unknown or not perceived to be a goal of the WPC program, please select N/A." Partner survey includes partners actively involved or with some involvement and excluded partners with limited involvement. Sample size for selection of goals ranged from 167 to 179 in PY 3, and 146 to 156 in PY 5 as partner organizations could select “unknown” when appropriate. EI Structure of WPC Pilots | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research . as December 2022 Health Economics and Evaluation Research Program Additionally, in PY 3 and PY 5 partner surveys, partners rated how effective the WPC program was at achieving aspects of care delivery from 0 (not effective) to 10 (extremely effective). Ratings increased between PY 3 and PY 5, indicating increased effectiveness of improving aspects of care delivery through WPC (Exhibit 27). Partners perceived WPC to have improved coordination of care and enrollee health and wellbeing (7.5, respectively), and improved the quality of care delivered to enrollees (7.3). Exhibit 27: Partners' Average Perceptions of WPC in Improving Aspects of Care Delivery, PY 3 and PYS 7.2 Improved enrolee Neath 20. es 7; ' ' PA Improved coordination of care for enrollees eee 7: Improved oe coy I 7: PY3 mPYS Sources: PY 3 Partner Survey (n=227), June-September 2018; PY 5 Partner Survey (n=166), June-August 2020. Notes: In response to the question "On a scale from 0 to 10, where 0 = Not effective and 10 = Extremely effective, please indicate the overall WPC Pilot’s effectiveness at achieving the following aspects of care delivery. If unknown or not perceived to be a goal of the WPC program, please select N/A." Partner survey includes partners actively involved or with some involvement and excluded partners with limited involvement. Sample size for selection of goals ranged from 167 to 179 in PY 3, and 146 to 156 in PY 5 as partner organizations could select “unknown” when appropriate. Whole Person Care Final Evaluation Report | Structure of WPC Pilots UCLA Center for Health Policy Research [py-te1un]o\-1ayAs yea . ‘ Health Economics and Evaluation Research Program Chapter 3: Health Information Technology and Data Sharing Infrastructure WPC Pilots were required to “improve data collection and sharing amongst local entities to support ongoing case management, monitoring, and strategic program improvements in a sustainable fashion.” Specifically, Pilots were required to: (1) share enrollee data with and between participating partners as needed for effective care coordination, (2) develop methodology for sharing Protected Health Information (PHI), particularly mental health, and/or substance use disorder information, (3) use innovative tools to support data sharing, and (4) create and adhere to an implementation plan for developing their data sharing infrastructure. WPC Pilots were also required to collect and report data on WPC interventions provided and enrollee health outcomes. This chapter expands upon initial progress described in the interim report which addressed: “to what extent did the Pilot (a) improve data collection and information sharing amongst local entities to support identification of target populations, ongoing case management, monitoring, and strategic program improvements in a sustainable fashion; and (b) achieve the approved application deliverables relating to data collection and information sharing?” Specific data sharing elements as outlined in prior UCLA assessments (e.g., PY 4 (2019) Care Coordination Policy Brief and the associated Pilot Case Studies) were identified as critical for facilitating effective cross-sector care coordination and included: (1) formal agreements that defined terms and conditions of data sharing with key partners; (2) a universal consent form to reduce barriers to sharing enrollee-level data; (3) use of an electronic data sharing platform that includes key information such as comprehensive care plans; (4) medical, behavioral health and social service use data; and (5) capacity to track and report care coordination activities. Ideally, care coordinators could also access this data sharing system to (6) view and enter data (7) remotely (e.g., in the field) and (8) in real-time. [1], [2], [3] Since the interim report, Pilots made significant progress in developing data sharing infrastructure and preparing their information technology platforms to support the transition to Cal-AIM. Data sources for this chapter included PY 3 (2018), PY 5 (2020), and PY 6 (2021) Lead Entity surveys and PY 6 follow-up interviews with leadership and frontline staff of all 26 Pilots. Additional qualitative data around challenges and solutions was provided in 25 WPC mid-year and annual narrative reports. The PY 5 and PY 6 data sources included both updates on program implementation since the interim report as well as clarification and further detail on activities conducted since the start of WPC. For additional detail on data sources and methodology, please see Appendices C, D, E, and F. By. Health Information Technology and Data Sharing Infrastructure | Whole Person Care Final Evaluation Report eer UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 28: Frequency of Data Sharing Agreements with Lead Entity and Specific Types of Key Partners, PY 5 Mec-cal managed cre plan 2: Health care organization [20 Mental health treatment agency [i 73 Ss Non-housing social service agency ns ;; Substance abuse treatment agency (I y 1> Source: PY 5 Lead Entity (LE) Survey (n=25), June-August 2020. Notes: Napa did not complete a PY 5 LE survey and therefore is not included in the analysis. “Non-housing social services agency” includes organizations such as: county and/or community-based social services, employment and human service agencies, aging and adult services. Additionally, enrollee consent was required to share private health data amongst care providers and participating partner organizations. Pilots took a wide variety of approaches to the development of consent forms, which often accompanied the process of enrolling into the program. Some Pilots, such as San Joaquin and Los Angeles, implemented a segmented consent form, which allowed enrollees to choose which types of data they felt comfortable sharing, such as consent to share medical, mental health, or substance use history. In PY 5 LE surveys, LEs reported using universal consent forms for data sharing with which key partners (Exhibit 29). Most LEs utilized universal consent forms with health care providers (18) and non-housing social service agencies (15). In PY 6 follow-up interviews, LEs emphasized access to substance use disorder (SUD) treatment data was often challenging due privacy restrictions under Title 42 of the Code of Federal Regulations (CFR) Part 2. Health Information Technology and Data Sharing Infrastructure | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Data Sharing Agreements and Enrollee Consents In the interim report, LEs reported using different mechanisms to facilitate data sharing with their partners, including Memorandums of Understanding (MOUs) and Business Associate Agreements (BAAs). These agreements ensured accountability to Health Insurance Portability and Accountability Act (HIPAA) regulatory requirements and created liability between the participating parties. As indicated in the PY 3 LE survey, few (4 of 27) LEs had established data sharing agreements with key partners prior to WPC. By the PY 5 LE survey, the majority of LEs (20 of 25) had data sharing agreements in place with all key partners and the other five had these agreements with some key partners. Key partners were defined as those who have a high awareness of the WPC program structure and goals. These partners were actively involved in the program, either through day-to-day implementation or strategic planning, and could include a combination of internal and external partners. “| think Whole Person Care has kind of set the precedent for using data from multiple sources because in the past each division kind of focused on their own data from their system.” -San Mateo By PY, in surveys, LEs most often reported having these agreements with Medi-Cal managed care plans (MCPs; 21 of 25), followed by health care providers (20) and mental health treatment agencies (18; Exhibit 28). Agreements with other key partners were less common, but not insignificant. Data sharing agreements with MCPs were notable because many LEs received enrollee level data from MCPs for the purposes of targeted identification, outreach, and engagement. During PY 6 and in follow-up interviews, LEs frequently described data sharing agreements as time-intensive to successfully implement for WPC due to a wide variety of Pilot-specific challenges. For example, LEs expressed difficulty working with some partner organizations that did not actively promote a data sharing culture and challenges reaching consensus amongst participating parties on appropriate language for formal contracts. Furthermore, LEs reported that it was often easier to share data within the county departments or internal organizations than with key partners that were outside their umbrella organization. Some Pilots, such as Contra Costa, Mendocino, and Sacramento, offered incentive payments for executing data sharing agreements, which encouraged participation particularly with community-based partners. Whole Person Care Final Evaluation Report | Health Information Technology and Data Sharing [IEE Infrastructure UCLA Center for Health Policy Research i i ey 1 \gPley ed Health Economics and Evaluation Research Program Soa all Exhibit 29: Frequency of Use of Universal Consent Form for Data Sharing by Key Partner Type, PY'S Heath care organicstion i Non-housing social service agency (ES ;; Housing gency TT 3: Mental health treatment agenry (ns :; Medi-Cal managed care plan [rs ; 3 Substance abuse treatment agenry (10 Source: PY 5 Lead Entity (LE) Survey, n=25, June-August 2020. Notes: Napa did not complete a PY 5 LE survey and therefore is not included in the analysis. “Non-housing social services agency” includes organizations such as: County and/or community-based social services, employment and human service agencies, aging and adult services. Exhibit 30 provides selected examples of how LEs implemented various data sharing agreements and enrollee consent forms to support WPC activities. Exhibit 30: Selected Examples of Data Sharing Agreements and Enrollee Consent in WPC, PY 6 WPC Pilot Selected Examples Santa Cruz In Santa Cruz, many agreements existed prior to WPC because of the county’s health information exchange. This previously established infrastructure facilitated data sharing for WPC care coordination activities. As a result of collaborative discussions facilitated through WPC, participating partners expanded upon existing data agreements to include data on social determinants of health, in addition to medical data. Contra Costa During initial WPC engagement, prospective enrollees signed (1) a consent for treatment form, which covered data sharing amongst all agencies within the comprehensive health system (e.g., behavioral health, public health, emergency medical services, and housing) and (2) a universal release form, modeled from an existing program in Contra Costa, which allowed the Pilot to share data amongst external and internal partners. San Joaquin San Joaquin utilized a segmented consent form which allowed enrollees to choose what agency’s data could be shared for the purposes of care coordination. Frontline staff emphasized that WPC demonstrated the necessity of such an approach as it facilitated comfort and trust building with enrollees. Los Angeles Los Angeles required partners to sign a business associate agreement with a data-sharing element. Enrollees were required to sign a universal consent form in order to participate in WPC, which was segmented to allow enrollees to opt- out of sharing particular data elements, such as data covered by the Code of Federal Regulations (CFR) Part 2, mental health history, and/or HIV test results. Whole Person Care Final Evaluation Report | Health Information Technology and Data Sharing [iE Infrastructure UCLA Center for Health Policy Research Decem vleyyd . ‘ Health Economics and Evaluation Research Program WPC Pilot Selected Examples The universal consent authorized Los Angeles to share data for a five-year period, even after disenrollment or graduation from the WPC program. Mendocino Enrollees in Mendocino signed a release of information form that was developed collaboratively by all partnering agencies. This form was later utilized for Project Roomkey and Project Homekey during pandemic response. Source: PY 6 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=26), June-September 2021. Data Sharing Platforms and Tools to Support Care Coordination In PY 5 LE surveys, Pilots reported frequently used multiple data sharing platforms and tools to support care coordination (Exhibit 31). The majority of Pilots (19 of 25) indicated they had acquired and/or developed a care management platform to facilitate daily workflows and ensure appropriate capture and tracking of important enrollee-level data such as demographic characteristics, encounter notes, and attempts to contact. Many of the care management platforms were intended to be web-based, which would allow the care coordination team to access enrollee data and case notes in the field and when working directly with the enrollee. Sixteen Pilots utilized electronic health or medical records (EHRs/EMRs) to support care coordination activities. Some case management platforms, as described above, were integrated into existing EHRs/EMRs. Smaller Pilots often had success with simple cloud-based storage, which allowed the care team to view and edit important enrollee documents, such as the care plan. This tool was used by 12 Pilots. Seven Pilots utilized centralized repositories, such as a Health Information Exchange (HIE), to access community-wide longitudinal enrollee records. Tools within data sharing platforms offered increased functionality. Seventeen Pilots utilized an event-based alert system for emergency department or hospital visits. This data allowed frontline staff to make real-time strategic and informed decisions regarding enrollees’ care. Ten Pilots utilized query-based exchanges to access individual enrollee level data. Streamlining access to enrollee data was a common goal of WPC. By PY 5, 17 Pilots reported they could access enrollee’s comprehensive care plan, needs assessment, and referrals in the same location (data not shown). Health Information Technology and Data Sharing Infrastructure | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research errr Health Economics and Evaluation Research Program Exhibit 31: Platforms and Tools Used to Support WPC Data Sharing, PY 5 Care management platfor) a 19 E Electronic medical/health record | 16 g = Simple cloud-based storage box or drive is 17 Centralized repository i 7 ui Event-based alert syste i 17 3 8 2 Query-based exchange tool I 10 Source: PY 5 Lead Entity (LE) Survey, n=25, June-August 2020. Note: Napa did not complete a PY 5 LE survey and therefore is not included in the analysis. Access to Data Sharing for Care Coordination Team and Other Staff Although access to care management platforms and event-based notifications varied by key partners, Pilots reported that access was most commonly granted directly to the care coordination team, followed by staff at county health care and mental health service agencies (Exhibit 32). No Pilots reported access by law enforcement or probation staff. Exhibit 32: Type of Staff or Partner and Access to Care Management Platform and Event-Based Notifications, PY 5 County health care or public health as 25 county mental eth TT | 4 Community based organization TT 11 County substance use disorder treatment County human services Local housing authority J 3 Access by staff outside of care team Care management platform (n=19) _ tl Event based notifications (n=17) Source: PY 5 Lead Entity (LE) Surveys, n=25, June-August 2020. Note: Napa did not complete a PY 5 LE survey and therefore is not included in the analysis. Whole Person Care Final Evaluation Report | Health Information Technology and Data Sharing [ERA Infrastructure errr) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program “Some of the technology investments will only continue to grow and deepen... when we first started, the default ... was ‘it's easier just not to do it... and because I'm not certain if | can share it or not, we're just not going to share it’... We've knocked down a few of those silos... [now] we have visibility into the behavioral health record and we actually do our documentation in their health record.” -Ventura For care team staff, the majority of Pilots reported having access to data on emergency department and hospitalizations (21), other medical care (19), temporary housing/shelter (17), and mental health encounters (17; Exhibit 33). Pilots less frequently reported point of care access for all the types of enrollee-level data inquired about in the survey. Exhibit 33: Type of Data Accessible to Care Coordination Staff, PY 5 Emergency department or hospital utilization Medical - a na © XN 8 ” 5 Other medical care B BI B q Temporary housing/shelter Social service benefits eligibility | » & Social services | 2 B Social service encounters 4 So a a Justice system involvement Justice Mental health encounters Bl So B Su Behavioral | © Substance use encounters Any access m Access at point of care Source: PY 5 Lead Entity (LE) Surveys, n=25, June-August 2020. Notes: Examples of "point of care" include ability to access in the field or during meetings with clients. “Other medical service encounters” includes those other than emergency department or hospital utilization. Examples of “social service encounters" include Child Protective Services, in-home supportive services, examples of "justice system involvement" include jail admission and discharge data. “..[We have] an immediate email notification system that tells us when someone has gone to the emergency room or to the hospital inpatient... ... that way we know when and how to help the most.” -Placer Health Information Technology and Data Sharing Infrastructure | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research i i ey lala Health Economics and Evaluation Research Program Seem Exhibit 34 provides selected examples of how case management software and real-time data sharing facilitated care coordination activities. Additional detail is provided in the Pilot specific mini analyses (see Appendix L). Exhibit 34: Selected Examples of Data Sharing Tools and Platforms to Support Care Coordination in WPC, PY 6 WPC Pilot Selected Examples Alameda Alameda’s primary mechanism for data sharing with partners was a community health record (CHR) that consolidated client data and was accessible by all partners upon establishment of a data sharing agreement. The CHR was powered by a social health information exchange platform that integrated data from the LE’s electronic health record (Epic) and case management tools, as well as the homeless management information system and county jail incarceration information. Alameda also utilized a tool called “EDie” to notify and alert frontline staff in real-time when WPC enrollees had an emergency department encounter. Contra Costa The primary mechanism for data sharing with external partners was a care management platform embedded within the electronic health record (EHR) called “Care Everywhere”, which integrated data across county departments and affiliated health system partners. Care coordinators in Contra Costa received real-time notifications when WPC enrollees visited the emergency department or an in-patient setting at any hospital within the local geographic area. Kings Kings adopted a care coordination platform called “Effort to Outcomes” (ETO) from Social Solutions. ETO allowed the care team to input case notes, record care coordination services, and build reports, with access to medical, behavioral health, and social services data in a single location. Los Angeles Los Angeles developed their case management platform “CHAMP”, which facilitated care coordination by providing eligibility screenings, enrollment documentation and assessments, stored enrollee documents (e.g., universal consent form) and care plan, and comprehensively documented case related information (e.g., attempted contacts with enrollees, case notes). Throughout the Pilot, Los Angeles made continuous improvements and modifications to the platform based on user feedback. The platform included applications that facilitated day-to-day workflows. For example, the team developed a dashboard that displayed enrollees’ “SMART” goals and associated action steps. Through the dashboard, the care team could communicate on these goals and monitor their status, reducing redundancy and preventing duplication of services. Marin Marin’s care coordination platform called “Wizard” was viewed as a critical tool for allowing the care coordination team to stay up to date about an enrollee’s current goals, appointments, progress, and future scheduling. Communication amongst the care team could occur through in-platform HIPAA compliant messages or through a chat function. The platform featured real time alerts for care coordination staff. Sacramento Sacramento utilized a care management platform called “Shared Care Plan” which helped share enrollee medical, behavioral health, and other information between designated staff at service partner organizations. Source: PY 6 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=26), June-September 2021. Whole Person Care Final Evaluation Report | Health Information Technology and Data Sharing Infrastructure erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Use of Incentives to Promote Data Sharing As indicated in PY 6 LE surveys, 18 LEs utilized contract incentives with partners to promote the development of data sharing infrastructure (e.g., increased functionality within existing or acquisition of new case management platform, EHR, or HIE; data not shown). Of all contracting incentives presented in the survey, incentives to promote the development of data sharing infrastructure were rated the highest as both having achieved their desired goals (7.5 out of 10) and in likelihood of continued use (8.7; where 0 = “not at all” and 10 = “highly”). Challenges Related to Data Sharing and Reporting Exhibit 35 summarizes the most frequently identified challenges related to data sharing and reporting by program year as presented by Pilots in bi-annual narrative reports. Overall, the most common theme across the duration of WPC was challenges related to lack of buy-in and/or readiness from partners and frontline staff for new data systems or integrating existing data systems (77 unique mentions across reporting periods by 23 Pilots; data not shown). Many partners had different and very particular data needs and it was challenging to find a platform that met everyone’s specifications. Frontline staff were resistant to access multiple systems in order to input required information for reporting and tracking of care coordination services. This theme was observed more frequently over time as Pilots formalized their data sharing systems, with five mentions in PY 2, 21 mentions in PY 4 and PY 5, and 19 mentions in PY 6. Pilots also expressed inability to access necessary data to facilitate WPC activities (68 unique mentions across reporting periods by 24 Pilots; data not shown). The majority of these Pilots did not have real-time access to Medi-Cal coverage which would be useful in verifying prospective enrollee’s eligibility and preventing unnecessary churn from Medi-Cal and the WPC program. There was an increase over time as Pilots ramped up outreach, engagement, and enrollment, with two mentions in PY 2, a peak of 20 mentions in PY 4, and 16 mentions in PY 6. Pilots reported inability to implement data sharing systems and/or integrate data from existing systems as intended (65 unique mentions across reporting periods by 22 Pilots; data not shown). WPC Pilots noted that data sharing often required integrating data from disparate sources. For example, frontline staff had to assimilate data from different electronic health records or administrative databases so they could comprehensively understand the needs of an enrollee in order to make an informed care decision on what the enrollee required. Vendor delays, designing and/or purchasing technology that allowed for real-time data storage, and access by multiple agencies and users were described as challenges, both in terms of cost and in terms of the identification and selection process. However, there was a degree of resolution il Health Information Technology and Data Sharing Infrastructure | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program over time, as WPC Pilots resolved issues with vendors and worked collaboratively with partners to achieve integration. There was a peak of 18 mentions in PY 3, and only 10 mentions in PY 6. A consistent theme across reporting periods was legal and cultural barriers to data sharing, such as risk aversion and differing interpretations of laws and regulations (60 unique mentions across reporting periods by 22 Pilots; data not shown). Fear of violating the HIPAA or other data privacy laws was cited as contributing to a reluctance to share data, even across departments within the same agency. WPC Pilots described misunderstandings and differing interpretations among partners regarding what data could be legally shared as a barrier to successful data sharing. Issues with data reporting (e.g., tracking care coordination activities and services provided through WPC) largely decreased over time, although it was a challenge that almost all Pilots faced (43 unique mentions across reporting periods by 24 Pilots; data not shown). WPC Pilots reported challenges in ensuring consistency of data being collected across partners and noted a considerable effort to reconcile different data sources and develop new documentation strategies. These efforts resulted in progress towards better data collection for reporting purposes (e.g., DHCS required metrics, internal dashboards for monitoring progress). The interim report and narrative report updates provide additional examples of data sharing and reporting challenges by Pilot. Whole Person Care Final Evaluation Report | Health Information Technology and Data Sharing [REHE Infrastructure UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 35: Data Sharing and Reporting Challenges Among WPC Pilots by Program Year, PY 2 — PY6 Lack of buy-in and/or readiness 2 Unable to access necessary data 7 20 16 Implementing data sharing systems and/or integrating data ” 18 as intended TL 10 0 Legal and cultural barriers to data sharing a i 12 16 oe 11 Data reporting issues PY2 MPY3 MIPY4 MPYS MIPY6 Source: WPC Mid-Year and Annual Narrative Reports, PY 2 (2017) - PY 6 (2021). Notes: Numbers indicate WPC Pilots that mentioned the thematic challenge at least once within the given program year. PY 2 = 2017, PY 3 = 2018, PY 4 = 2019, PY 5 = 2020, and PY 6 = 2021. Successes in Data Sharing and Reporting In PY 5 LE surveys, LEs perceived relatively high impact of WPC on improving data sharing between the LE and partners (7.9 out of 10; data not shown). Health Information Technology and Data Sharing Infrastructure | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research eer) Health Economics and Evaluation Research Program Exhibit 36 summarizes the most frequently identified successes related to data sharing and reporting by program year as presented by Pilots in bi-annual narrative reports. Successes in data sharing and reporting often directly reflected a response to the challenges detailed above. Overall, the most common theme across the duration of WPC was progress in sharing data across sectors, particularly between LEs and Medi-Cal managed care organizations, local homeless management information systems (HMIS), substance use disorder programs, and county behavioral health departments (108 unique mentions across reporting periods by all 25 Pilots; data not shown). Pilots consistently reported successes in this area in each reporting period (range of 19 to 24 Pilots per reporting period). Pilots also reported successes in developing new software, data sharing platforms, and/or data repositories (105 unique mentions across reporting periods by all 25 Pilots; data not shown). These included: developing a new care management platform, utilizing temporary data systems while longer-term solutions were still being developed, moving forward with procurement processes for data systems, and/or expanding functionality within existing systems including developing additional forms and prompts within EHR. Pilots also consistently reported successes in this area in each reporting period (18-23 Pilots per reporting period). Pilots also emphasized setting up infrastructure needed to support data-informed decision making or quality improvement efforts (93 unique mentions across reporting periods by all 24 Pilots; data not shown). For example, providing instant notifications when enrollees checked into the ED or dashboards to help track enrollee progress on relevant metrics allowed frontline staff and management to make real time strategic and informed decisions regarding enrollee care. Use of these tools increased over time as Pilots formalized and better integrated data systems into existing workflows, with 22 Pilot mentions in PY 6 (compared to only 11 in PY 2). Less common themes related to successes in data sharing included: meeting external reporting requirements (e.g., enrollment, utilization, and metrics to DHCS) and implementing data sharing agreements and consents with WPC partners. Pilots often found early success with these components benefited them throughout the course of WPC. Whole Person Care Final Evaluation Report | Health Information Technology and Data Sharing [REE Infrastructure December 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 36: Data Sharing and Reporting Solutions Among WPC Pilots by Program Year, PY 2 — PY 6 Data sharing across multiple systems ao 20 : 23 Developing a new software, platform, and/or repository 2 20 i Using data informed decision making to support 5 implementation processes or quality improvement efforts 23 22 16 Required reporting (e.g., enrollment, utilization, metrics) 16 9 5 Implementation of data sharing agreements and/or “ universal consents 8 7 PY2 mPY3 mPY4 mPYS MPY6 Source: WPC Mid-Year and Annual Narrative Reports, PY 2-PY 6. Notes: Numbers indicate WPC Pilots that mentioned the thematic challenge at least once within the given program year. PY 2 = 2017, PY 3 = 2018, PY 4 = 2019, PY 5 = 2020, and PY 6 = 2021. Please refer to the interim report and narrative report updates for specific examples of data sharing and reporting solutions as presented by Pilot. C2 Health Information Technology and Data Sharing Infrastructure | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research en eT r) Health Economics and Evaluation Research Program Chapter 4: WPC Enrollment Processes, Size, and Patterns WPC Pilots were required to identify eligible Medi-Cal beneficiaries using pre-defined inclusion criteria, enroll them in WPC, and engage enrollees in care. This chapter reports on strategies used by Pilots to identify, enroll, and engage eligible Medi-Cal beneficiaries in WPC, as well as summarizes facilitators, barriers, and lessons learned. In addition, this chapter reports on the resulting enrollment size and patterns for the overall program and by target population. Key findings from the interim report are summarized when data have not changed. Data sources for this chapter include PY 5 (2020) and PY 6 (2021) Lead Entity (LE) surveys and PY 6 follow-up interviews with leadership and frontline staff of 26 Pilots. Data from 25 narrative reports submitted by Pilots to DHCS were also included in the following analyses. The PY 5 and PY 6 data sources included clarification on identification, engagement, and enrollment activities conducted since the start of WPC. Since the interim, new and further detail is available. The data source for enrollment size and pattern analyses were WPC Quarterly Enrollment and Utilization Reports from PY 2 (2017) to PY 6. For additional detail on data sources and methodology please see Appendices A and B. WPC Processes for Identification, Engagement, and Enrollment of Eligible Medi-Cal Beneficiaries Identifying Prospective Enrollees In PY 6 LE surveys, WPC Pilots reported using a range of strategies to identify eligible Medi-Cal beneficiaries. Nearly all Pilots (24 of 26) utilized referrals from WPC partner agencies, which came from diverse sources such as Medi-Cal managed care plans, hospitals, clinics, and law enforcement. Many Pilots (20) also accepted referrals from other agencies not participating in WPC. In PY 6 follow-up interviews, Pilots emphasized the importance of developing and “Some of these folks have never been engaged ... We're finding people on the streets who've been homeless for 20 years and have not been engaged in care for that length of time. ... | think a lot of Pilots learned ... that there is an unknown group of very vulnerable people out there who weren't accessing services because we were all focused on the high utilizers. We inadvertently found these low utilizers with extremely high needs.” -San Mateo Whole Person Care Final Evaluation Report | WPC Enrollment Processes, Size, and Patterns erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program maintaining relationships with other agencies (e.g., hospitals, emergency departments) to establishing strong referral streams. As indicated in PY 6 LE surveys, the next most commonly used strategy for identifying eligible beneficiaries was through shelter/street- or other field-based (e.g., hospital/medical care delivery facility) outreach (22). Half the Pilots (13), including Kings, Santa Cruz, and Sonoma, also allowed potential enrollees to refer themselves or their peers into the program based on interest and individual assessment of eligibility. Less common identification methods included: target population lists provided by Medi-Cal managed care plans (10) and predictive modeling or risk-based algorithms/scores (8). Exhibit 37 shows the perceived effectiveness of these strategies for identifying prospective enrollees on a scale from 0 to 10 (where 0 = not at all effective and 10 = highly effective). Pilots rated referrals from WPC partner agencies as more effective (average rating of 7.7 out of 10) than referrals from other (non-WPC partner) community-based agencies (6.5). In PY 6 follow-up interviews, Pilots noted that WPC partner agencies often had a better understanding of Pilot enrollment criteria (e.g., primary target populations) and program offerings and thus were more likely to make appropriate referrals. Some Pilots, such as Mendocino, iteratively edited form fields on WPC referral forms to clarify eligibility criteria with partners and ensure receipt of appropriate referrals. In PY 6 LE surveys, Pilots also rated field-based outreach (e.g., at hospitals) as highly effective (average rating of 7.5 out of 10), with the added benefit of allowing for warm-handoffs to WPC. Pilots rated use of predictive modeling or risk-based algorithms and target population lists provided by Medi-Cal managed care plans to identify prospective enrollees slightly lower in terms of effectiveness (6.9 and 6, respectively), due to challenges with follow-up and engagement of prospective enrollees. A handful of Pilots, such as Contra Costa, experienced higher effectiveness with risk-based algorithms. Prior to WPC, Contra Costa had already integrated data from multiple systems. Allowing individuals to refer themselves or peers was considered least effective (4.4), as these individuals often did not meet Pilot eligibility criteria. Ea WPC Enrollment Processes, Size, and Patterns |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 Exhibit 37: Most Common Strategies for Identifying Prospective Enrollees and Pilot Perceived Effectiveness, PY 6 Maximum rating Referrals from WPC partner agencies (n=24) Hospital or other medical care delivery facility outreach (n=22) Street- or shelter-based outreach (n=22) Predictive modeling or risk-based algorithm/scores (n=8) Referrals from other (non-WPC partner) agencies in the community (n=20) Target population lists provided by Medi-Cal managed care plans (n=10) Allowing individuals to refer themselves or peers (n=13) SS « eS 77 7: 7: aT < ; eS « a Source: PY 6 Lead Entity (LE) Survey (n=26), May-June 2021. Notes: Numbers in parenthesis represent the number of Pilots who indicated they utilized a given strategy. If the Pilots used the identification strategy, they were asked to rate effectiveness on a scale from 0 to 10, where 0 = not at all effective and 10 = highly effective. Whole Person Care Final Evaluation Report | WPC Enrollment Processes, Size, and Patterns 97 December 2022 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program “.. One thing that really helped is we were able to really get buy-in from our hospital partners... we had workflows in place specifically for the hospitals where we would try to get a CHW out there within a couple of hours so that we could do a warm handoff before the individual ...[left] the [ED]. The hospitals were so bought into that, that they created their own referral form. ...we played a really big part .... And | do think that was a huge success for us because they were really bought into it including, not just our main points of contact with the community engagement folks, but all the way through the discharge workers at the hospitals. -Sacramento Exhibit 38 highlights specific approaches by Pilots to identify prospective enrollees; these examples demonstrate the variety of strategies utilized across WPC Pilots. Exhibit 38: Selected Examples of WPC Pilot Strategies to Identifying Prospective Enrollees Strategy Pilots that Utilized Strategy Selected Examples Referrals from WPC partner agencies (n=24) All Pilots, except Contra Costa San Bernardino Marin relied on their partnership with federally qualified health centers to receive referrals and real-time data on prospective enrollees. Mendocino relied heavily on partner referrals, particularly medical and behavioral health providers. Mendocino’s referral form clearly outlined program eligibility criteria and encouraged the referring party to gauge the prospective enrollee’s interest and potential for engagement with WPC prior to submitting the referral. Prospective enrollees were already educated on the basics of WPC by the referring partner, which facilitated enrollment and future engagement. Hospital or other medical care delivery facility outreach (n=22) All Pilots, except Mendocino Riverside San Francisco Santa Cruz Sacramento attempted to respond to referrals from emergency department visits within two hours and to respond to referrals of hospital inpatients within 24 hours, which allowed them to identify and engage prospective enrollees while they were still in systems of care and to receive a warm handoff from the provider or care team to WPC frontline staff. Alameda utilized care transitions nurses at the County's Community Health Center to evaluate whether individuals entering the hospital or transitioning to a skilled nursing facility met WPC enrollment criteria. If enrollment criteria were met, the individual would be connected directly with a WPC community health worker. Street- or shelter- based outreach (n=22) All Pilots, except Contra Costa Mendocino Riverside Santa Cruz Santa Clara partnered with the Valley Homeless Healthcare Program, which used mobile vans to conduct regular visits to areas with relatively high concentrations of homeless individuals. This increased WPC enrollment through in-field outreach. In San Francisco, street medicine and shelter health worked to identify prospective enrollees for WPC in places where individuals experiencing homelessness typically frequented, Ei WPC Enrollment Processes, Size, and Patterns | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ely ed Health Economics and Evaluation Research Program Strategy Pilots that Utilized Strategy Selected Examples including shelters and overnight residences, as well as on the street and in encampments. Target population lists provided by Medi-Cal managed care plans (n=10) Mariposa (SCWPCC) San Benito (SCWPCC) San Bernardino San Joaquin Santa Clara Solano Sonoma Ventura Kern Due to law enforcement’s strong working relationship with the Kings King’s WPC program, many justice-involved individuals referred Los Angeles themselves to the program after hearing positive outcomes Mariposa (SCWPCC) and success stories through word-of-mouth. Mendocino Allowing individuals Monterey To identify prospective enrollees for their substance use to refer themselves San Benito (SCWPCC) | programs, Los Angeles utilized their substance abuse services or peers (n=13) San Diego help hotline. At the end of the call, a high-level overview of Santa Clara WPC was provided, and callers were asked whether they were Santa Cruz interested in WPC. If the caller expressed interest, the Solano prospective enrollee was assigned to a community health Sonoma worker for subsequent follow-up. Ventura Kern Kern received lists of individuals who met WPC enrollment Los Angeles criteria from managed care plans; they matched those lists with daily reports of people who were released from the local county jail to identify eligibility for WPC. Predictive modeling or risk-based algorithms/scores (n=8) Contra Costa Kern Los Angeles Placer San Bernardino San Diego Santa Clara Sonoma Contra Costa employed a predictive risk model to identify prospective enrollees. The model factored in utilization of services, health records, behavioral health issues, and social factors to generate a list of the top 23,000 adults expected to have an avoidable emergency department visit or hospitalization. The higher risk individuals were prioritized for WPC enrollment. The model was refined throughout WPC, integrating lessons learned. Until PY 6, San Bernardino employed a scoring mechanism based off data from the health system, public health, and Medi-Cal managed care plans, which ranked prospective enrollees based on utilization of emergency department, inpatient hospital stays, and urgent care visits. Source: PY 6 Follow-up Interviews with Lead Entities and Frontline Staff (n=26), June-September 2021. Note: SCWPCC is the Small County Whole Person Care Collaborative. Whole Person Care Final Evaluation Report | WPC Enrollment Processes, Size, and Patterns ee errr UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Determining Eligibility In PY 6 LE surveys, Pilots were asked to identify their methods for determining WPC eligibility. Pilots most often utilized existing data to determine eligibility, including electronic medical records (EMRs) or other medical data (21 of 26) and information provided by WPC partners (e.g., SMI/SUD diagnosis, homelessness indicators; 21). Other common methods for determining eligibility included staff assessment using standardized tools (20) and care coordinator assessments (18). Exhibit 39: Method for Determining WPC Eligibility Following Identification of Prospective Enrollees, PY 6 Electronic medical record or other medical data [i 2: Information provided by WPC partners |i 2) Staff assessment using standardized too] I 20 Care coordinator assessment [I 13 Source: PY 6 Lead Entity (LE) Survey (n=26), May-June 2021. Hele WPC Enrollment Processes, Size, and Patterns |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ely Health Economics and Evaluation Research Program Methods for determining WPC eligibility varied by target population (Exhibit 40). Within the target population of high utilizers, they were most often identified using EMRs or other medical data (82%), followed by information provided by WPC partners (76%). Staff standardized screening were most often used within the SMI/SUD target population (90%) and homeless or at-risk-of-homelessness target populations (64% and 78%, respectively). Exhibit 40: Method for Determining Eligibility for WPC within Primary Target Population, PY 6 82% High utilizers (n=17) ry 78% 41% 43% 50% . SMI/SUD (n=10) 90% 60% 33% At-risk-of-homelessness (n=9) 7 44% 50% 75% Chronic conditions (n=4) 50% a5% 25% Electronic medical record or other medical data m@ Information provided by WPC partners @ Staff assessment using standardized tool Care coordinator assessment Source: PY 6 Lead Entity (LE) Survey (n=26), May-June 2021. Notes: Numbers in parenthesis indicate the number of Pilots who indicated a given target population as a primary target population. The primary target population is defined as a key demographic of focus, one that WPC Pilots designed their services, infrastructure, and processes around; Pilots could serve multiple primary target populations. SMI/SUD is serious mental illness/substance use disorder. Whole Person Care Final Evaluation Report | WPC Enrollment Processes, Size, and Patterns JileH errr UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Enrollment Approach In PY 6 LE surveys, the majority of WPC Pilots indicated enrolling directly at health care facilities (20 of 26) or on the street, at shelters, or community-based locations (20; Exhibit 41). Pilots rated these enrollment methods as the most effective (average rating of 8.2 and 7.7 out of 10, respectively). Pilots emphasized partnership networks and structure developed through WPC greatly facilitated this in PY 6 follow-up interviews. Nineteen Pilots utilized warm handoffs at co-located organizations (data not shown). Pilots reported they would co-locate WPC staff at points of care or transition (e.g., hospitals, clinics, jails) when possible and use warm handoffs as an opportunity to establish relationships and build trust. Fewer Pilots utilized strategies such as telephonic outreach and auto-enrollment (i.e., enrollment based on defined criteria and notification by mail; 15 and 3, respectively). These methods were used in attempts to expand program reach but were considered least effective, likely due to lack of personal engagement and connection established through in-person contact. Exhibit 41: Pilot Perceived Effectiveness of WPC Enrollment Method, PY 6 Voxnvn tn 00 Enrollment at health care facilities (n=20) (NN 2 On street, at shelter, or other community based location Ss 7) (n=20) Enrollment by telephone (n=15) (T0008 so Auto-enrollment and opt-out (n=3) (NN s.2 Source: PY 6 Lead Entity (LE) Survey (n=26), May-June 2021. Notes: Numbers in parentheses represent the number of Pilots who indicated they utilized a given enrollment method. If the Pilots used the enrollment method, they were asked to rate effectiveness on a scale from 0 to 10, where 0 = not at all effective and 10 = highly effective. Heyy WPC Enrollment Processes, Size, and Patterns |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erry Health Economics and Evaluation Research Program Enrollee Engagement and Retention After enrollment into WPC, care coordination staff employed engagement techniques to ensure enrollee retention in the program. As highlighted in the interim report, WPC Pilots reported performing a variety of activities to engage beneficiaries in the WPC program, including in- person one-on-one meetings, phone calls, text conversations, street outreach, and/or home visits. Sustained enrollee engagement was an important focus of Pilots due to the nature of WPC’s vulnerable and often transient target populations. In PY 6 interviews, Pilots reported challenges in maintaining enrollee engagement, including lack of regular communication with enrollees due to inaccurate or outdated contact information and lack of cell phones, particularly amongst the homeless and the justice-involved target population. As a result, it was important for Pilots to engage enrollees in a variety of locations and through different modalities. Many Pilots commented on the importance of developing rapport and trust with enrollees. For example, Placer and San Joaquin addressed immediate needs (e.g., transportation, hygiene) before moving towards a discussion about other needs (e.g., health outcomes). “| would say the other part that’s important is really building trust and getting to know the patients. ... you must reach so many people by a certain day in order to get reimbursed. And outreaching to somebody, sometimes it takes... | don’t know how many times, months to do it, right? And that’s something that WPC has enabled us to be able to do... we have a whole process of trying to create some trust, a whole pre-outreach review, some best practices around having some ideas what a patient wants without being too overly prescriptive of what they probably want... If you know the person doesn’t come in, that might be a question, or, ‘Oh, are you needing transportation?’ So right away, you know some things and aren’t expecting the patient to just open up and tell you their entire life and every single thing that they need....” -Alameda Another key factor in engaging and promoting rapport with enrollees was having enthusiastic and dedicated care coordinators and ensuring consistent care coordinator assignment. In PY 5 surveys, 13 Pilots indicated having a single, dedicated care coordinator. Having staff with lived experience (e.g., CHWs, peer support specialists) like that of the target population was another strategy utilized to build trust. Whole Person Care Final Evaluation Report | WPC Enrollment Processes, Size, and Patterns [iBK0E} [pyre] o\-1ayAe yee UCLA Center for Health Policy Research Health Economics and Evaluation Research Program “This sub-population has a lot of trauma... So that is part of the reason why it's so hard to establish that trust and that relationship. And | think a lot of them, when they do achieve stability, that it is partly because of those relationships, that they do have that person that they can turn to whena crisis arises, that they can turn to somebody who they trust.” -Santa Clara Exhibit 42 provides selected examples of these specific strategies WPC Pilots employed to promote and maintain engagement of enrollees. Exhibit 42: Selected Examples of Strategies for Engagement of WPC Enrollees Engagement Elements WPC Pilot Selected Examples Multiple points of contact Orange Orange engaged prospective enrollees in various points of contact, including the hospital and clinics. The care coordinator also attended appointments or assisted in transportation for their enrollees. Riverside Riverside embedded a nurse in the probation office to keep in constant communication with the probation officer, so the care team was able to reach the enrollee when needed. Developing trust and rapport San Bernardino San Bernardino emphasized hiring for key traits in care coordination staff, including kindness, compassion, and respect, in order to foster relationships with their enrollees. San Joaquin San Joaquin highlighted the importance of addressing the immediate needs of prospective enrollees in order to increase trust and rapport. Consistent care coordinator assignment Kern Kern utilized a consistent care coordinator, who was responsible for initial and subsequent engagement. The consistent contact allowed for trust and rapport building throughout the life of the enrollee’s participation in WPC. Los Angeles Each enrollee in Los Angeles was assigned to a specific community health worker, which ensured consistency of communication and engagement throughout WPC enrollment. Community health workers maintained contact with enrollees through a variety of mechanisms but primarily by phone (ideally once a week). Source: PY 6 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=26), June-September 2021. ule WPC Enrollment Processes, Size, and Patterns | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research en etre) Health Economics and Evaluation Research Program “.. a lot of these people are very skeptical. They have been in and out of the system. The system has failed them over and over and over and over again, and they are very skeptical initially of how are you going to be any different? What are you going to do for us that's any more help than any other entity that I've been referred to in the past that has failed me? So we really do try to make sure that .. from the very onset ... they're following through, and that they are continuing to experience a level of continuity that they never had before.” -Kern Source: PY 6 follow-up interviews. Challenges and Successes Extensive discussion of challenges and successes related to identification, engagement, and enrollment are presented in the interim report and bi-annual narrative report updates. As discussed in these reports, early program challenges were around initial enrollment of eligible Medi-Cal beneficiaries into WPC and with maintaining enrollee engagement over time. These challenges were often attributed to the complex needs and/or transient nature of WPC target populations. Some target populations presented more complex challenges to work with, such as individuals experiencing homelessness (e.g., no permanent address, transient nature, lost phone) and justice-involved target populations (e.g., unpredictability around timing of release and difficulty contacting/locating after release from jail). Some Pilots also identified poor timeliness or accuracy of data, which was needed to support outreach and enrollment efforts. Over time, Pilots reported successfully enrolling eligible beneficiaries by employing solutions that were often directly the result of policy and procedure changes, which were motivated by observed challenges. Enrollment generally increased as Pilots’ staffing capacity and program processes improved (e.g., formalized contracts with community partners, creation of clear guidelines and protocols for referring agencies that outlined WPC Pilot goals and enrollment criteria, utilization of warm handoffs to facilitate enrollee trust and buy-in). Analyses of trends over time indicated that both challenges and successes related to identification, engagement, and enrollment were more prevalent in early reporting periods. These challenges and successes decreased in late PY 5 as LEs focused on existing enrollment as they approached the program end (December 2021) and maintained their response to the COVID-19 pandemic. During the COVID-19 pandemic, there was unanticipated improvement in enrollee engagement as Pilots found synergy with COVID-19 response and short-term housing programs. For example, Project Roomkey provided an opportunity for WPC staff to identify and consistently Whole Person Care Final Evaluation Report | WPC Enrollment Processes, Size, and Patterns [iges] erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program engage eligible enrollees while they were temporarily housed. Building upon existing partnerships, some Pilots coordinated with community-based organizations for offerings such as vaccination, testing, education, and personal hygiene pods, which provided additional opportunities for WPC outreach and engagement. WPC Enrollment Size and Patterns Enrollment into WPC began during program year 2 (PY 2), with enrollment beginning in or after January 2017 for Pilots that began implementing in January 2016 and in or after July 2017 for Pilots that began implementing in July 2016. WPC Pilots submitted Quarterly Enrollment and Utilization Reports to DHCS each quarter, from January 2017 to December 2021. These reports contained monthly records for each individual that participated in WPC. Data included enrollment status, enrollment date, disenrollment date, disenrollment reason, target population(s), homeless status, and WPC service utilization. UCLA combined data from all WPC Pilot reports, and used this data for analyses of enrollment size and patterns. UCLA defined enrollment in WPC as any individual that a WPC Pilot reported as enrolled and had an enrollment start date. The Quarterly Enrollment and Utilization Reports also included individuals that received a limited set of services from WPC Pilots (e.g., outreach and stays ina sobering center), but ultimately did not enroll into a WPC Pilot. These individuals were not included in the analysis in this chapter, as they were not enrollees, but are examined in Chapter 5: WPC Services Offered and Delivered. A number of other enrollees were also excluded from the analyses in this chapter. There were 576 individuals enrolled in more than one WPC Pilot at the same time and unknown to the Pilots. This was likely in part due to moving from one county to another. However, 1,491 enrollees with non-overlapping enrollment periods were not excluded. The final number of enrollees across Pilots was 249,378 out of a total of 247,887 unique individuals ever reported in the program. UCLA did not report data based on 10 or fewer enrollees to protect confidentiality. In addition, 11,775 (4.7%) unique enrollees had no target population reported and are not included in analyses of enrollees by target population. Enrollment Size Based on the Quarterly Enrollment and Utilization Reports of the 25 WPC Pilots, seven began enrolling in January 2017 (Exhibit 43). By the end of 2017, 16 more Pilots began enrolling. Two Pilots, San Diego and Sonoma, started enrollment during PY 3 (2018). San Diego needed additional time to establish administrative and delivery infrastructure prior to enrolling, and Sonoma delayed their enrollment due to significant wildfires in their community around the time of implementation. The Small County Whole Person Care Collaborative (SCWPCC) was ulelsay WPC Enrollment Processes, Size, and Patterns | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research . 7 December 2022 Health Economics and Evaluation Research Program formed among three counties, Mariposa, Plumas and San Benito, and started enrollment in December 2017. In September 2018, Plumas County dropped out of the SCWPCC. Due to the COVID-19 pandemic, WPC was extended for additional year (PY 6). Two Pilots, SCWPCC and Solano, dropped out of WPC at the end of PY 5. Exhibit 43: Timeline of the Start of WPC Enrollment by Pilot, PY 2 to PY 3 Jan Alameda Dec Contra Costa Mendocino Los Angeles scwPCcc Monterey (Mariposa, Orange Plumas, and San Francisco Aor dun Aug: Oct San Benito) Feb May San Mateo Placer San Bernardino —Kern Riverside San Diego Sonoma Program Year 2 (2017) Ruud ec) Mar May dul Sep ‘Nov Santa Clara Shasta Napa Kings Marin Solano San Joaquin Sacramento Santa Cruz Ventura Source: Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Enrollment start was the first month that each WPC Pilot enrolled individuals and provided services. SCWPCC is the Small County Whole Person Care Collaborative. Plumas County dropped out of SCWPCC in September 2018. SCWPCC and Solano dropped out of WPC in January 2021. Whole Person Care Final Evaluation Report | WPC Enrollment Processes, Size, and Patterns [ieleyg ee ee By the end of PY 2 (2017), a total of 50,202 individuals were enrolled in WPC (Exhibit 44). By the end of PY 6, the cumulative total to have ever enrolled in WPC increased to 247,887, with 96,416 enrolled in that month (91,001 existing enrollees and 5,415 newly enrolled in December 2021). Peak enrollment in the program occurred in June 2021 with 100,968 enrollees. As the program came to an end, the monthly current enrollment decreased for the first time starting in July 2021. Monthly new enrollment in the program ranged from 1,432 in February 2017 to 8,502 in January 2017. The average new enrollment per month was 5,068 (data not shown). Exhibit 44: Unduplicated Monthly and Cumulative WPC Enrollment, PY 2 to PY 6 mam Existing Enrollees per Month Mm New Enrollees per Month 247,887 163,646 —— Cumulative Enrollment 214,917 108,864 100,968 96 aig 50,202 tr | | | | | | SIPIRIS (BBS 8S SIS/AiS/S BA S/S\BIRIBA 5 3 “7 a PY2 PY3 PY4 PYS PY6 Source: Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 247,887 unique first enrollments into any WPC Pilot. Does not include re-enrollments or enrollments in a second WPC Pilot. Excludes individuals who received outreach or other WPC services, but did not enroll. He) WPC Enrollment Processes, Size, and Patterns |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program HISSS HIRSH! Exhibit 45 shows total WPC enrollment during the program ranged from 143 enrollees in the SCWPCC to 76,107 enrollees in Los Angeles. Of the 25 WPC Pilots, nine Pilots had enrollment numbers under 1,000 enrollees and six Pilots had enrollment over 10,000 enrollees. Given the staggered implementation of the program, the length of time that each WPC Pilot was actively enrolling individuals into their Pilots varied. Exhibit 45: Total Enrollment in WPC by Pilot, PY 2 to PY 6 LOS ADCS LE 76,107 Contra COSt2 aan 57) S)() Alancd A 30) 727 San FranciscO A 22,749 Orange EE 13,861 Riverside 13,531 Santa Clara 7,431 Sonoma mum 4,181 San Mateo mmm 4,163 SanJoaquin mmm 3,201 Kern mm 2,773 Sacramento mm 2,345 Marin mm 1,881 San Bernardino m 1,552 Ventura m 1,520 Kings ™ 1,037 San Diego = 958 Monterey = 836 Napa = 771 Santa Cruz § 603 Shasta § 581 Placer § 501 Mendocino 1 494 Solano | 247 SCWPCC 143 Source: Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 249,378 unique first enrollments into a WPC Pilot. Excludes individuals who received outreach or other WPC services but did not enroll. SCWPCC is the Small County Whole Person Care Collaborative. Whole Person Care Final Evaluation Report | WPC Enrollment Processes, Size, and Patterns Jee) ee err UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Enrollment Patterns As of the end of WPC (December 2021), 29% of WPC enrollees had stayed continuously enrolled in the program since their initial enrollment (Exhibit 46). The percent of enrollees that stayed continuously enrolled varied by Pilot, with some Pilots having less than 10% of enrollees continuously enrolled (SCWPCC, Shasta, Orange, Solano, and Contra Costa) and other Pilots having over 80% of enrollees continuously enrolled (Kern and Alameda; data not shown). Exhibit 46: Patterns of Enrollment and Disenrollment in WPC, PY 2 to PY 6 Pieler Stayed Disenrolled, 54% el ia ue Enrolled Multiple 14% Times, 17% te Leroy itd Cel Cay Enrolled, 29% Source: Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 249,378 unique enrollments into a WPC Pilot. Continuously enrolled includes individuals that never disenrolled from the program. Over the course of the program, 71% of WPC enrollees disenrolled at least once (Exhibit 46). Enrollees could reenroll into the program if they met the criteria for enrollment at a future date. Data showed that most enrollees disenrolled and stayed disenrolled (54%) while others enrolled multiple times (17%). Of those that enrolled multiple times, most enrolled twice into the program, but 3% of enrollees enrolled three or more times into the program. HK WPC Enrollment Processes, Size, and Patterns |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program HISSS HIRSH! Given the staggered enrollment of enrollees into WPC and the different approaches to graduation by Pilot, the length of enrollment by enrollee ranged from 1 to 60 months (data not shown). Exhibit 47 displays the percent of enrollees by their length of enrollment in WPC. Over one-third of enrollees were enrolled for 6 months of less (38%), with 11% of enrollees only enrolled for one month (data not shown). Nearly one-fifth (19%) were enrolled for 7-12 months. The mean, median, and mode length of enrollment in the program was 14.2, 9, and 1 month(s), respectively (data not shown). Length of enrollment varied by Pilot, with mean length of enrollments from 5.8 months in Shasta to 29.7 months in Marin (data not shown). Exhibit 47: Length of Enrollment of WPC Enrollees, PY 2 to PY 6 38% 19% 13% 9% 9 6% 4% 4% 4 2% 2% 2% a |_| pit) = = 1-6 7-12 13-18 19-24 25-30 31-36 37-42 43-48 49-54 55-60 months months months months months months months months months months Source: Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Note: Includes 249,378 unique enrollments into a WPC Pilot. Whole Person Care Final Evaluation Report | WPC Enrollment Processes, Size, and Patterns uml December 2022 Disenrollment UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 48 shows the number of disenrollments each quarter from PY 2 to PY 6. This number ranged from 583 in first quarter of PY 2 (2017) to 14,699 in the third quarter of PY 6 (2021). Exhibit 48: Quarterly Disenrollments from WPC, PY 2 to PY 6 m@ 583 ai 3,605 1,595 Q2 a3 Py2 MEE. «6,688 a4 8,409 10,977 11,071 12,075 14,288 Qi Q2 Q3 Q4 Qi PY3 12,704 13,032 Q2 a3 PY4 13,435 13,996 a4 ai 11,074 9,437 Q2 a3 PYS 4,699 11,296 10,720 11,238 9,452 Q4 Ql Q2 Q3 a4 PY6 Source: Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Note: Includes 200,734 unique disenrollments from WPC, with some enrollees disenrolling more than once. shy) WPC Enrollment Processes, Size, and Patterns |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Tr) Health Economics and Evaluation Research Program WPC Pilots reported reason for disenrollment in the Quarterly Enrollment and Utilization Repots using a standardized set of disenrollment reasons. An additional reason for disenrollment, “Graduated” was not added until PY 3. Of the 200,734 disenrollments from WPC (some enrollees had more than one disenrollment), the most common reasons for disenrollment were “Lack of Engagement” (26%), “WPC Services No Longer Needed” (23%), “Other” (21%), and “Not Eligible for Medi-Cal” (16%; Exhibit 49). Less frequent reasons included “Graduated” (6%) and Beneficiary Request” (5%). Prior to the inclusion of “Graduated,” many WPC Pilots reported that they used the “WPC Services No Longer Needed” reason when their enrollees had met their goals and were ready to leave the Pilot. As a result, the “WPC Services No Longer Needed” is a mix of enrollees that were not appropriate or did not benefit from services provided through WPC and those that successfully developed the skills to independently manage their own care. Exhibit 49: Reason for Disenrollment from WPC, PY 2 to PY 6 ae Rly Graduated, 6% Medi-Cal, 16% a Beneficary eh aes) _— Request, 5% Moved from Service Area, 2% Deceased, 1% WPC Services No Cao Engagement, 26% Longer Needed, 23% Source: Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Note: Includes 200,734 unique disenrollments from WPC with standardized disenrollment reasons. Whole Person Care Final Evaluation Report | WPC Enrollment Processes, Size, and Patterns Juke} erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Enrollment Size and Patterns by Target Population Classification of enrollees into target populations varied by WPC Pilot. Some WPC Pilots classified enrollees into the target population(s) that was used to initially identify the individual as eligible, while others used patient assessment data to classify enrollees into additional target populations that were not the primary reason for their enrollment. Overall, inclusion in a particular target population indicated that an enrollee fit the criteria for that target population. However, exclusion from a target population did not guarantee that an enrollee did not meet the criteria. For example, Napa’s primary target population was the homeless, and all enrollees in the Pilot were categorized only as homeless, and very few were categorized in other target populations. In contrast, Santa Cruz used health records and assessments to categorize their enrollees in up to seven target populations, even though the primary target populations were only those with chronic physical conditions and/or SMI/SUD. The COVID-19 target population was added in PY 5 and could have included both enrollees with known COVID-19 infection and/or those at-risk of infection. While some Pilots only used the target population to provide services to those with specific COVID-19 needs, other Pilots used the broadest definition of at- risk of infection and classified all enrollees in the COVID-19 target populations. UCLA identified which Pilots reported at least ten enrollees in each target population in Exhibit 50. Exhibit 50: WPC Pilots Reporting at Least Ten Enrollees by Target Population, PY 2 to PY 6 2 : ¢ a » “wow . a a ges 3 cea re os Sr g sz a a saa 2S a BS 8 2) Beg ss e re eis = = eee BS > 5 Gis = S ; 8 3 3 5 Oas6 S 2 gz SE 8 WPC Pilot 2 Alameda x x x x Contra Costa x Kern x x x x x Kings x x x Los Angeles x x x x Marin x x x Mendocino x x x Monterey x x x x x Napa x x x Orange x x x x x Placer x x x x x Riverside x x x x x x x Sacramento x x x x x San Bernardino x x San Diego x x x x x x San Francisco x x San Joaquin x x x x x x KEE WPC Enrollment Processes, Size, and Patterns | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program _# o ooo 2 236 5 3 o8 . a3 3 og = o oO sue 22 a gfe = E gee a8 5 GES = S ; oS 32 ° oo a = ax TS ° WPC Pilot = San Mateo x x Santa Clara x x x x x x Santa Cruz x x x x x Shasta x x x x SCWPCC x x x x x x Solano x x x x x x Sonoma x x x x Ventura x x x x Total 24 18 19 23 20 15 9 Source: Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 237,603 unique enrollees in WPC Pilots with a target population reported. When count for a target population was less than ten individuals, it was not reported. SMI/SUD is serious mental illness and/or substance use disorder. SCWPCC is the Small County Whole Person Care Collaborative. The most commonly reported target populations were high utilizers (24 Pilots of 25) and homeless (23). The next most commonly reported target populations were at-risk-of- homelessness (20), SMI/SUD (19), and chronic physical conditions (18). The least often reported target populations were justice-involved (15) and COVID-19 (9). Whole Person Care Final Evaluation Report | WPC Enrollment Processes, Size, and Patterns Jule) December 2022 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Of the 237,603 individuals who ever enrolled in WPC, Pilots classified 57% as high utilizers and 53% as homeless (Exhibit 51). The next most common target populations that enrollees were classified as were justice-involved (25%), SMI/SUD (24%) and at-risk-of-homelessness (22%). Enrollees were least often classified in the COVID-19 (16%) and chronic physical conditions (10%) target populations. Exhibit 51: WPC Enrollee Target Population Classifications, PY 2 to PY 6 High Utilzers i 57% Homeless i 53% Justice-Involved 25% SM/SU0 24% At-Risk-of-Homelessness Ss 22°, CoviD-19 icx Chronic Physical Conditions J 10% Source: Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 237,603 unique enrollees in WPC Pilots with at least one reported target population. Enrollees may be reported in more than one target population. SMI/SUD is serious mental illness and/or substance use disorder. MK WPC Enrollment Processes, Size, and Patterns |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Length of enrollment by target population was influenced by when Pilots started enrollment, the graduation protocols, and the level of need of the enrollee. Ultimately, UCLA found that the enrollees classified in the COVID-19, chronic physical conditions, and SMI/SUD target populations had the longest average length of enrollment (Exhibit 52), ranging from 17.2 to 20.0 months. Enrollees classified in the at-risk-of-homelessness and homeless target populations had the shortest average length of enrollments, ranging from 13.8 to 14.9 months. Exhibit 52: WPC Length of Enrollment in Months by Target Population, PY 2 to PY 6 Target Population Mean 25% Percentile Median 75% Percentile High Utilizers 16.4 4 11 25 Homeless 14.9 3 10 22 Justice-Involved 16.0 3 10 26 SMI/SUD 17.2 4 11 27 At-Risk-of-Homelessness 13.8 2 8 24 COVID-19 20.0 11 18 24 Chronic Physical Conditions 17:7 5 12 29 Source: Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 237,603 unique enrollees in WPC Pilots with at least one reported target population. Enrollees may be reported in more than one target population. SMI/SUD is serious mental illness and/or substance use disorder. Whole Person Care Final Evaluation Report | WPC Enrollment Processes, Size, and Patterns Jaws UCLA Center for Health Policy Ri h Health coBRarAIeS aa rvallntlon Ressateh Program Chapter 5: WPC Services Offered and Delivered WPC Pilots were expected to improve beneficiary health and wellbeing by coordinating their use of health, behavioral health, and social services in a patient centered manner. However, WPC did not predefine the specific types of services to be offered and delivered by Pilots. This chapter addresses the following evaluation question: “what services did WPC enrollees receive through WPC?” Data sources for this chapter include WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6, PY 5 (2020) LE survey, WPC applications (n=25), and WPC Annual Invoices from PY 2 to PY 6. The WPC Quarterly Enrollment and Utilization Reports were used to identify enrolled individuals, their identified target populations, and their use of WPC services across the length of the entire program as reported through utilization of per-member, per-month (PMPM) bundled services or individual service reimbursed as fee-for-service (FFS). The specific services offered through each PMPM bundles and FFS category included in the WPC Quarterly Enrollment and Utilization Reports were identified by Pilots in the PY 5S (2020) LE survey. WPC Annual Invoices were used to identify the cost of each PMPM and FFS category per year. Lastly, the WPC applications were used to identify the amount paid to WPC Pilots during PY 1, prior to the start of enrollment and the submission of annual invoices. WPC Services Offered Pilots had the flexibility to offer services that would best fit the needs of their target populations and could be delivered with existing or newly developed infrastructure and resources. While no single service was specifically required by the program, all Pilots were expected to provide care coordination and housing support services as needed to address the needs of beneficiaries. Additionally, Pilots had the flexibility to determine whether funding for these services would be provided through capitated payments for bundled services (per- member, per-month [PMPM]) or single payments for defined services (fee for service [FFS]). Pilots reported WPC service utilization per enrollee using PMPM and FFS categories identified in WPC Quarterly Enrollment and Utilization Reports. Pilots included multiple services under these service categories. Pilots differed in the number of categories, and categories were not comparable across Pilots. Specifically, category descriptions frequently did not identify types of services that were included therein. Therefore, UCLA asked Pilots to report on inclusion of 20 different services in each FFS and PMPM bundle in the PY 5 (2020) LE survey. UCLA then grouped the 20 possible services into 11 service categories for analysis. Exhibit 53 shows how the 20 specific services were grouped. UCLA used the individual- KEE) =WPC Services Offered and Delivered | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program level utilization data in the WPC Quarterly Enrollment and Utilization Reports from PY 2 (2017) to PY 6 (2021) to assess enrollee-level service use for each of the 11 service groups. Exhibit 53: WPC Services Offered by Pilots as of PY 5 WPC Services Groups Description of Specific Services Offered per Category Outreach Outreach to prospective enrollees in the field including at homes, homeless encampments, shelters, Emergency Departments, etc. Outreach to prospective enrollees through telephone, in-office visits, email or mail. Care Coordination Conduct needs assessments as part of care coordination services. Develop care plans as part of care coordination services. Link or refer patient to needed services and then follow up on referrals as needed as part of care coordination services. Provide frequent communication with enrollees and follow up on referrals as part of care coordination services. Provide warm hand-offs to other providers. Housing Support Provide housing navigation services, which includes applying for, connecting to, and accessing housing services. Provide supportive housing services, which includes successful linkage to services that increase housing stability through tenancy services, housing transition services, legal support, and coaching for successful housing skills. Benefit Assistance Assess enrollees for eligibility for public benefits services (e.g., SSI, CalFresh, etc.). Actively assist with benefit applications and appeals. Employment Assistance Provide one-on-one coaching, training or education programs to assist enrollees in finding and securing employment. Actively refer and place enrollees in job opportunities. Sobering Center Provide sobering center services. Medical Respite Medical respite or recuperation services for 48 hours or less. Medical respite or recuperation services for greater than 48 hours. Whole Person Care Final Evaluation Report | WPC Services Offered and Delivered SK) errr UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Transportation Coordinate or provide transportation to enrollees for appointments or services. Health Education Actively refer to or provide educational opportunities (e.g., classes) designed to teach enrollees about improving their health and well- being. Legal Services Actively refer to or provide legal services or legal assistance (e.g., related to their criminal charges or other legal needs). Re-entry Services Run educational programs (e.g., one-on-one or in groups) specifically designed to assist in adjusting to life post-incarceration. Source: PY 5 Lead Entity (LE) Surveys, n=25, June-August 2020. Note: UCLA developed the WPC service list using knowledge of WPC Pilot design and set of interventions. Exhibit 54 shows the frequency with which Pilots offered WPC services. All Pilots offered outreach, care coordination, housing support, benefit assistance and transportation. The majority of Pilots also offered health education (92%) and legal services (84%). However, sobering centers and re-entry services were the least often offered (56% and 28% of Pilots, respectively). Exhibit 54: Percentage of WPC Pilots Offering Each Service Group Cts 100% Care Coordin ati) ns 100% Housing Support 100% Benefit Assistance Ls 100% Employment Assistance ns 75% Sobering Centers 56% Medical Respite 72°, Transportation ns 100% Health Education a 92%, Legal Services EA 84% Re-Entry Serviccs 28% Source: PY 5 Lead Entity (LE) Surveys, n=25, June-August 2020 and Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Two counties in the Small County Whole Person Care Collaborative (SCWPCC) (Mariposa and San Benito) were counted separately as they reported unique combinations of services. Napa and Plumas counties were excluded from this service analysis because they did not respond to the LE Survey, and they dropped out of WPC in PY 3, respectively. ute WPC Services Offered and Delivered | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ely Health Economics and Evaluation Research Program The number of PMPM and FFS service categories reported in WPC Quarterly Enrollment and Utilization Reports, are shown in Exhibit 55 and vary with Pilot. Pilots offered as many as 16 and as few as 1 PMPM bundles. They also offered as many as 21 and as few as 1 individual services (FFS). Some Pilots disaggregated services into numerous bundles and individual services (e.g., Alameda) and others relied on very few (e.g., San Mateo, Solano). Pilots differed in type of services bundled together. For example, San Mateo provided all of their services through two PMPM bundles that included a range of services (e.g., care coordination, benefit assistance, sobering center, transportation, and health education). Conversely, Los Angeles provided sobering centers to WPC enrollees, but only as a stand-alone service funded through an FFS mechanism, and other WPC services were bundled in program-specific PMPM bundles. Exhibit 55: Number of Bundles (PMPM) and Individual (FFS) Services Offered by WPC Pilots, PY 2 to PY6 eco 5 Contra Costa Ke ne (cs Angeles LS [ Mai Mariposa >a Mendocino [>i Monterey Cane Placer Riverside Sacramento San Benito >a San Bernardino 2m Sandiego San Francisco ey San Joaquin a a San Mateo [Pail Santa C2. Santa Cru Shasta Solano 5Ue mPMPM Categories Offered mFFS Categories Offered Source: Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Whole Person Care Final Evaluation Report | WPC Services Offered and Delivered erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Notes: Two counties in the Small County Whole Person Care Collaborative (SCWPCC) (Mariposa and San Benito) were counted separately as they reported unique combinations of services. Napa and Plumas counties were excluded from this service analysis. WPC Services Delivered UCLA reported the proportion of enrollees that utilized a service category at any point during the program overall and among seven target populations. The COVID-19 target population was added in the second half of 2020. Pilots did not uniformly define or apply assignment criteria to this new target population. Some Pilots retroactively assigned enrollees and others used the broadest definition of at-risk for COVID-19 and reassigned all enrollees to this target population. Due to these inconsistencies, UCLA included any enrollee that was ever assigned to the COVID-19 target population in the following analyses. Therefore, the findings reflect the overall experience of these enrollees and are not restricted to the second half of PY 5 and PY 6 (July 2020 to December 2021). In addition, UCLA reported service use for the small proportion of beneficiaries who were not formally enrolled in WPC but received outreach or sobering center services. The data used for the analyses in this section reflect the bundle of services delivered to specific enrollees, but does not guarantee receipt of each service under a bundle. For example, an enrollee who received a bundle that included both care coordination and benefit assistance may not have received benefit assistance if they were not eligible or it was not needed. Furthermore, UCLA analyzed the services provided by the two counties in the Small County Whole Person Care Collaborative (SCWPCC) Pilot (San Benito and Mariposa) separately as each used different bundles of services. Two Pilots were excluded from these analyses due to non- response to the PY 5 LE survey and subsequent lack of information regarding services (Napa) and discontinuation of WPC involvement in PY 3 (Plumas). Outreach Nearly three-quarters of the enrollees (73%) received outreach services (Exhibit 56). Among the WPC target populations, the SMI/SUD target population was most often offered outreach services (91%) and the COVID-19 population was the least often offered outreach services (42%). Of the 25 Pilots offering the service, outreach was funded through PMPM by 17. KP WPC Services Offered and Delivered | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research eer Health Economics and Evaluation Research Program Exhibit 56: Outreach Services Delivered to WPC Enrollees by Enrollment Status and Target Population, PY 2 to PY6 89% 87% 79% . | | | Enrolled High Utilizers Chronic SMI/SUD Homeless Risk of Justice COVID-19 Physicial Homelessness Involved Conditions WPc Enrolled Target Populations Population Source: PY 5 Lead Entity (LE) Surveys, n=25, June-August 2020 and Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 248,599 unique individuals that received services through WPC and were enrolled. Enrollees are included in target population if ever assigned to that target population during program. COVID-19 target population was added in PY 5. SMI/SUD is serious mental illness and/or substance use disorder. PY 2 is 2017 and PY 6 is 2021. Pilots varied in their outreach approach. For example, Sacramento used outreach navigators to identify potential enrollees and refer them for WPC eligibility determination and enrollment, while Monterey provided targeted outreach services in conjunction with other services to help establish trust and rapport with enrollees. More detailed information regarding overall activities of Pilots in the identification, enrollment, and engagement efforts are provided in Chapter 4: WPC Enrollment Processes, Size, and Patterns. Care Coordination The great majority (89%) of WPC enrollees received care coordination services (Exhibit 57). This estimate included those newly enrolled who were being assessed prior to receipt of care coordination services as well as a subset of enrollees who were linked to other providers without using care coordinator services. Among the enrolled WPC target populations, estimated care coordination rates were high among all populations. The COVID-19 population had the lowest rate of estimated care coordination at 79%. All 25 Pilots offering care coordination funded the service through at least one PMPM. More detailed information regarding overall activities of Pilots in care coordination efforts is provided in Chapter 6: WPC Care Coordination. Whole Person Care Final Evaluation Report | WPC Services Offered and Delivered errr UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 57: Care Coordination Services Delivered to WPC Enrollees, Overall and by Target Population, PY 2 to PY6 91% 90% 89% 92% 87% oar om 1% 3 2 3 2 8 a 7 a 3 5 g S 3 § g in 2 2 ge @ 2 5 3 Q % ge E 8 = -5 6 = E° 5 3 & 3 2 z Wec Enrolled Target Populations Population Source: PY 5 Lead Entity (LE) Surveys, n=25, June-August 2020 and Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Notes: Includes 248,599 unique individuals that received services through WPC and were enrolled. Enrollees are included in target population if ever assigned to that target population during program. COVID-19 target population was added in PY 5. SMI/SUD is serious mental illness and/or substance use disorder. PY 2 is 2017 and PY 6 is 2021. Housing Support The majority (70%) of WPC enrollees received housing support services (Exhibit 58). Receipt of housing support services varied somewhat by target population, with 91% of justice-involved enrollees receiving services that included housing support but only 38% of COVID-19 enrollees receiving services that included housing support. Of the 25 Pilots offering the service, housing support was funded through PMPM by 24. PLE WPC Services Offered and Delivered | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Tr) Health Economics and Evaluation Research Program Exhibit 58: Estimated Delivery of Housing Support Service to WPC Enrollees, Overall and by Target Population, PY 2 to PY 6 77% ‘ z 71% 72% 70% a 75% «ase wec Enrolled Target Populations Population Enrolled Homeless COVID-19 High Utilizers Chronic Physicial Conditions Justice Involved Risk of Homelessness Source: PY 5 Lead Entity (LE) Surveys, n=25, June-August 2020 and Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Notes: Includes 248,599 unique individuals that received services through WPC and were enrolled. Enrollees are included in target population if ever assigned to that target population during program. COVID-19 target population was added in PY 5. SMI/SUD is serious mental illness and/or substance use disorder. PY 2 is 2017 and PY 6 is 2021. Based on interviews with Pilot lead entities and frontline staff, WPC Pilots often used specialized staff (e.g., social workers) to provide housing support services, which often focused on helping enrollees live in the least restrictive community-based setting appropriate to their needs. Staff providing housing support services typically focused on identifying and mitigating barriers to housing placements and facilitating enrollee access to short-term shelters, coordinated entry systems, or to other housing benefits. Many Pilots had staff that also worked directly with landlords to mediate disputes, encourage renting to enrollees with negative rental histories, and/or assist landlords in accessing programs that reward them for renting their properties to underserved populations. Some Pilots also set aside funds to directly support enrollees with a range of housing-related financial needs that if not addressed, would negatively impact their ability to accept or maintain housing placement. For example, funds could be used to help pay security deposits, set-up fees for utilities or service access, first month utilities, outstanding utility bills, furniture, moving costs, cleaning services prior to move- in, home modifications needed to have their medical needs met in the home, medically necessary services (e.g., hospital beds or lifts), credit repair, criminal record expungement, etc. Further detail on housing services can be found in the chapter on enrollees experiencing homelessness. Selected examples of housing support services are provided in Exhibit 59. Whole Person Care Final Evaluation Report | WPC Services Offered and Delivered 1B-Tor-T 4) 2022 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 59: Selected Examples of Housing Support in WPC WPC Pilot Example of Housing Support Alameda Alameda’s housing transition service bundle included elements essential for enrollees’ transition to attaining housing. Funds were used for security deposits, set-up fees for utilities or service access, first month utilities, furniture, moving costs, cleaning services prior to move-in, home modifications (e.g., A/C and/or heater), medically necessary services (e.g., hospital beds or lifts). Kern Kern initially sent housing referrals to the Kern Housing Authority (KHA), and by PY 4, the increasing volume of referrals resulted in an updated process wherein WPC staff conducted warm hand-offs with KHA. This allowed WPC staff to be involved with KHA in the process of scheduling, documentation assistance, and coordination of services for the enrollee. Marin Marin had a housing-based case management component where enrollees who were homeless or precariously housed were supported by a case manager who worked to secure and sustain housing while also promoting awareness and teaching strategies that reduced the likelihood of a return to homelessness in the future. Napa Napa provided training on housing rights (e.g., occupancy and eviction issues) for people with disabilities, families with children, and other classes protected in the Fair Housing Act. Placer Placer provided a housing services bundle for homeless or individuals at-risk-of homelessness that worked towards obtaining housing and developing daily living skills to remain stable in their new living situation. Services included housing assessments, developing an individualized housing support plan, assistance with the housing application, and identifying and securing available resources to assist with subsidizing rent. Riverside Riverside’s housing bundle included financial assistance to provide money to landlords for up to a triple security deposit. Landlords were usually skeptical of providing housing to new probationers. Through the deposit, however, landlords were incentivized to provide housing to this population. San Benito (SCWPCC) San Benito provided financial assistance for credit repairs and/or criminal record expungement in order to better position enrollees for housing. Santa Cruz Santa Cruz enrollees met with WPC staff up to twice daily or weekly to address poor tenancy skills, which affected their ability to maintain stable, housing situations. Source: Whole Person Care Pilot Applications (n=25), 2016 and WPC Mid-Year and Annual Narrative Reports, PY 2 (2017) - PY 6 (2021) and Follow-up Interviews with Lead Entity (LE) and Frontline Staff from PY 2 to PY 6. Note: SCWPCC is the Small County Whole Person Care Collaborative Benefit Assistance Among WPC enrollees, 79% received benefit assistance (Exhibit 60). Among the various target populations, risk of homelessness, chronic physical conditions, and SMI/SUD were most likely to receive benefits assistance (97%, 96%, and 95%, respectively). The COVID-19 target population was the least likely to receive benefit assistance (36%). Of the 25 Pilots offering the service, benefit assistance was funded through PMPM by 24. nA WPC Services Offered and Delivered | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ely Health Economics and Evaluation Research Program Exhibit 60: Benefit Assistance Services Delivered to WPC Enrollees, Overall and by Target Population, PY 2 to PY6 89% 91% 94% 90% 79% wec Enrolled Target Populations Population 80% 81% 65% Enrolled Justice Involved High Utilizers Chronic Physicial Conditions SMI/SUD Homeless Risk of Homelessness Source: PY 5 Lead Entity (LE) Surveys, n=25, June-August 2020 and Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 248,599 unique individuals that received services through WPC and were enrolled. Enrollees are included in target population if ever assigned to that target population during program. COVID-19 target population was added in PY 5. SMI/SUD is serious mental illness and/or substance use disorder. PY 2 is 2017 and PY 6 is 2021. Benefit assistance included a range of services such as assistance with applications for Supplemental Security Income/Social Security Disability Insurance (SSI/SSDI), Medi-Cal, CalFresh, and/or CalWorks (e.g., completing applications, obtaining critical eligibility documents such as certified mail and identification cards, preparing medical summary reports), benefits advocacy (e.g., appealing initially rejected applications), transportation to appointments, and other miscellaneous services. For example, Contra Costa provided enrollees with temporary phones, while Kern offered childcare services so enrollees could attend needed appointment and services. Other selected examples of benefit assistance services are found in Exhibit 61. Exhibit 61: Selected Examples of Benefit Assistance Services in WPC WPC Pilot Example of Benefit Assistance Services Alameda Alameda held trainings informing participants how to identify and secure public benefits. Kings Kings developed a screening tool to send referrals for participants applying for public benefits. Kings was also able to monitor the status of applications to better manage the application process. Solano Solano assisted enrollees in obtaining Supplemental Security Income/Social Security Disability Insurance (SSI/SSDI) Advocacy. This included assistance with obtaining critical eligibility documents (e.g., birth certificates, identification cards, certified mail), preparing detailed Medical Summary Reports, gathering and paying for potential costs for health records, and appealing initially rejected applications. Whole Person Care Final Evaluation Report | WPC Services Offered and Delivered JRA errr UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Source: Whole Person Care Pilot Applications (n=25), 2016 and WPC Mid-Year and Annual Narrative Reports, PY 2 (2017) - PY 6 (2021) and Follow-up Interviews with Lead Entity (LE) and Frontline Staff from PY 2 to PY 6. Employment Assistance Over one-third (39%) of WPC enrollees received employment assistance (Exhibit 62). Receipt of employment assistance was highest among high utilizers (53%), and lowest in the COVID-19 target population (8%). Of the 19 Pilots offering the service, employment assistance was funded through PMPM by 18. Exhibit 62: Employment Assistance Services Delivered to WPC Enrollees, Overall and by Target Population, PY 2 to PY6 59% 50% 42% 36% 36% 27% 29% ] | [| LI o Enrolled High Utilizers Chronic = SMI/SUD _—_ Homeless Risk of Justice COVID-19 Physicial Homelessness Involved Conditions wec Enrolled Target Populations Population Source: PY 5 Lead Entity (LE) Surveys, n=25, June-August 2020 and Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 248,599 unique individuals that received services through WPC and were enrolled. Enrollees are included in target population if ever assigned to that target population during program. COVID-19 target population was added in PY 5. SMI/SUD is serious mental illness and/or substance use disorder. PY 2 is 2017 and PY 6 is 2021. Employment assistance focused on helping enrollees develop skills and connections that would improve their chances of obtaining employment. For example, Kern provided enrollees with training on personal finance, resume building, interview skills, application assistance, and other supportive services. Napa connected clients with the local Workforce Development Board’s “America’s Job Center,” which offered free internet access, a resource library, resume building assistance, and employment readiness workshops. Solano hired an Employment Specialist who offered enrollees one-on-one coaching on how to secure a job and maintain employment. nEIN WPC Services Offered and Delivered | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ely Health Economics and Evaluation Research Program Sobering centers were used as a safe space to recover from the acute effects of alcohol and Sobering Centers drug intoxication and as an alternative to placement in ED, emergency psychiatric services, hospitals, or incarceration. Among overall WPC enrollees, 14% received sobering center services. Those in the risk of homelessness, chronic physical conditions, and justice-involved target populations had the highest rates of estimated sobering center use at 31%, 29%, and 29%, respectively. One-quarter (25%) of the SMI/SUD target population received the service (Exhibit 63). Of the 14 Pilots offering the service, sobering centers were funded through PMPM by 7. Exhibit 63: Sobering Centers Services Delivered to WPC Enrollees by Enrollment Status and Target Population, PY 2 to PY6 49% 45% 43% 34% 21% 14% 14% 10% a Of al 5 Enrolled High Utilizers. Chronic» SMI/SUD _—_—_ Homeless Risk of Justice coviD-19 Physicial Homelessness Involved Conditions wec Enrolled Target Populations Population Source: PY 5 Lead Entity (LE) Surveys, n=25, June-August 2020 and Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 248,599 unique individuals that received services through WPC and were enrolled. Enrollees are included in target population if ever assigned to that target population during program. COVID-19 target population was added in PY 5. SMI/SUD is serious mental illness and/or substance use disorder. PY 2 is 2017 and PY 6 is 2021. Pilots had different criteria for the individuals that used their sobering centers and the services offered within the center. Some Pilots offered specific services to patients with SUD and a co- occurring mental illness, while other Pilots offered more comprehensive, multidisciplinary services. Most Pilots with sobering centers only permitted enrollees to stay for 24 hours or less, with the exception of Kings, which required enrollees to stay for a longer period of time (e.g., average of three days) to complete detox. Exhibit 64 highlights selected examples of sobering center services in WPC Pilots. Whole Person Care Final Evaluation Report | WPC Services Offered and Delivered erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 64: Selected Examples of Sobering Center Services in WPC WPC Pilot Example of Sobering Center Services Contra Costa Contra Costa included a 24/7 sobering center in order to provide a safe environment for uncomplicated, acute intoxicated individuals to receive detoxification services along with comprehensive care services such as basic hygiene, identification and management of urgent care needs, transportation, etc. Los Angeles Los Angeles provided onsite services such as medical triage, point-of-care lab testing, client beds, oral rehydration and food service, nausea treatment, wound care and dressing changes, shower and laundry facilities, substance use counseling, and linkage to health and behavioral health services. Santa Clara Mission Street Sobering Center in Santa Clara used their own transportation and worked with local law enforcement to transport participants to the sobering center. Sobering center staff were trained on administering screenings to identify homelessness and housing eligibility and screening results were documented in the participant’s record. Source: Whole Person Care Pilot Applications (n=25), 2016 and WPC Mid-Year and Annual Narrative Reports, PY 2 (2017) - PY 6 (2021) and Follow-up Interviews with Lead Entity (LE) and Frontline Staff from PY 2 to PY 6. Medical Respite Medical respite was viewed as a critical tool for helping reduce over-utilization of ED visits and hospitalizations. Medical respite included acute and post-acute medical care for enrollees in unstable living situations who were not sufficiently ill to remain in a hospital or skilled nursing facility but too ill to recover without adequate shelter. Among WPC enrollees, 6% received services that included medical respite or recuperation care (Exhibit 65). Among the target populations, enrollees with chronic physical conditions had the highest rate of receiving these services (22%). Of the 18 Pilots offering the service, medical respite was funded through PMPM by 8. Exhibit 65: Medical Respite Services Delivered to WPC Enrollees by Enrollment Status and Target Population, PY 2 to PY6 28% 15% .% 15% 6% 5% | 7% 6% 566 Enrolled High Utilizers Chronic SMI/SUD Homeless Risk of Justice covID-19 Physicial Homelessness Involved Conditions wec Enrolled Target Populations Population nicfey WPC Services Offered and Delivered | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ely Health Economics and Evaluation Research Program Source: PY 5 Lead Entity (LE) Surveys, n=25, June-August 2020 and Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 248,599 unique individuals that received services through WPC and were enrolled. Enrollees are included in target population if ever assigned to that target population during program. COVID-19 target population was added in PY 5. SMI/SUD is serious mental illness and/or substance use disorder. PY 2 is 2017 and PY 6 is 2021. Length of stay in medical respite varied considerably across Pilots. Kings provided medical respite for an average of one to three days, but expected enrollees to utilize the service more than once while enrolled in WPC, while Ventura estimated an average enrollee length of stay at 12 days. By contrast, multiple other Pilots (Orange, Los Angeles, Placer, San Francisco, and San Joaquin) permitted stays of up to three months. Transportation Transportation services were often offered in conjunction with other services. Among WPC enrollees, 63% received transportation as part of a bundle of services or alone (Exhibit 66). Among the target populations, SMI/SUD enrollees and high utilizers had the highest rates of services that included transportation (81% and 76%, respectively). Of the 25 Pilots offering the service, transportation was funded through PMPM by 23. Exhibit 66: Transportation Services Delivered to WPC Enrollees by Enrollment Status and Target Population, PY 2 to PY 6 . 69% 15% 64% 61% se 55% % six | [ L L Enrolled | High Utilizers Chronic SMI/SUD Homeless Risk of Justice coviID-19 Physicial Homelessness Involved Conditions wec Enrolled Target Populations Population Source: PY 5 Lead Entity (LE) Surveys, n=25, June-August 2020 and Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 248,599 unique individuals that received services through WPC and were enrolled. Enrollees are included in target population if ever assigned to that target population during program. COVID-19 target Whole Person Care Final Evaluation Report | WPC Services Offered and Delivered [ESI errr UCLA Center for Health Policy Research Health Economics and Evaluation Research Program population was added in PY 5. SMI/SUD is serious mental illness and/or substance use disorder. PY 2 is 2017 and PY 6 is 2021. Many Pilots used existing infrastructure and processes to improve transportation availability for enrollees, while other Pilots developed new technology to coordinate transportation. For example, Kings worked with Anthem Blue Cross to understand which free transportation options were available for enrollees and created a medical transportation guide to give providers and enrollees more information about transportation options. Solano worked with Partnership Health Plan of California to leverage their transportation resources and improve access to healthcare appointments. Contra Costa implemented a new ridesharing platform that linked to an enrollee’s electronic health record and gave providers the ability to coordinate a ride for the enrollee. Health Education Pilots provided health education services to give enrollees tools to improve their health status and understand how to navigate the healthcare system. Among WPC enrollees, 39% received health education on its own or under a bundle of services (Exhibit 67). The high utilizer target population had the highest rates of health education service (56%), followed by enrollees with chronic physical conditions and SMI/SUD (50%). Of the 23 Pilots offering the service, health education was funded through PMPM by 22. Exhibit 67: Health Education Services Delivered to WPC Enrollees by Enrollment Status and Target Population, PY 2 to PY6 68% 61% 61% 47% 40% 30% 32% | [| l u Enrolled High Utilizers Chronic_~=—- SMI/SUD _—_ Homeless Risk of Justice cOVID-19 Physicial Homelessness Involved Conditions wec Enrolled Target Populations Population Source: PY 5 Lead Entity (LE) Surveys, n=25, June-August 2020 and Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 248,599 unique individuals that received services through WPC and were enrolled. Enrollees are included in target population if ever assigned to that target population during program. COVID-19 target KY WPC Services Offered and Delivered | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program population was added in PY 5. SMI/SUD is serious mental illness and/or substance use disorder. PY 2 is 2017 and PY 6 is 2021. Health education services often focused on improving patients’ ability to navigate the healthcare system, teaching skills to address specific conditions, and educating patients about preventative care resources as alternatives to frequent hospital and emergency department utilization. Exhibit 68 shows selected examples of health education services. Exhibit 68: Selected Examples of Health Education Services in WPC WPC Pilot Example of Health Education Services Kern Kern developed six care coordination classes to improve enrollees’ relationships with their care coordinator as well as to increase self-sufficiency in addressing all aspects of their health. The classes included Health Literacy, Hospital Relapse Prevention, Job and Volunteer Readiness, Basic Nutrition, Household Budgeting, and Life Skills. Kings Kings developed a Medical Education Brochure to inform patients of the importance of regular preventative care visits and of alternative options to emergency department utilization. Santa Clara Santa Clara implemented screenings and nutrition classes to support their pre- diabetic population. Source: Whole Person Care Pilot Applications (n=25), 2016 and WPC Mid-Year and Annual Narrative Reports, PY 2 (2017) - PY 6 (2021) and Follow-up Interviews with Lead Entity (LE) and Frontline Staff from PY 2 to PY 6. Whole Person Care Final Evaluation Report | WPC Services Offered and Delivered [EE] errr UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Legal Services Legal services included providing or referring enrollees to assistance related to any legal needs surrounding topics such as public benefits, housing, immigration, and criminal charges. Among WPC enrollees, 68% received legal services alone or as part of a bundle (Exhibit 69). The SMI/SUD and high utilizer target populations had the highest rates of services including legal service (79% and 74%, respectively). Of the 21 Pilots offering the service, legal services were funded through PMPM by 19. Exhibit 69: Estimated Delivery of Legal Service to WPC Enrollees by Enrollment Status and Target Population, PY 2 to PY6 67% 65% oo" oe% 64% 56% L Enrolled High Utilizers Chronic = SMI/SUD —_ Homeless Risk of Justice COVID-19 Physicial Homelessness Involved Conditions wec Enrolled Target Populations Population Source: PY 5 Lead Entity (LE) Surveys, n=25, June-August 2020 and Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 248,599 unique individuals that received services through WPC and were enrolled. Enrollees are included in target population if ever assigned to that target population during program. COVID-19 target population was added in PY 5. SMI/SUD is serious mental illness and/or substance use disorder. PY 2 is 2017 and PY 6 is 2021. Many Pilots developed partnerships with legal aid organizations to connect WPC enrollees with legal assistance. Contra Costa worked with Bay Area Legal Aid to develop and administer a survey for WPC enrollees to identify those who needed legal assistance, conduct classes to educate case managers on legal issues, and provide WPC enrollees free legal services. Class topics included Housing Law, Immigration and Survivors of Interpersonal Violence, SSI and Other Public Benefits, Health Consumer Law, Small Claims Court Processes, Reentry, Wills & Trusts, and Consumer Debt. Los Angeles also had a Medical Legal Partnership program to connect enrollees with legal aid often related to claims denials. KEV WPC Services Offered and Delivered | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research . 7 December 2022 Health Economics and Evaluation Research Program Among all WPC enrollees, 10% received re-entry services (Exhibit 70). As expected, the justice involved target population had the highest rates of these services (34%) while all other target populations received very few re-entry services. Of the 7 Pilots offering the service, re-entry services were funded through PMPM by 4. Re-Entry Services Exhibit 70: Estimated Delivery of Re-entry Services to WPC Enrollees by Enrollment Status and Target Population, PY 2 to PY 6 45% 9 13% i 0% 2% o% us 0% rs] — — |_| = Enrolled High Utilizers. Chronic SMI/SUD Homeless Risk of Justice COVvID-19 Physicial Homelessness Involved Conditions wec Enrolled Target Populations Population Source: PY 5 Lead Entity (LE) Surveys, n=25, June-August 2020 and Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 248,599 unique individuals that received services through WPC and were enrolled. Enrollees are included in target population if ever assigned to that target population during program. COVID-19 target population was added in PY 5. SMI/SUD is serious mental illness and/or substance use disorder. PY 2 is 2017 and PY 6 is 2021. Re-entry services varied by Pilot, but both Kern and Kings offered life skills classes with Kings providing enrollees with a life skills manager to coordinate training and participation in educational classes. Services without Enrollment Of the individuals identified in WPC Quarterly Enrollment and Utilization Reports to have received services, 67,580 individuals were never formally enrolled into WPC by the end of the program. These individuals were identified by Pilots during outreach but were not enrolled either due to lack of engagement or did not meet the eligibility criteria. Pilots provided outreach (initial contact with potential enrollee) and/or short-term stays in sobering centers. Of the 25 WPC Pilots, 20 reported these individuals. Of the 17 Pilots that had more than 10 such individuals, the numbers varied from 22,629 in Los Angeles to 113 in San Joaquin (Exhibit 71). All (100%) individuals receiving services without enrollment in Los Angeles received outreach services, but 15% received a stay in a sobering center (data not shown). Kern initially used Whole Person Care Final Evaluation Report | WPC Services Offered and Delivered errr UCLA Center for Health Policy Research Health Economics and Evaluation Research Program administrative data from the managed care plans to identify individuals as potential enrollees and then screened these individuals to determine their eligibility. They found that this system was not successfully identifying their target populations and switched to a referral-based system. Exhibit 71: Individuals Receiving Services through WPC without Enrollment by Pilot, PY 2 to PY 6 7 a se 2 q a a ou 3 a 2 wy oS £ non mao . RS £ 82g es w we F © Hh FS TF Rg wv oom we 7 2 KR SF ow a ow om il “44a 8 2© = & & GS Eid + HM N 4 Eo i oO S @ @B@ GO @ O© A ~O Ww BS WB OC Ll WN xs é eo os ¥ < s LE PFE EF SF EE EEE EE ES 8 ee ec ° EF SS ¥ SE SSC SS i e* y Ss & & 9 S % Source: Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 67,580 individuals reported as receiving services but never enrolled in the WPC. Excludes two Pilots that reported less than eleven individuals that received services without enrollment. WPC Expenditures and Payment for WPC Services UCLA calculated the amounts paid to Pilots for WPC using WPC Applications and WPC Annual Invoices from PY 2 to PY 6. The amount paid to Pilots in PY 1 to start implementation of the program prior to enrollment was equivalent to the approved budget amount for PY 2 detailed in their WPC applications and only once their WPC application was approved and baseline metric data was submitted. Following the start of enrollment in PY 2, Pilots were paid based on infrastructure requirements (administrative and delivery infrastructure), the amount of WPC services delivered to enrollees (PMPM and FFS), and for meeting predefined goals (pay for reporting, pay for outcomes, and incentive payments). Exhibit 72 shows the total amounts paid to WPC Pilots. This includes overall payments and amount per program year across Pilots, in addition to the median and range of amounts paid to individual Pilots. Overall, nearly $3.6 billion was paid to WPC Pilots, ranging from $6.2 million (Solano) to $1.5 billion (Los Angeles) per Pilot. Annual payments increased from $361 million in nila WPC Services Offered and Delivered | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program PY 2 to $778 million in PY 5. Payments were lower in PY 6 or when WPC was extended for one year and two Pilots discontinued their Pilots. Sonoma did not start enrollment in PY 2 as planned due to delays in implementation that resulted from wildfire activity in their area and as a result did not receive any payment in PY 2. Exhibit 72: Program Year and Overall WPC Payments to Pilots, PY 1 to PY 6 eee aselaames Median Pilot Te mdi ela Maximum Pilot Payment Ey 4 Payment PY1 $498,967,343 $4,907,400 $933,402 $180,000,000 PY2 $361,336,345 $3,057,092 $0 $137,003,935 PY3 $546,238,400 $5,638,780 $802,183 $226,215,249 PY4 $766,371,449 $6,241,763 $825,319 $367,243,307 PYS $778,374,868 $7,585,920 $1,708,800 $346,299,925 PY6 $642,848,405 $6,242,833 $1,419,352 $279,499,004 PY1-PY6 $3,594,136,811 $31,888,477 $6,164,396 $1,536,261,420 Source: WPC Annual Invoices, PY 2 to PY 6. Notes: For PY 2, Sonoma did not receive payment in PY 2 because they had zero enrollment during PY 2. SWPCC and Solano did not participate in WPC during PY 6. Following enrollment in PY 2, WPC Pilots submitted invoices broken down into budget categories to receive payment (Exhibit 73). Data showed that the largest payment category was WPC services (53%), followed by 20% for incentives, and 10% for pay for outcomes categories. There was large variation in the breakdown of payments by budget category among Pilots (data not shown). Exhibit 73: Proportion of Overall WPC Payments to Pilots by Budget Category, PY 2 to PY6 Administrative Infrastructure, 7% _ Os Delivery Infrastructure, 5% UatetTa) NYC Nu eed WPC Services, 53% Pay for Outcomes, a 10% Pay for Reporting, J 5% Source: WPC Annual Invoices, PY 2 to PY 6. Note: SWPCC and Solano did not participate in WPC during PY 6. Whole Person Care Final Evaluation Report | WPC Services Offered and Delivered errr UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Pilots were reimbursed for WPC services based on the reported use of bundles (PMPM) and individual services (FFS). PMPM bundles were paid for each month that an enrollee was included in that bundle and FFS was paid every time an enrollee used that service. Exhibit 74 shows the percent of total WPC service payments made to WPC Pilots that were paid under PMPM or FFS for each Pilot. Twenty Pilots mainly received payments through PMPM, with two Pilots (Placer and San Mateo) only receiving payments through PMPM. Five Pilots received payments mainly through FFS. Pilots used different strategies and designs to create their set of interventions and payment structure for these services. For example, Alameda largely worked with existing programs and organizations to provide WPC services and relied on FFS to pay for these services. Other Pilots, like Contra Costa and San Mateo, developed largely new infrastructure to provide WPC services and bundled these services into a few PMPMs and had none or few individual services paid through FFS. KIEEI) WPC Services Offered and Delivered | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 Exhibit 74: Proportion of Total WPC Services Payments under PMPM and FFS Reimbursement Methods by Pilot, PY 2 to PY 6 All Pilots Alameda 34% 66% Contra Costa Kern Kings Los Angeles Marin 91% 9% Mendocino Monterey cw Ea Napa Orange Placer Riverside Sacramento San Bernardino San Diego San Francisco tA CTA San Joaquin San Mateo Santa Clara Santa Cruz SCWPCC Shasta Solano 70% 30% Sonoma A 84% Ventura PA CA mPMPM as a Percentage of Service Budget. —smFFS.as.a Percentage of Service Budget Source: WPC Annual Invoices, PY 2 to PY 6 and Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: SCWPCC is the Small Counties Whole Person Care Collaborative. PMPM is per-member, per-month payments for a bundle of services and FFS (fee for service) is payment for specific services. rot) Whole Person Care Final Evaluation Report | WPC Services Offered and Delivered JE) errr UCLA Center for Health Policy Research Health Economics and Evaluation Research Program UCLA calculated the average payment to Pilots per enrollee for WPC services from PY 2 to PY 6 overall and by target population (Exhibit 75). On average, WPC Pilots received $6,272 per enrollee and $743 per beneficiaries not formally enrolled. Average payments for SMI/SUD enrollees were highest at $13,541, followed by those with chronic physical conditions ($11,666). The COVID-19 target population had the lowest average payment ($5,629). Exhibit 75: Average Overall Payment for Services per WPC Enrollees by Enrollment Status and Target Population, PY 2 to PY6 $13,541 $11,666 $7,563 $8,481 $9,321 $6,272 a9 I ssars i ee [] 2 z 2 = 2 8 2 3 2 a 3 3 gs , 8 3 3 g a S S = os z & s 2 & & 5 as = 3 a z 6 “ = os a o o i. 3 3 & 2e E E 3 2 = 68 6 6 2 £ = = 7 & 5 3 z WPC Population Enrolled Target Populations Source: WPC Annual Invoices, PY 2 to PY 6 and Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes all payments for WPC services across all years of the program and includes services received prior to enrollment. Includes 289,417 unique individuals that received services through WPC: 224,632 enrolled and 64,785 never enrolled. Enrollees are included in target population if ever assigned to that target population during program. COVID-19 target population was added in PY 5. SMI/SUD is serious mental illness and/or substance use disorder. PY 2 is 2017 and PY 6 is 2021. UCLA also calculated the average monthly payment per enrollee for WPC services to account for different lengths of enrollment (Exhibit 76). On average, WPC Pilots were paid $397 per enrollee per month for all WPC enrollees. WPC Pilots were paid the most for the SMI/SUD target population ($670 per enrollee per month) and the least for the COVID-19 population ($241 per enrollee per month). ale WPC Services Offered and Delivered | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program Exhibit 76: Average Monthly Payment per WPC Enrollees Receiving Services for WPC Services Overall and by Target Population, PY 2 to PY 6 670 $605 $ $493 9512 $397 oer $407 UJ [J | J : Enrolled High Utilizers_ Chronic SMI/SUD Homelessness _ Risk of Justice covip 19 Physical Homelessness Involved Conditions wec Enrolled Target Populations Population Source: WPC Annual Invoices, PY 2 to PY 6 and Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021 Notes: Includes 224,632 unique individuals that received services through WPC and were enrolled. Enrollees are included in target population if ever assigned to that target population during program. COVID-19 target population was added in PY 5. SMI/SUD is serious mental illness and/or substance use disorder. PY 2 is 2017 and PY 6 is 2021. Whole Person Care Final Evaluation Report | WPC Services Offered and Delivered [BUSI UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Chapter 6: WPC Care Coordination A major goal of WPC was to “increase coordination and appropriate access to care for the most vulnerable Medi-Cal beneficiaries.” This chapter addresses the following evaluation question: “to what extent did WPC Pilots (a) improve comprehensive care coordination, including real- time coordination, across participating entities; and (b) achieve the approved application deliverables relating to care coordination?” In the Care Coordination Policy Brief (see Appendix K), published in October 2019, UCLA developed a conceptual framework that identified key elements needed for effective care coordination under WPC. This framework was developed following the Agency for Healthcare Research and Quality (AHRQ) definition of care coordination, interviews with Pilots, and a review of the literature on cross-sector care coordination. The interim report included 25 Pilot- specific case studies to highlight the activities of each Pilot according to this framework. The key elements of the framework included infrastructure needed to support effective care coordination, as well as specific care coordination processes. Infrastructure elements include: (1) care coordination staffing that meets patient needs, (2) data sharing capabilities to support care coordination, (3) standardized organizational protocols to support care coordination, and (4) financial incentives to promote cross-sector care coordination. Care coordination processes include: (5) ensuring frequent communication and follow-up to engage patients, (6) conducting needs assessments and develop comprehensive care plans, (7) actively linking patients to needed services across sectors, and (8) promoting accountability within the care coordination team. This framework was used to measure the progress Pilots made in implementing effective care coordination through WPC in the interim, as well as ensuring sustainability of the infrastructure and processes beyond the life of the Pilot. This chapter is structured around that conceptual care coordination framework, providing updates and additional nuanced detail since the WPC interim report. Data sources for this chapter included PY 3 (2018), PY 5 (2020), and PY 6 (2021) Lead Entity surveys and PY 6 follow-up interviews with leadership and frontline staff of all 26 Pilots. Additional qualitative data around challenges and solutions was obtained from WPC mid-year and annual narrative reports. The PY 5 and PY 6 data sources included updates on program implementation since the interim report as well as clarification and further detail on activities conducted since the start of WPC. For additional detail on data sources and methodology please see Methods Section and Appendices C, D and E. KEY) WPC Care Coordination | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Care Coordination Infrastructure Care Coordination Staffing that Meets Patient Needs In PY 3 LE surveys, the majority of Pilots (24 of 27) reported using shared care navigators or care coordinators across two or more participating WPC organizations to develop comprehensive care plans and coordinate care. In PY 5 LE surveys, UCLA asked about specific organizational involvement of these shared care coordinators. Most often shared care coordinators were from a health care organization (12 of 25), behavioral health care organization (11), and/or social service agency (9). Diversification of care coordinators allowed teams to access a broader range of resources for their enrollees. Most Pilots reported using community health workers, peer coaches/support specialists, or other staff with lived experience relevant to enrollees to provide care coordination services (18). These services were often provided in consultation with or under the supervision of staff with clinical expertise such as physicians, nurses, or social workers. Additionally, eight Pilots offered care coordination services outside of typical business hours (e.g., evenings or weekends). “Lived experience is a big one. Having a CHW who has been in your shoes and that you can identify with ... has been really critical... | personally believe that that takes a very special type of person... | do think that we did provide certain resources over the years about self-care, setting boundaries, trauma-informed care, how to take care of yourself...| think some of the CHWs who have been in the program since the beginning... are persistent and dedicated.” —Sacramento Average caseload ranged from approximately five, to over 300 enrollees per care coordinator depending on the structure of the program and the needs of the enrollees. For example, Contra Costa offered three tiers based on enrollee acuity, whereas Tier 1 was high acuity and had primarily field-based case management with a 1:80 case ratio. Tier 2 was moderate acuity, with enrollees receiving primarily telephonic support by community health workers with a 1:300 case ratio and Tier 3 was highest acuity with short-term and high-intensity case management focused on emergency department and inpatient hospital diversion and had a 1:25 case ratio. Median caseload across all Pilots was approximately 20 to 30 enrollees per care coordinator; specific breakdowns of caseload by Pilot is presented in Exhibit 1 in the WPC Snapshot Policy Brief. Whole Person Care Final Evaluation Report | WPC Care Coordination [B53 errr UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Additional detail on specific staffing models is provided below in the Care Coordination Staffing section of this chapter. Data Sharing Capabilities to Support Care Coordination Pilots demonstrated progress in data sharing capabilities from the interim report or PY3, in PY 5 LE surveys (Exhibit 77). For example, while all Pilots had established data sharing agreements with some partners, they reported an increase in such agreements with their key partners (20 of 25; compared to 15 of 27 in PY 3). Key partners were defined as those who have a high awareness of the WPC program structure and goals. As of PY 5, Pilots had the capability to access enrollees’ comprehensive care plans (21), needs assessments (19), and referrals (18) electronically in a single database (data not shown). Exhibit 77: Number of WPC Pilots Participating in Select Data Sharing Capabilities to Support Care Coordination, PY 3 and PY 5 Established data sharing agreements with all key partners 7 B a Used a single intergrated data system to track and report on 10 care coordination activities B Su Provided care coordination staff real-time access to data (e.g., notifications or alerts of enrollee events) b a mPY3 mPYS Sources: PY 3 Lead Entity (LE) Survey (n=27), June-September 2018; PY 5 Lead Entity (LE) Survey (n=25), June- August 2020. Notes: Key partners were defined as those who have a high awareness of the WPC program structure and goals. HUB WPC Care Coordination |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program In PY 6 follow-up interviews, most Pilots identified data and information technology infrastructure to support care coordination (e.g., case management platforms, real-time alerts, data sharing agreements) as a strategic priority of WPC and noted significant improvements from the Pilot’s inception. Pilots reported that frontline care coordination staff recognized benefits in their day-to-day workflows with efficiency, ability to see an enrollee’s history, and communication with multi-disciplinary partners. Information on how Pilots developed such infrastructure is provided in Chapter 3: Health Information Technology and Data Sharing Infrastructure. “A pretty big game changer. We used to do all of our assessments on paper, and then securely store those and write a summary online. But now we can actually complete them digitally. And we have more of an opportunity to show that work to other clinic staff. That wasn't as possible with our old system... we're getting a lot of information about a patient. The [primary care provider] can go just check out that encounter and see what happened with that patient. And that's a brand-new thing for us.” —Alameda Standardized Organizational Protocols to Support Care Coordination Developing standardized procedures and protocols to support care coordination was a priority for many Pilots. Standardized protocols helped to minimize undesirable variation in delivery of care coordination services, while improving staff workflows and data reporting. In PY 3 LE surveys, one third of Pilots reported that prior to WPC they had standardized protocols in place for referring enrollees to services (9 of 27). As indicated in PY 5 LE surveys, WPC increased the proportion of Pilots with protocols in place, with the majority of Pilots reporting they had standardized protocols for referring enrollees to medical, behavioral health, or social services (20 of 25), or had standardized protocols for monitoring and following up on whether enrollees needed services (16). Financial Incentives to Promote Cross-Sector Care Coordination All Pilots used per-member-per-month (PMPM) funding to support care coordination activities. In PY 5 LE surveys, 15 Pilots reported that their PMPM bundles were stratified by the risk or level of need of enrollees. Most Pilots contracted out some or all care coordination services for delivery by partner organizations (19); the remaining Pilots delivered care coordination services in-house, and did not contract out to partners. Whole Person Care Final Evaluation Report | WPC Care Coordination [BZ errr UCLA Center for Health Policy Research Health Economics and Evaluation Research Program In PY 6 LE surveys, 18 of 26 Pilots indicated that they provided financial incentives to partner organizations for engagement in WPC activities (e.g., stakeholder meetings, reaching specified milestones). On a scale from 0 (not effective) to 10 (extremely effective), Pilots rated these incentives as effective (6.8 of 10). More specifically, incentives to promote development of data sharing infrastructure within participating partner organizations and for Pilots to achieve set process targets were considered most effective. Care Coordination Processes Ensuring Frequent Communication and Follow-Up to Engage Patients In PY 6 follow-up interviews, Pilots emphasized the importance of using a patient-centered approach to communication that accommodated enrollee needs and preferences. All of the Pilots required care coordinators to regularly contact enrollees at least once per month. As indicated in PY 5 LE surveys, many Pilots (21 of 25) reported that the most common type of contact between care coordinators and enrollees was in-person, rather than by phone or other modes of communication. In PY 6 follow-up interviews, Pilots emphasized the importance of field-based and in-person communication for engaging enrollees in WPC, particularly those experiencing homelessness. While there were limitations to in-person engagement due to the COVID-19 pandemic, Pilots reported that several opportunities, such as Project RoomKey, emerged that allowed for more concentrated engagement of vulnerable populations. Needs Assessment and Comprehensive Care Planning Processes All Pilots were required to conduct needs assessments to identify target population needs and evaluate enrollee health progress over time. Specific needs assessment tools and their comprehensiveness varied, particularly when it came to evaluating social needs. In PY 5 LE surveys, 15 of 25 Pilots indicated utilizing a “homegrown” tool to assess enrollee’s non-medical needs and these were often tailored specifically to Pilot’s WPC enrollment criteria and program goals (data not shown). Fourteen Pilots reported using the VI-SPDAT (Vulnerability Index — Service Prioritization Decision Assistance Tool). Pilots also varied in whether they administered formal needs assessments to enrollees once per year, or more frequently (as indicated by 16 of 27 Pilots in PY 3). Outside of medical needs, information on housing and housing stability (all Pilots; 25 of 25) was most often collected as part of the needs assessment process, followed by access to other government benefits (23), food access (22), social supports (22), and interpersonal safety (18; Exhibit 78). KES WPC Care Coordination | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program Exhibit 78: Information Systematically Collected as Part of Needs Assessment Process in WPC Housing and housing stability [| 25 Access to other government benefits [i 33 Food access ee 22 Social supports ns 22 Interpersonal safety ey 18 Source: PY 5 Lead Entity (LE) Survey (n=25), June-August 2020. Oftentimes, needs assessments directly informed the development of comprehensive care plans. Almost all Pilots (23) reported that enrollees had a single, comprehensive care plan that was shared across all or some partners. Actively Linking Enrollees to Needed Services Across Sectors Linking enrollees to services to meet their health and social needs was a foundational component of care coordination in all WPC Pilots. In PY 5 LE surveys, Pilots reported using active referral strategies, such as providing/arranging transportation to and from appointments (24 of 25); ensuring warm hand-offs to other providers (24); and follow-up with enrollees and/or service providers to monitor referral status (23; Exhibit 79). Whole Person Care Final Evaluation Report | WPC Care Coordination [RY/ errr UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 79: Specific Approaches Used to Actively Link WPC Enrollees to Services and Integrate Care Ensure warm hand-offs to other providers i >, Provide or arrange transportation to and from appointments (i >. Follow-up with enrollees and/or service providers tO es >; monitor status of referrals Provide education or coaching around patient Sel a >> management education Accompany enrollees to appointments Ty 22 Regularly review data on enrollees with specific health risks ' i ee 20 to identify potential problems and gaps in care Assist with medication management and adherence TT 15 Implement disease management programs and/or strategies 7 for select health conditions Source: PY 5 Lead Entity (LE) Survey (n=25), June-August 2020. “.. our care managers are so amazing and work together so well, because they have their partner, which is their screening nurse... They give them real time warm handoffs. Like, you know, ‘This is the client. This is his number’ Sometimes they even call them right there in the office, if they don't have anybody waiting for them, as a warm handoff, so they get to know them, so they know it's a real person on the other end. And | know that a lot of my nurses, within 24 hours, they try to call them back, because they know that window of opportunity is right there and then..” —Riverside Promoting Accountability Within the Care Coordination Team Care coordination is most effective when accountability for different activities is clearly defined and monitored. In PY 5 LE surveys, many reported co-locating or otherwise embedding care coordinators within partner organizations (14 of 25). The most common types of co-located organizations were health care organizations (12), followed by mental health treatment agencies (10) and (non-housing) social service agencies (8). As emphasized in PY 6 follow-up interviews, WPC Pilots developed a variety of strategies to facilitate communication within care coordination teams. The primary mechanism for team HESS WPC Care Coordination |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program communication was regular in-person meetings, followed by phone calls, emails, and sometimes even text messages. Exhibit 80 illustrates the variety of strategies used by Pilots to promote accountability among care coordination teams, as indicated in PY 5 LE surveys. Data show 18 of 25 Pilots required staff to document, log, or otherwise track care coordination encounters and 18 Pilots had regular team meetings which promoted discussion by different stakeholders involved in a specific enrollee’s care. Exhibit 80: Number of WPC Pilots Engaging in Selected Strategies to Increase Care Coordination Team Accountability Requiring staff to document, log or otherwise track care ee coordination encounters Regular team meetings in which different stakeholders involved in enrollee care jointly discuss care of specific NS 3: enrollee At least weekly clinical supervision meetings involving care coordinators and supervisor Bb R Care coordinators held accountable for meeting pre- established targets in performance review 10 Source: PY 5 Lead Entity (LE) Survey (n=25), June-August 2020. Care Coordination Staffing Pilots developed multidisciplinary teams with relevant and diverse clinical expertise to address enrollee needs. As indicated in PY 5 LE surveys, across all Pilots, the most common roles involved in care coordination included: housing navigators (22 of 25), licensed social workers (19), community health workers or other staff with lived experience (18), and nurses (18). Exhibit 81 shows the types of staff involved in care coordination by Pilot. Outside of care coordination, staff may also have been involved in outreach, providing clinical consults, and/or supervision, depending on the structure of the Pilot. Most often community health workers or staff with lived experience (18) and housing navigators (15) conducted outreach. Licensed social workers (18) and nurses (17) most often provided clinical consults, and licensed social workers (13) and nurses (9) provided care team supervision (data not shown). Whole Person Care Final Evaluation Report | WPC Care Coordination [RUE) erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 81: Types of Staff Involved in WPC Care Coordination by Pilot other staff with lived experience [Medical assistant or equivalent (Alcohol or drug counselor or [Housing navigator or equivalent Physician or nurse practitioner (Community health worker or equivalent [Mental health counselor or Licensed social worker (e.g., equivalent |Unlicensed social worker [Benefits support staff Clinical psychologist >< _|Nurse (RN or LVN or PHN) Alameda Contra Costa < >< [PS [>< |MSW or LCSW) Kern >< [>< f>< ]>< Kings x x Xx x Los Angeles x Marin Mendocino x Monterey Orange Placer x Riverside x |< fox fox [>< ]>< fox Sacramento San Bernardino x San Diego San Francisco >< [>< [>< [>< ]>< fox [>< foe [>< ox fox San Joaquin x San Mateo >< [>< fox |>< ]>< fox Joe Jo fox fos }>< [>< fox x >< [>< fox >< [>< Jo< ]>< >< [>< fox [>< Santa Clara >< [>< fos foe [>< ]>< fox ]>< x |< fx fo [>< |< Santa Cruz x fx fox [>< [>< fox fox Shasta x >< [>< fos fox Small County — Mariposa Small County — San Benito >< [>< fox fox >< J>< Solano Sonoma xX x x x x Ventura x x x x x x Overall 18 12| 18 19 | 16 14 16 22 | 12 4 5 Source: PY 5 Lead Entity survey (n=25), June-August 2020. Notes: RN is registered nurse. LVN is licensed vocational nurse. PHN is public health nurse. MSW is Master of Social Work. LCSW is licensed clinical social worker. Kiley) WPC Care Coordination | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program Pilots reported difficulty in recruitment and retainment of different types of staff. Generally, Pilots found it most challenging to recruit nurses and/or licensed social workers. Pilots found it most difficult to retain licensed social workers, housing navigators, and community health workers (data not shown). In PY 6 follow-up interviews, Pilots noted that the most common challenge faced by staff was the demanding nature and high emotional burden associated with inherent responsibilities of the job. Based on geographic location, some Pilots mentioned staff challenges related to high cost of living and long commute times. Pilots offered a wide variety of supports for staff responsible for care coordination (Exhibit 82). As indicated in PY 5 surveys, all Pilots provided opportunities for shared learning via collaborative care planning or joint discussion of cases. Other common offerings included: clinical skills training (23 of 25); team training or inter-personal training (23); shadowing of other care coordinators/providers (22); and clinical supervision by a formally designated supervisor (20). Exhibit 82: Resources in Place to Support Staff Responsible for Care Coordination Opportunities for shared learning via Collaborative Care EE 25 planning or joint discussion of cases Clinical skills training (e.g. trauma-informed Care, ES 23 motivational interviewing) Team training or inter-professional training IT 23 Shadowing of other care coordinators or providers ET 27 Clinical supervision by a formally designated supervisor I 20 Supportive supervision provided by a formally designated pe 15 supervisor Standardized protocols for how communication about ee 17 training will be disseminated to staff Formal orientation for new hires that lasts longer than One es 14 day Source: PY 5 Lead Entity (LE) Survey (n=25), June-August 2020. Notes: Clinical supervision is defined as opportunities for supervisor and supervisee discuss specific cases, determine courses of action, and resolve problems related to a case; whereas supportive supervision is defined as a focus on discussing non-clinical issues, decrease job-related stress, improve staff motivation and morale. Whole Person Care Final Evaluation Report | WPC Care Coordination erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Challenges and Successes Exhibit 83 summarizes the most frequently identified challenges related to care coordination by program year as presented by Pilots in bi-annual narrative reports. Overall, the most common theme across the life of WPC was challenges related to limited availability and/or accessibility of services (72 unique mentions across reporting periods by 24 Pilots; data not shown). WPC Pilots most commonly referenced housing-related issues, including: long wait times for existing permanent housing stock, limited housing options available within the county, poor quality and fit for enrollees among the available housing units, and how the lack of housing prevented other desired health and social outcomes among enrollees. Additional examples of challenges WPC Pilots discussed regarding limited availability and accessibility of services included: increased referrals on an already overburdened system prevented access to needed services for WPC enrollees and a lack of specialty care, substance use, and mental health treatments within county limits. However, the prevalence of this challenge became less dominant in later reporting periods (PY 5 and PY 6), as Pilots became more familiar with access and referral pathways to services through partnerships. With the COVID-19 pandemic, there was also an increase in the availability of temporary and short-term housing options for vulnerable populations. There was a peak of 22 mentions in PY 4, with 10 mentions in PY 6. Pilots also expressed difficulty engaging appropriate interdisciplinary partners as a barrier to care coordination (67 unique mentions across reporting periods by all 25 Pilots; data not shown). For example, multiple WPC Pilots reported that partners were unwilling or hesitant to engage due to their competing priorities with other programs or initiatives. Initially, WPC LEs mentioned limited trust and buy-in from partners to the WPC program. However, the prevalence of this challenge became less dominant in later reporting periods (PY 5 and PY 6), as partnership networks strengthened and strategic goals aligned. There was a peak of 20 mentions in PY 4, with five mentions in PY 6. Pilots experienced staffing issues including recruitment, training, retention, and turnover which negatively impacted care coordination activities (57 unique mentions across reporting periods by 20 Pilots; data not shown). Multiple WPC Pilots explicitly attributed staffing challenges to cumbersome county hiring and/or contracting processes (e.g., background checks, requirements for open search). These challenges required WPC Pilots to plan far ahead when developing project timelines, which was challenging early in the implementation process. Later in the implementation process, staff questioned their job security with the inevitable end of the Pilot, which may have led to turnover. There was a peak of 17 mentions in PY 4, and six mentions in PY 6. Ky) ~WPC Care Coordination | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program A somewhat consistent theme across reporting periods was challenges in understanding WPC target populations and how to address their complex and evolving needs (46 unique mentions across reporting periods by 21 Pilots; data not shown). Oftentimes, staff found that enrollees were of particularly high acuity or had undocumented diagnoses. This theme was reported by 11 to 12 Pilots in key implementation years of PY 3 to PY 5. Competition or confusion with other similar programs was a less common theme related to challenges in care coordination (32 unique mentions across reporting periods by 18 Pilots; data not shown). Care coordination and case management services were often offered through a variety of agencies and organizations, such as behavioral health departments and Medi-Cal managed care plans, which created confusion regarding WPC scope and concerns around non- duplication of services. This theme had nine mentions in PY 2, a peak of 11 mentions in PY 4, with four mentions in PY 6. Exhibit 83: Commonly Identified Challenges in Care Coordination Among WPC Pilots, by Reporting Period, PY 2 to PY 6 Limited availability and/or accessibility of services being 22 coordinated Engagement of appropriate interdisciplinary partners Staffing issues Understanding the population and how to address needs and complexity Competition or confusion with other programs | — 1 PY2 mPY3 mPY4 MPYS MPY6 Source: WPC Mid-Year and Annual Narrative Reports, PY 2-PY 6. Notes: Numbers indicate WPC Pilots that mentioned the thematic challenge at least once within the given program year. Themes are presented in order of overall prevalence across reporting periods. Program Year (PY) 2 = 2017, PY 3 = 2018, PY 4 = 2019, PY 5 = 2020, and PY 6 = 2021. Whole Person Care Final Evaluation Report | WPC Care Coordination erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Successes in implementing care coordination services and programs often directly reflected a response to the challenges detailed above (Exhibit 84). Across reporting periods, all Pilots reported solutions related to implementation of new or improved care coordination services; many of these efforts focused on improvements in the day-to-day activities of frontline staff (110 unique mentions across reporting periods by 25 Pilots; data not shown). Commonly identified examples of successes within the delivery of care coordination services included: organizing regular case conferences with partners and managed care plans to discuss high-need enrollees, prioritization of services or housing for WPC enrollees including reserved appointments, set-aside housing vouchers, and effective communication across the entire care team. This theme was consistently reported with 23-25 mentions in each period from PY 3 to PY 6. Pilots also reported successes in using data systems to support care coordination activities (65 unique mentions across reporting periods by 24 Pilots; data not shown). Many WPC Pilots reported having procured care management platforms, which helped to streamline important care coordination activities and share relevant enrollee information amongst multiple users involved in the enrollee’s care. This theme was consistently reported across all reporting periods. Pilots described successes in working with partners in new ways that improved understanding of mutual goals for shared clients (e.g., warm handoffs of enrollees after an emergency department visit, direct communication through electronic platforms; 60 unique mentions across reporting periods by 24 Pilots; data not shown). WPC Pilots emphasized proactive and consistent communication amongst partners, and formalized contracts to facilitate implementation of care coordination activities among partners with historically limited interaction. This theme had nine mentions in PY 2, a peak of 11 mentions in PY 4, with four mentions in PY 6. Pilots reported successes for WPC enrollees as a result of effectively utilizing synergies with existing programs and initiatives, particularly because many programs have similar goals and provide care to the same populations (44 unique mentions across reporting periods by 20 Pilots; data not shown). Typically, these successes involved the Pilots working with other programs to identify and delineate their respective roles and responsibilities with WPC enrollees. One particularly successful complementary initiative was Project Roomkey, a part of comprehensive COVID-19 response. This theme was consistently reported from PY 3 to PY 6. Pilots also defined care coordination and worked to comprehensively understand care coordination needs across agencies including alignment of enrollee assessment tools across partners, tracking of metrics, and establishment of referral pathways (31 unique mentions KES) WPC Care Coordination | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research errr Health Economics and Evaluation Research Program across reporting periods by 18 Pilots; data not shown). This theme had a peak of 10 mentions in PY 3 when WPC was becoming established with partners, and seven mentions in PY 6, likely with preparation for the transition to Cal-AIM. Exhibit 84: Commonly Identified Successes in Care Coordination Among WPC Pilots, by Reporting Period, PY 2 to PY 6 24 5 23 s 16 11 9 £ 3 8 8 =, Defining care coordination and understanding needs across agencies PY2 mPY3 mPY4 MPYS MPY6 Source: WPC Mid-Year and Annual Narrative Reports, PY 2-PY 6. Notes: Numbers indicate WPC Pilots that mentioned the thematic challenge at least once within the given program year. Themes are presented in order of overall prevalence across reporting periods. Program Year (PY) 2 = 2017, PY 3 = 2018, PY 4 = 2019, PY 5 = 2020, and PY 6 = 2021. Whole Person Care Final Evaluation Report | WPC Care Coordination Reenter UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Chapter 7: WPC Quality Improvement and Program Monitoring DHCS provided several forms of support to Pilots to promote successful implementation of WPC. DHCS contracted with external organizations and provided support from a DHCS analyst to assist with preparing data and reports. Pilots were also required to engage in regular performance improvement activities and submit bi-annual Plan-Do-Study-Act (PDSA) reports to DHCS documenting Pilot-led efforts to improve workflows and metric performance. This chapter outlines Pilots’ involvement in PDSAs and technical assistance provided to Pilots from DHCS. This chapter also examines the frequency and extent to which stakeholder engagement influenced design, implementation, and evaluation of Pilots. Additional detail on performance improvement and program monitoring was provided in the interim report. Data sources for this chapter include PY 6 LE surveys and follow-up interviews with leadership and frontline staff. Data from bi-annual PDSA Reports is also included in the following analyses. For additional detail on data sources and methodology please see Appendices G. Pilot-Initiated Quality Improvement All Pilots were required to monitor progress on selected performance measures and to utilize a quality improvement approach known as “Plan-Do-Study-Act” (PDSA) to improve Pilot performance. The bi-annual Pilot reports included the PDSA activities that were implemented during that reporting period. PDSA Types WPC Pilots submitted several different categories of PDSAs to DHCS reflecting their WPC program goals, target populations, and infrastructure and process goals. The categories of PDSAs reported by Pilots included: (1) ambulatory care, (2) care coordination, (3) comprehensive care plan, (4) data, (5) inpatient utilization, and (6) other (as cited in WPC STCs). DHCS required four PDSAs on ambulatory care, inpatient utilization, and comprehensive care plan per year and two PDSAs on data and care coordination per year. DHCS did not set specific criteria on the length of quality improvement efforts and used the term PDSA to refer to a variety of quality improvement activities. All Pilots conducted at least one PDSA that was considered long-term and had different stages depending on program planning and implementations phases. sEY WPC Quality Improvement and Program Monitoring | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program The data show that ambulatory care PDSAs typically focused on efforts to reduce use of the emergency department for ambulatory care sensitive conditions. A second category of PDSAs were around creation of a comprehensive care plan. Comprehensive care plans were to be developed and accessible to the entire care team to outline goals and services once enrolled into WPC. Across all Pilots, as part of a universal metric, the goal was for comprehensive care plans to be accessible within a 30-day timeframe. Care coordination PDSAs focused on how to improve coordination of care. Some elements of care coordination explored through PDSAs included navigation infrastructure, coordinated entry, common assessment tools used among participating entities, collection and use of social determinants data, and increased access to social services. Data and reporting PDSAs were usually intended to improve methods for capturing and storing data, particularly as it related to reporting to DHCS. Inpatient utilization PDSAs were projects aimed to reduce inpatient utilization; some Pilots focused on a particular target population with high rates of inpatient utilization. Appendix G provides an example of PDSAs by each category type, since the interim report. Whole Person Care Final Evaluation Report | WPC Quality Improvement and Program Monitoring El ery UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Volume of PDSAs Conducted by WPC Pilots, PY 3-PY 6 Multiple PDSAs were submitted during each reporting period across each category; the number of PDSA reports submitted to DHCS varied by WPC Pilot per reporting period. On average, Pilots completed nine PDSAs per reporting period. Overall, 2,133 PDSAs reports were submitted to DHCS through reporting periods PY 2 mid-year and PY 6 annual. Of those 2,133 reports submitted, the most common categories submitted included: ambulatory care PDSAs (19%, 398 reports), followed by care coordination PDSAs (18%, 381 reports), and inpatient utilization PDSAs (17%, 370 reports; Exhibit 85). The “other; metrics” category was created based on PDSAs that were submitted that did not fit into any of the provided categories but were metric-specific. Examples of PDSAs from the “other” category included projects that Pilots wished to pursue but that did not neatly fit into existing categories. Exhibit 85: WPC PDSA Category Types Across Reporting Periods, PY 2 to PY 6 Ambulatory care in 398 Care coordination ie 3:1 Inpatient utilization in 370 Comprehensive care plan [i 357 Cote 305 Other i 215 Other (metrics) 104 Source: Bi-annual PDSA Reports, PY 2-PY 6 (n=25). WPC Quality Improvement and Program Monitoring |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program In PY 6 follow-up interviews, some Pilots provided additional detail on other quality and performance improvement and monitoring activities that were not captured through PDSA reports submitted to DHCS. Selected examples are provided in Exhibit 86. Exhibit 86: Selected Illustrative Examples of WPC Quality and Performance Improvement and Monitoring Activities Pilot Selected Example Santa Cruz Santa Cruz conducted a Lean Six Sigma Green Belt training with all WPC staff, as well as CBO partners, to collectively gather and develop strategies on process improvement. A key focus of this training was to strengthen the ability of organizations to work together. Santa Cruz also conducted a “root cause” analysis, which provided insights into the complexity of underlying challenges faced by the program. The conclusions from this training were used to inform strategic goals for the future. San Bernardino San Bernadino held "WAR conferences" (Whole Person Care Accountability Review), in which all care team members discussed critical issues facing each individual client. This process helped to illuminate “best practice” strategies, with generalizable lessons learned that informed care team interactions with enrollees. Riverside When determining areas of focus for required PDSA reports to DHCS, Riverside program management obtained feedback from frontline staff who worked directly with enrollees. PDSA reporting facilitated important conversations between frontline staff and program management. Napa Napa created an annual participation survey to assess enrollee satisfaction with WPC services. Napa also received feedback through their partners by holding semi-annual interviews on WPC’s progress and areas for improvement. Napa discussed feedback and used it to improve the program. Marin Marin partnered with a consulting firm to perform a qualitative evaluation, which included interviews with case managers and organizational leadership. Based on the evaluation, Marin was able to self-assess and make improvements to their Pilot. Source: PY 6 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=26), June-September 2021. Technical Assistance Since the interim report, DHCS along with the Learning Collaborative team from Aurrera Health (previously Harbage Consulting) continuously checked in with the LEs through surveys, phone calls, virtual meetings, and email communications to better understand the issues that were of most interest and concern to help guide Learning Collaborative content. An online portal was created to share information across Pilots and participating organizations. The portal was managed by Center for Health Care Strategies (CHCS). In PY 6, the Learning Collaborative primarily supported the conclusion of the WPC Pilots and transition to new Medi-Cal benefits and services under the state’s California Advancing and Innovating Medi-Cal (CalAIM) initiative, including the new Enhanced Care Management (ECM) benefit and Community Supports (CS). Additional information on this technical assistance is provided in WPC Transition to CalAIM chapter. Whole Person Care Final Evaluation Report | WPC Quality Improvement and Program Monitoring eerie 6 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program In PY 6 follow-up interviews, Pilots expressed that they would have benefited from additional technical support from DHCS around standardizing data collection, particularly considering metrics and reporting requirements. Stakeholder Engagement on Quality Improvement Activities Many Pilots attempted to integrate and elevate stakeholder perspectives into their Pilot. In PY 6 surveys, Pilots were asked about stakeholder engagement in the design, implementation, and evaluation of key WPC activities. Eighteen of 26 Pilots felt they had allocated sufficient resources (e.g., time, staff, compensation) to capture key stakeholder input (e.g., frontline staff, enrollees, other community members) throughout their WPC Pilot (data not shown). “We did host a lot of focus groups where a lot of staff were able to come to those focus groups and voice what they've been experiencing with their clients. And then we took that information and built workflows and protocols for all Staff to how to assist with that. And then we did trainings on those report flows and protocols to make sure everybody was on the same page.” -Contra Costa Exhibit 87 shows the frequency of stakeholder involvement during various stages of the WPC Pilot. Across all three stakeholder categories, reported involvement was highest during the Pilot design phase, with enrollees and other community members engaging often (e.g., once a month). All groups were less involved during the implementation phase, but occasionally (e.g., quarterly) were involved in aspects of the evaluation phase. Overall, enrollees and other community members were most frequently involved, while frontline staff were reported to be the least involved. aI) WPC Quality Improvement and Program Monitoring |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ely) Health Economics and Evaluation Research Program Exhibit 87: WPC Pilots’ Rating of Frequency of Involvement of Stakeholders in Aspects of Quality Improvement Activities 4 4 33 3 3 29 2.2 2 15 Design Implementation Evaluation Frontline staff mEnrollees Other community members Source: PY 6 Lead Entity (LE) Survey (n=26), May-June 2021. Notes: Ratings on scale of 1=Never, 2=Rarely/Once each year, 3=Occasionally/Once each quarter, 4=Often/Once each month, 5=Always/At every decision-making point, regarding frequency of involvement. “Frontline staff” is defined as those responsible for delivering WPC services, such as community health workers, care managers, peer support within LE or partner organizations and “other community members” is defined as individuals not enrolled in WPC but that could represent perspectives of communities that could benefit from WPC services. Whole Person Care Final Evaluation Report | WPC Quality Improvement and Program Monitoring LAO TPE UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Despite being less frequently involved, frontline staff were perceived by Pilots as having greater influence in aspects of quality improvement efforts for design, implementation, and evaluation, whereas enrollees were perceived by Pilots as having the least amount of influence (Exhibit 88). Exhibit 88: WPC Pilots’ Rating of Extent of Stakeholder Influence on Quality Improvement Activities 77 73 53 5.4 53 51 41 3.4 U Design Implementation Evaluation Frontline staff ™ Enrollees Other community members Source: PY 6 Lead Entity (LE) Survey (n=26), May-June 2021. Notes: Ratings on scale of 1=not at all and 10=great extent, regarding extent of influence of involvement. “Frontline staff” is defined as those responsible for delivering WPC services, such as community health workers, care managers, peer support within LE or partner organizations and “other community members” is defined as individuals not enrolled in WPC but that could represent perspectives of communities that could benefit from WPC services. aI-y) WPC Quality Improvement and Program Monitoring |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Chapter 8: WPC and COVID-19 The COVID-19 pandemic began early in PY 5 (2020), and significantly impacted Pilots and enrollees. Due to the pandemic, in December 2020, DCHS received approval from the Centers for Medicare and Medicaid Services (CMS) to extend WPC for one year, through December 31, 2021. Furthermore, DHCS added a new COVID-19 target population in the third quarter of 2020, which could be retroactively applied to enrollees if Pilot elected to use it. UCLA presented initial findings on the impact of COVID-19 through the end of 2020, including progression of the COVID-19 in WPC counties, the estimated prevalence of COVID-19 among WPC enrollees, and the changes in healthcare service utilization during the pandemic compared to the year prior, in a related policy brief. The analysis presented in this chapter updates some of these findings to include data from 2021. This chapter addresses the following evaluation questions, which were added post-pandemic as part of the WPC extension: (1) how did WPC infrastructure and processes facilitate Pilot’s COVID-19 response? (2) What were the changes to WPC implementation due to COVID-19? (3) What was the impact of the COVID-19 pandemic on WPC enrollment, utilization of healthcare services, and services offered? This analysis is further needed to assess whether the impact of COVID-19 was similar on WPC enrollees and the control group when measuring the impact of WPC program. Data sources for this chapter include the PY 5 COVID-19 impact survey, PY 6 (2021) follow-up interviews with leadership and frontline staff, Medi-Cal enrollment and claims data, and Quarterly WPC Enrollment and Utilization Reports. Additional qualitative data around challenges and solutions was provided in the 25 WPC mid-year and annual narrative reports by Pilots. For more detail on data sources and methodology please see Appendices C, D, and E. Whole Person Care Final Evaluation Report | WPC and COVID-19 Reaper 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Progression of COVID-19 in WPC Counties Over 5.5 million confirmed COVID-19 cases and 76,448 resulting deaths were reported in California through December 2021 with peaks occurring at different time points throughout the pandemic (data not shown). When examining 14-day average daily case rate in WPC counties, we found four distinct peaks: late July 2020 (21 confirmed cases per 100,000), early January 2021 (79 confirmed cases per 100,000), late August 2021 (35 confirmed cases per 100,000) and late December 2021 (65 confirmed cases per 100,000; Exhibit 89). Most WPC counties had peaks in the same time frame, but there were variations in the magnitudes of these peaks by county. Trends in 14-day average daily hospitalizations from COVID-19 mirrored trends in confirmed cases, with the average rate in WPC counties peaking between 14 and 37 hospitalized for COVID-19 per 100,000 around the time of the peak in cases. Exhibit 89: 14-Day Average Daily Confirmed COVID Cases and Hospitalizations per 100K for WPC Counties, April 2020 to December 2021 200 180 160 140 120 100 80 60 40 20 Sf Pry YY YD DY YP DY YD YD o oO oO PP oP Do oD # PP Pr PP DO : PF FPF HK FM KP KK MH KW ® oO PP & wer sv 2 Low Average —==High + +++++ Average (Hospitalizations) Source: Daily new cases and hospitalizations report by the Los Angeles Times and the July 2019 U.S. Census population estimates. Note: Low, average and high are the lowest, average and highest county-specific rates of COVID cases among WPC- participating counties per 100,000 county residents. Includes all 27 WPC counties. Informed by daily rates from March 29, 2020 to December 31, 2021. Ate) WPC and COVID-19 | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Impact of COVID-19 on WPC Implementation and Infrastructure UCLA assessed how infrastructure and processes established through WPC may have helped with Pilots’ COVID-19 response and the potential impact of the COVID-19 pandemic on WPC elements such as staffing, engagement, and care coordination processes and workflows. Early pandemic impacts were measured by UCLA in a rapid survey administered in April 2020 (PY 5) and subsequently reported in a Health Affairs blog. How WPC Infrastructure and Processes Facilitated COVID-19 Response In the PY 5 COVID-19 impact survey, Pilots were asked to indicate how WPC informed or otherwise impacted their COVID-19 response on a scale of one (not at all) to five (great extent; Exhibit 90). Pilots reported that all WPC elements impacted COVID-19 response, although to varying degrees. Most WPC elements (7 of 8) had a mean impact score greater than four, suggesting that existing WPC infrastructure and processes impacted Pilots’ COVID-19 response efforts. On average, WPC staff had the highest degree of impact (4.7) while relationships with housing providers had the lowest (3.7). Exhibit 90: WPC Informing or Impacting COVID-19 Response by Program Element, PY 5 Number.of pilots (te24) Mean Extent to Which the that Reported the Element Informed/Impacted WPC Element Element Informed or (dencvatele 2 eres Impacted COVID-19 ee extent) Response WPC staff offered skills and expertise 96% 47 WPC care coordination processes influenced ag COVID-19 workflows a8 46 Existing relationships with health and behavioral 88% 46 health partners facilitated COVID-19 response “ Existing relationships with social service partners 88% 46 facilitated COVID-19 response ° - Other WPC services (i.e., outside of care 75% 46 coordination) offered additional resources = ° Existing relationships with Medi-Cal managed care 88% 44 plans facilitated COVID-19 response “ WPC inf tion technol: ted dati information technology promoted data 96% 43 sharing Existing relationships with housing providers 96% 37 facilitated COVID-19 response " i Source: PY 5 COVID-19 Impact Survey (n=25,), April 2020. Notes: 24 of 25 Pilots reported that the elements informed/impacted COVID-19 response; percentages presented are with 24 as the denominator. "Care coordination processes" includes items such as intake and assessment, development of comprehensive care plan, and referrals. “Other WPC services” includes services such as recuperative care, sobering centers, and medical transportation. Elements were rated on a scale of 1 to 5, where 1 = “not at all”, 2 = “very little”, 3 = “somewhat”, 4 = “moderate”, and 5 = “great extent”. Whole Person Care Final Evaluation Report | WPC and COVID-19 ery UCLA Center for Health Policy Research Health Economics and Evaluation Research Program “Prior to WPC, care was provided primarily through a medical lens and has [now] been expanded to include social determinants of health... While WPC alone did not create all changes, it was a strong contributing focus to the cultural shift underway. The skills and resources are transferrable... [and has been] particularly beneficial during the COVID-19 crisis. WPC has helped to build increased knowledge, relationships, resources, and coordination across many of the distinct programs within the health system and its’ community partners.” -Santa Clara Exhibit 91 shows the breakdown of impact score by WPC program element. Most Pilots reported that using WPC staff greatly impacted their ability to respond to the pandemic (18 Pilots providing a score of 5); fewest Pilots (10) reported it greatly improved their relationships with housing providers. Exhibit 91: Reports of WPC Informing or Impacting COVID-19 Response by Program Element and Extent, PY5 w i} WPC staff offered skills and expertise WPC information technology promoted data sharing Existing relationships with housing providers facilitated COVID-19 response = rs wu me w & 10 Existing relationships with health and behavioral health partners facilitated COVID-19 response a WPC care coordination processes influenced COVID-19 workflows 2 Existing relationships with social service partners facilitated COVID-19 response Existing relationships with Medi-Cal managed care plans facilitated COVID-19 response eS ny 5 ny N a Other WPC services (i.e., outside of care coordination) offered additional resources @1=Notatall m2=Verylittle m™3=Somewhat m4 = Moderate 5 = Great extent Source: PY S COVID-19 Impact Survey (n=25), April 2020. Notes: "Care coordination processes" includes items such as intake and assessment, development of comprehensive care plan, and referrals. “Other WPC services” includes services such as recuperative care, sobering centers, and medical transportation. Elements were rated on a scale of 1 to 5, where 1 = “not at all”, 2 = “very little”, 3 = “somewhat”, 4 = “moderate”, and 5 = “great extent”. nies) WPC and COVID-19 | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program WPC Staff Offered Skills and Expertise Through WPC, staff had been formally trained in outreach and engagement, screening, and referrals and had experience working with vulnerable populations that would be at highest risk for COVID-19 (e.g., homeless, individuals with chronic conditions). Skills developed through WPC may have helped find and house or shelter high-risk homeless individuals, provide operational support for isolation hotels for high-risk individuals experiencing homelessness, and inform screening processes for COVID-19. Ongoing case management was necessary for proactively managing enrollees and individuals most at-risk for COVID. As a result of this, many WPC staff were directly involved in their County’s coordinated COVID-19 response. WPC Information Technology Promoted Data Sharing Data sharing agreements and platforms were utilized to identify individuals at highest risk of COVID-19 and plan COVID-19 response. Systems were used to create dashboards and monitor COVID-19 cases, as well as provide updates on hospital and clinic capacity. Other WPC Services Offered Additional Resources Other WPC services, particularly existing networks for providing medical transportation, proved helpful. In some cases, Pilots redirected resources in mental health transitional care, recuperative care, and sobering centers; they used these resources to expand hospital capacity for COVID-19 patients. Relationships with Partners Facilitated COVID-19 Response Pilots reported that preexisting relationships allowed counties to leverage WPC resources (e.g., outreach to vulnerable populations, care coordination for COVID-19 patients, understanding legal requirements for obtaining consent) in confronting the pandemic. Existing relationship networks were utilized for communication and dissemination of public health messaging, as well as to assess need and develop plans (e.g., emergency department protocols, acquiring and distributing personal protective equipment). Key relationships included those with health and behavioral health partners, social service agencies, Medi-Cal managed care plans, and housing providers. Whole Person Care Final Evaluation Report | WPC and COVID-19 eAEOel 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 92 highlights illustrative examples from Pilots on how each WPC element was incorporated into their COVID-19 response efforts. Pilots continually emphasized the advantages of WPC to counties because it had helped establish the infrastructure, staff, relationships, and experiences needed for an effective COVID-19 response. “The value of having this kind of program cannot be understated. The services provided reduce overall costs to the system in everyday practice and the way our program works helps the county respond more effectively and more efficiently in a crisis situation.”-Placer WPC Staff Offered Skills and Expertise Through WPC, staff had been formally trained in outreach and engagement, screening, and referrals and had experience working with vulnerable populations that would be at highest risk for COVID-19 (e.g., homeless, individuals with chronic conditions). Skills developed through WPC may have helped find and house or shelter high-risk homeless individuals, provide operational support for isolation hotels for high-risk individuals experiencing homelessness, and inform screening processes for COVID-19. Ongoing case management was necessary for proactively managing enrollees and individuals most at-risk for COVID. As a result of this, many WPC staff were directly involved in their County’s coordinated COVID-19 response. WPC Information Technology Promoted Data Sharing Data sharing agreements and platforms were utilized to identify individuals at highest risk of COVID-19 and plan COVID-19 response. Systems were used to create dashboards and monitor COVID-19 cases, as well as provide updates on hospital and clinic capacity. Other WPC Services Offered Additional Resources Other WPC services, particularly existing networks for providing medical transportation, proved helpful. In some cases, Pilots redirected resources in mental health transitional care, recuperative care, and sobering centers; they used these resources to expand hospital capacity for COVID-19 patients. Relationships with Partners Facilitated COVID-19 Response Pilots reported that preexisting relationships allowed counties to leverage WPC resources (e.g., outreach to vulnerable populations, care coordination for COVID-19 patients, understanding legal requirements for obtaining consent) in confronting the pandemic. Existing relationship networks were utilized for communication and dissemination of public health messaging, as Aist3) WPC and COVID-19 | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program well as to assess need and develop plans (e.g., emergency department protocols, acquiring and distributing personal protective equipment). Key relationships included those with health and behavioral health partners, social service agencies, Medi-Cal managed care plans, and housing providers. Exhibit 92: Illustrative Examples of How WPC Informed or Impacted COVID-19 Response WPC Element Examples Pilot WPC staff offered skills and expertise Social workers and nurses had developed extensive experience working with vulnerable and medically complex populations, particularly with homeless individuals who were at high risk of COVID-19. Training and protocols for WPC effectively translated to COVID-19 response. Placer Santa Clara deployed WPC staff in partnership with team members from the Office of System Integration and Transformation to support COVID-19 operations at the hospital command center. Staff members were selected due to their subject expertise, leadership, and established interagency relationships. Santa Clara WPC information technology promoted data sharing Mendocino utilized their data sharing platform developed through WPC for COVID-19 response, which allowed WPC staff to identify and manage information for high risk, vulnerable individuals experiencing homelessness. It further enabled WPC staff to identify and contact enrollees that qualified for early access to COVID-19 vaccination based on demographics and health status. Mendocino Santa Clara created dashboards for WPC staff which provided regular updates on COVID-19 guidelines and best practices. The platform had a question-and-answer feature. Santa Clara WPC care coordination processes influenced COVID-19 workflows WPC staff assisted the county in screening the general population for COVID-19 at drive-through locations. WPC registered nurses also helped determine emergency housing eligibility for enrollees. Riverside Alameda modified existing WPC referral protocols for referrals to COVID-19 homeless isolation hotels. Alameda Other WPC services (i.e., outside of care coordination) offered additional resources San Diego expanded medical respite capacity to decrease hospitalization and emergency department visits for WPC high utilizers; this allowed for increased capacity for hospitals to manage COVID-19 patients. San Diego WPC shower pods were used to screen and engage with people experiencing homelessness, connecting them to WPC resources. Ventura Relationships with partners facilitated COVID-19 response Orange leveraged health plan relationships to assist with additional medical oversight of shelters and alternate care sites with heightened COVID-19 activity. Orange Ventura continued working with their health and behavioral partners while developing new ways to coordinate support for Ventura Whole Person Care Final Evaluation Report | WPC and COVID-19 ery UCLA Center for Health Policy Research Health Economics and Evaluation Research Program WPC Element Examples Pilot hotel sites. For example, they delivered medication assistance treatment/addiction medicine services directly to hotel sites to support social distancing. Source: PY 5 COVID-19 Impact Survey (n=25), April 2020 and PY 6 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=26), June-September 2021. Impact of COVID-19 on WPC In the PY 5 COVID-19 impact survey, Pilots were also asked to indicate if specific WPC processes, procedures, or policies were impacted by COVID-19. Most Pilots reported an impact on staffing policies and procedures (21 of 24; Exhibit 93), which included shifts to telework and protocols for use of personal protective equipment (PPE). Twenty Pilots indicated changes in engagement of eligible beneficiaries or enrollees in WPC services. The remote model often resulted in fewer engagements due to reduced face-to-face interactions, particularly with hard-to-reach populations such as homeless individuals who might not have reliable and consistent access to a phone. “Our program is 100% outreach. We do communicate with the clients via telephone, text, and e-mail, but this is only a temporary solution and a hindrance to the services we provide our clients. Nothing will replace the personal connections of the in-person encounters.”-San Bernardino Nineteen Pilots indicated changes in care coordination policies or processes. These Pilots reported shifting at least some care coordination activities to be done remotely, over phone or video conferencing. Pilots noted mixed results with some that found enrollees demonstrated increased independence in fulfilling their healthcare needs and others that had challenges understanding enrollee needs and progress without in-person interactions. Specific enrollee factors and demographics could promote or hinder success of remote care coordination. Less than half of Pilots (11) reported an impact on enrollment of eligible beneficiaries in WPC and identifying beneficiaries eligible for WPC (10). Despite the pandemic, criteria for identifying eligible beneficiaries for WPC didn’t significantly change because it often already included the most vulnerable individuals. Some Pilots did broaden criteria to include individuals who tested positive or were at highest risk for COVID-19, but frequently found overlap with existing target populations. hfe WPC and COVID-19 | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 Exhibit 93: Pilot Reports of COVID-19 Impact on WPC Processes, Procedures, or Policies, PY 5 Staffing policies and procedures Engagement of eligible beneficiaries or enrollees in WPC services Care coordination policies or processes Enrollment of eligible beneficiaries in WPC Identifying beneficiaries eligible for WPC Source: PY 5 COVID-19 Impact Survey (n=24), April 2020. Exhibit 94 highlights illustrative examples from Pilots on how each WPC process, procedure, or policy was impacted by COVID-19. Exhibit 94: Illustrative Examples of COVID-19 Impact on WPC Processes, Procedures, or Policies Process/Policy/Procedure Examples Pilot Staffing policies and procedures (e.g., shift to telework, protocols for use of PPE) In Contra Costa, many staff were disaster service workers who were deployed to work in command centers, testing sites, and alternative care sites, shifting attention away from WPC roles. Contra Costa Placer felt the shift to telework increased efficiencies for staff, reducing commute times and allowing for additional flexibility. Placer Engagement of eligible beneficiaries or enrollees in WPC services (e.g., field-based outreach) San Francisco continued engagement in shelters and on the streets, incorporating social distancing and safety measures. San Francisco San Benito discontinued field-based outreach due to the COVID- 19 pandemic. Instead, they engaged with their enrollees through telephone or at shelters while wearing masks and social distancing. San Benito San Joaquin shifted their focus to populations who were at highest risk for COVID-19; they placed emphasis on providing education about and support around COVID-19 when engaging enrollees. San Joaquin Care coordination policies or processes (e.g., frequency, modality, location in which provided) Alameda experienced an increased willingness from partners to share data, along with increased access to remote trainings, because of the pandemic. Their consumer experience team also noted new opportunities in community building structures for the homeless isolation hotels. Alameda Whole Person Care Final Evaluation Report | WPC and COVID-19 December 2022 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program to shelter individuals who tested positive for COVID-19. WPC service integration teams conducted telephone screenings of all individuals in the hotels for enrollment into WPC, if eligible. These efforts occurred in addition to continued response to community- based referrals, warm hand-offs from program partners, and referrals from 2-1-1. Process/Policy/Procedure | Examples Pilot Ventura expanded medication-assisted treatment (MAT) to hotel_| Ventura sites for high-risk individuals experiencing homelessness, and enhanced coordination between WPC staff and MAT providers. Enrollment of eligible Alameda worked to directly enroll eligible enrollees on-site at Alameda beneficiaries in WPC COVID-19 isolation hotels. San Diego obtained approval from their Health and Human San Diego Services Agency Compliance Office for contractors to allow verbal consent for the enrollment and creation of digital records in ConnectWellSD for enrollees. Identifying beneficiaries Mendocino expanded their target population criteria to include Mendocino eligible for WPC those at risk for or who tested positive for COVID-19. San Diego contracted with local hotels through Project Roomkey San Diego Source: PY 5 COVID-19 Impact Survey (n=25), April 2020. Keay \WPC and COVID-19 | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 Anew COVID-19 target population was added by DHCS to WPC starting in the third quarter of 2020, and Pilots could retroactively report enrollees in this target population starting at the beginning of 2020. The new target population was designed to include “those at risk of contracting COVID-19, those who have contracted COVID-19, and those recovering from COVID- 19.” Only nine out of the 25 Pilots elected to report individuals in this target population (Exhibit 95). Three Pilots (San Francisco, Solano, and Small Counties) used the broadest definition and assigned nearly all of their new enrollees to this target population. COVID-19 Target Population Exhibit 95: WPC Pilots Reporting Enrollees in COVID-19 Target Population Month Starting to Total Number of Proportion of New Enrollees Since WPC Pilot Report COVID-19 Target | Enrollees in COVID-19 July 2020 Assigned to COVID-19 Population Target Population Target Population Alameda March 2020 18,582 46% Kings July 2020 12 1% Riverside January 2021 97 1% San Francisco January 2020 16,717 99% San Joaquin July 2020 468 21% Santa Clara January 2020 3,395 50% Santa Cruz September 2020 25 49% SCWPCC January 2020 80 100% Solano July 2020 61 100% Source: UCLA analysis of WPC Quarterly Enrollment Utilization Reports from January 2020 to December 2021. Note: Enrollees could be assigned to more than one target population. Impact of the COVID-19 Pandemic on WPC Enrollment Exhibit 96 illustrates the trends in monthly enrollment and the total new enrollment per quarter during WPC, including the pandemic. Monthly enrollment in WPC continued to grow throughout 2020, increasing from 76,015 in December 2019 to 95,866 in December 2020. There was a small increase to 96,416 in December 2021 or the end of WPC. Total new enrollment in the last two quarters of 2020 was lower than it had been in the same quarters in 2019. As the program came to an end during 2021, quarterly new enrollment was also lower compared to the same quarters during any other year of the program. There was a 16% decline in average monthly disenrollment in months during the pandemic (March 2020-December 2021) compared to 2019 (data not shown). Whole Person Care Final Evaluation Report | WPC and COVID-19 err UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 96: Monthly Enrollment and Total Quarterly New Enrollment in WPC, January 2017 to December 2021 COVID-19 Pandemic 95,886 96,416 mmm Total New Enrollment per Quarter | | 76,015 Monthly Enrollment ¥ 61,427 38,750 1 ° is N od s. 1 iN oO 8 §& 83 8 € 8 & i SF es fs BK fe c $8 * ds fF eg ao 8 I F tt Gamaeo ¥ 4 sos ¢ a 7 9 a 86 = = a S bititlibithtiripiia Sse e Fess FP eFsRessraselsesrgssszsr Be Qi Q2 Q3 a4 Qi Q2 Q3 Q4 Qi Q2 Q3 Qa Q Q2 Q3 a4 ai Q2 Q3 a4 2017 2018 2019 2020 2021 Source: UCLA analyses of WPC Quarterly Enrollment and Utilization Reports from January 2017 to December 2021 Notes: 23 of 25 pilots started enrolling throughout 2017, and two pilots started enrolling in early 2018. Characteristics of WPC Enrollees before and after the COVID-19 Pandemic Exhibit 97 shows the characteristics of WPCs enrollees prior to the start of the pandemic (January 2017 to February 2020) and during the pandemic (March 2020 to December 2021). Compared to before the pandemic, WPC enrollees that enrolled during the pandemic were more often younger (less than 34 years old) and less often white or black. They were also less likely to be high users of acute care services and have three or more chronic conditions. YZ) WPC and COVID-19 | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research . De 1 \gPAey ed Health Economics and Evaluation Research Program ba Exhibit 97: Characteristics of WPC Enrollees at Baseline Enrolled Before and During the COVID- 19 Pandemic Before Pandemic During Pandemic <18 1% 5% 18-34 31% 34% Age at Enrollment (Years) 35-49 28% 26% 50-64 33% 26% 65+ 7% 9% Gender Male 56% 55% White 28% 21% Hispanic 26% 32% Black 25% 21% Race/Ethnicity Asian 12 eth American Indian/Alaska Native 4% 7% Hawaiian and Other Pacific Islander 2% 2% Other 9% 11% Unknown 7% 5% At-Risk 24% 33% Acute Care Low 34% 34% Utilization during | Medium 25% 20% Baseline High 11% 8% Super 7% 5% Count of Chronic |0 35% 43% Conditions at 1-2 36% 34% Baseline 34 29% 22% Source: UCLA analysis of Medi-Cal enrollment and claims data, January 2015 to December 2021 Notes: Before pandemic is January 2017 to February 2020 and during pandemic is March 2020 to December 2021. Baseline is the two years prior to WPC enrollment. Chronic conditions are based on Chronic Condition Warehouse definitions. At risk for high utilization is defined as no ED utilization or hospitalizations 24 months prior to enrollment, low utilization is less than 2 ED visits and less than 1 hospitalizations per year, moderate utilization is 2 or more ED visits or 1 or more hospitalizations per year, high utilization is 5 or more ED visits or 2 or more hospitalizations per year, and super utilization is 10 or more ED visits or 4 or more hospitalizations per year. Whole Person Care Final Evaluation Report | WPC and COVID-19 alr) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Estimated Prevalence of COVID-19 among WPC Enrollees The diagnosis code for COVID-19 was developed and utilized by providers starting in late March 2020. To estimate the likely prevalence of COVID-19 among WPC enrollees and the control group, UCLA analyzed Medi-Cal claims starting in April 2020 and identified individuals with services for which COVID-19 was the primary or secondary diagnosis. Overall, 10% of enrollees and 8% of controls used a service with a COVID-19 diagnosis (data not shown). The rate of COVID-19 diagnosis per 1,000 Medi-Cal member months for enrollees and controls by month is shown in Exhibit 98. Rates peaked during the same months that cases peaked statewide and trends were similar among WPC enrollees and controls. Exhibit 98: Rate of COVID Diagnosis per 1,000 Medi-Cal Member-Months for WPC Enrollees and their Controls from April 2020 to December 2021 30 25 ull ih il I & LS of PPP COE SE SES EEE SL ye s e we ss S ca SF Pg © e we SF SK FT MW w ° mEnrollees mControls Source: UCLA analysis of Medi-Cal claims data from April 2020 to December 2021. Notes: COVID-19 diagnosis was identified using ICD code U07.1 in primary or secondary diagnosis per claim. sisi \WPC and COVID-19 | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program COVID-19-Related Health Service Use of WPC Enrollees UCLA examined the types of health services for COVID-19-related care utilized by WPC enrollees and their controls with a COVID-19 diagnosis from April 2020 to December 2021. Enrollees and controls had similar used of COVID-19-related services. They most frequently used hospitalizations (25% and 24%, respectively), followed by primary care services (18% and 21%), emergency department visits (17% and 14%), stays in long-term care facilities (11% and 10%), lab tests (8% and 8%), and specialty services (7% and 7%; Exhibit 99). Exhibit 99: Proportion of COVID-19-Related Health Services by Service Type among WPC Enrollees and their Controls with a COVID-19 Diagnosis 25% Primary Care Services 18% mEnrollees ™ Controls 11% 8% 8% 7% Source: UCLA analysis of Medi-Cal claims data from April 2020 to December 2021. Notes: COVID-19 diagnosis was identified using ICD code U07.1 in primary or secondary diagnosis per claim. 21% Lab Tests Whole Person Care Final Evaluation Report | WPC and COVID-19 entree 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Changes in Healthcare Utilization from COVID-19 UCLA assessed service utilization patterns among WPC enrollees and their controls before and during the pandemic, and found similar patterns for both groups. In particular, both enrollees and their controls had a decline in April 2020 compared to April 2019 for primary and specialty care (Exhibit 100). By December 2020, however, rates of primary care and specialty service utilization were similar to those in December 2019. There is a known delay in Medi-Cal claims and encounter reporting, with some reporting of claims and encounters taking more than six months. These delays likely explain why rates declined at the end of 2021 for both enrollees and controls. Exhibit 100: Monthly Utilization of Primary Care and Specialty Care Services per 1,000 Member Months among WPC Enrollees and their Controls, 2019 Compared to 2020 and 2021 WPC Enrollees Controls 400 400 350 350 PrimaryCare 300 Services 300 Pandemic Start 250 250 Pandemic Start 200 200 250 250 200 200 Specialty Care Services 150 Parjdemic Start 100 100 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec === 2019 Pre-Pandemic === 2020 Pre-Pandemic === 2020 Pandemic === 2021 Pandemic Source: UCLA analysis of Medi-Cal claims data from January 2019 to December 2021. Notes: Member-months were based on Medi-Cal enrollment. KWi3y \WPC and COVID-19 | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly Health Economics and Evaluation Research Program In contrast to primary care and specialty care, the number of both ED visits and hospitalizations declined in April 2020 relative to April 2019, and the utilization maintained at lower levels throughout the remaining months of 2020 and all of 2021 (Exhibit 101). Exhibit 101: Monthly Utilization of Emergency Department Visits and Hospitalizations per 1,000 Member Months among WPC Enrollees and their Controls, 2019 Compared to 2020 and 2021 WPC Enrollees Controls 200 200 i Oe Sof pandemic start =“ 150 150 Pandemic Emergency Start Department ae visits 100 100 50 50 50 50 40 t= Ser | ax A Hospitalizati 7 lospitalizations. 30 | p ° Pandemic Start *° ger wee eee eet ee. 20 20 ——_—_—— 10 10 Pandemic Start ° ° Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec ——= 2019 Pre-Pandemic_ ——— 2020 Pre-Pandemic 2020 Pandemic 2021 Pandemic Source: UCLA analysis of Medi-Cal claims data from January 2019 to December 2021. Notes: Member-months were based on Medi-Cal enrollment. Whole Person Care Final Evaluation Report | WPC and COVID-19 Reaper 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Further analyses found that fewer than 0.1% of primary care and specialty services were delivered by telehealth prior to the pandemic (Exhibit 102). Starting in the second quarter of 2020, between 11% and 18% of primary care services for WPC enrollees were provided through telehealth. The proportion of specialty care services that were provided through telehealth were slightly lower, between 8% and 11%. Overall, controls had similar trends with only slightly higher rates of primary care telehealth services compared to enrollees (data not shown). Exhibit 102: Proportion of Primary Care and Specialty Services that were Provided through Telehealth for WPC Enrollees, 2019 to 2021 x x Ss mt xs Ps a Pd S 0 & 8 i ; | | a a a | 2019 2020 2020 2020 2020 2021 2021 2021 2021 2019 2020 2020 2020 2020 2021 2021 2021 2021 Qi a2 Q3 Q4 Qi Q2 Q3 a4 Ql a2 03 Q4 Ql Q2 Q3 a4 18% & & a 15% 16% 14 12% 11% 11% 9%, <0.1% <0.1% Primary Care Services Specialty Care Services Source: UCLA analysis of Medi-Cal claims data from January 2019 to December 2021. Challenges, Successes, and Lessons Learned Related to COVID-19 The COVID-19 pandemic impacted WPC system capacity and access to health care. Exhibit 103 highlights the most frequently identified challenges and successes related to COVID-19 by reporting period as highlighted in bi-annual narrative reports. Across all themes in both challenges and successes, there was an increase in mentions in PY 5 annual, with a decrease in the PY 6 reporting period. This can likely be explained by Pilots’ adaptation to the ongoing pandemic and establishment of routinized workflows to accommodate for increases in telehealth and social distancing. The most frequently reported challenges were related to the transition to telehealth and Pilots’ inability to provide WPC services in-person (e.g., enrollees often did not have access to the appropriate technology to support telehealth or to engage with WPC staff remotely; 52 mentions across 21 unique LEs); limited staff capacity due to reassignment of WPC staff employed by county agencies to support broader community COVID-19 emergency response, KE:{oy WPC and COVID-19 | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program county-wide hiring freezes (48 mentions across 21 unique LEs); and/or inability to connect enrollees to services (e.g., due to facility closures or reduced provider capacity; 40 mentions across 18 unique LEs). Some Pilots noted that relationships with WPC partners and with enrollees were hindered by the remote work environment, which in turn negatively impacted enrollee engagement. Just over one half of Pilots cited increased service demand coupled with limited funding or resource availability as a challenge. Despite these challenges, many Pilots continued to report successes in WPC, often by integrating WPC activities with COVID-19 response efforts. For example, in some Pilots, COVID- 19 emergency housing projects expanded short-term housing availability for WPC enrollees and facilitated care coordination through co-located medical, behavioral, and social services. Through programs such as Project Roomkey, Pilots were able to consistently locate and engage WPC enrollees (44 mentions across 21 unique LEs). In PY 6 annual narrative reports, many Pilots also reported collaborative efforts to transition short-term emergency COVID-19 housing projects to long-term supportive housing programs. Furthermore, infrastructure previously established through WPC facilitated counties' response to the COVID-19 pandemic for their populations of focus. Pilots leveraged existing WPC partnerships and provider networks (e.g., there was a deepened level of cross-departmental collaboration in emergency operations structures) and utilized WPC-developed data systems and information technology (e.g., COVID-19 risk-based algorithms to provide focused outreach). Additionally, many Pilots adapted internally and/or expanded partner collaborations to provide pandemic-related services like vaccination, testing, education, personal hygiene pods, equity- driven outreach efforts, and increased telephonic check-ins (36 mentions across 20 LEs). Whole Person Care Final Evaluation Report | WPC and COVID-19 err UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 103: Commonly Identified Challenges and Successes Related to the COVID-19 Pandemic among WPC Pilots, PY 5—PY 6 Transition to telehealth and/or lack of in-person service —— 20 isi 14 provision = 8 i. Staff reassignment, hiring freezes, limited staff bandwidth - v7 & 1 2 = 3 7 é Drain a 5 7 9 Limited funding and/or resource availability . 3 9 Expanded short-term housing or shelter availability - ay 6 a Partnerships facilitated COVID-19 response = 15 g 3 8 3 1 Outreach and engagement ae 18 8 Data systems and IT assisted COVID-19 response a 2 MPYSMY MPYSAnnual MPY6MY PY 6 Annual Sources: PY 5 Mid-Year, PY 5 Annual (n=25), PY 6 Mid-Year, and PY 6 Annual Narrative Reports (n=23). Notes: Program Year 6 did not include reports for Small County Collaborative and Solano, as they discontinued WPC participation in PY 6. “MY” denotes mid-year report. WPC and COVID-19 | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research nee lpr) Health Economics and Evaluation Research Program Chapter 9: Enrollee Demographics, Health Status, and Prior Health Care Utilization WPC Pilots were required to “receive support to integrate care for a particularly vulnerable group of Medi-Cal beneficiaries who have been identified as high users of multiple systems and continue to have poor health outcomes.” This chapter addresses the following evaluation question: “What were the demographics of pilot enrollees?” In addition, UCLA examined the health status of enrollees and their utilization of services prior to enrollment in WPC. Whenever possible, this information is provided for the overall enrollee population and by target population. The data sources included Medi-Cal enrollment and claims data between January 2015 and December 2021 and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6 (2017 through 2021). Of the 247,887 total WPC enrollees during program implementation, 235,547 enrollees had Medi-Cal eligibility data and 233,332 of these enrollees had claims data, which allowed for assessment of their health status and health care use. UCLA included these enrollees when reporting on health status and health care utilization prior to enrollment for WPC overall. Assessment of demographics, health status, and health care use by target population can be found in Appendix T, which includes 228,680 enrollees that had an assigned target population and Medi-Cal data. The prevalence of chronic conditions was identified using the CMS Chronic Conditions Data Warehouse for WPC enrollees with Medi-Cal claims data, using the primary and secondary diagnosis at each encounter. UCLA calculated standardized rates of utilization to account for variations in length of enrollment in Medi-Cal and to facilitate comparisons across analytic groups. Utilization was calculated per 1,000 full-scope Medi-Cal member months for six-month intervals in the two years prior to an enrollees’ first WPC enrollment date. Age was time-variant and was identified at the time of WPC enrollment. Time-invariant demographics such as race/ethnicity were identified using the most frequently reported value in enrollment data during the 24 months prior to enrollment into the program. Health status was measured as the presence of a condition at any point within 24 months prior to enrollment. For additional detail on data sources and methodology please see Appendix A. Whole Person Care Final Evaluation Report | Enrollee Demographics, Health Status, and Prior Health BRE) Care Utilization ee 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Demographics Medi-Cal enrollment data indicated that over 90% of WPC enrollees were between the ages of 18 and 64, including a greater concentration of those who were 18-34 (32%) and 50-64 (31%) years old compared to 35-49 (28%; Exhibit 104). Enrollees were more often male (56%), Hispanic (28%), or preferred English as their primary communication language (86%). Half (51%) of enrollees experienced homelessness. Examining these characteristics by target population indicated differences (see Appendix T). For example, justice-involved enrollees were most frequently ages 18-34, were male, used English as their primary communication language, and experienced homelessness prior to WPC enrollment. Those in the homeless target population were most often ages 50-64 and either white or black. Exhibit 104: Demographics of WPC Enrollees Prior to WPC Enrollment 2% Age at enrollment 32% 28% 31% £579) 0-17 18-34 m35-49 m50-64 m65+ = Male _& Female 1% 2% Race/Ethnicity Ply 3 28% ply Ey 10% 7% @ White Hispanic Black BAsia @ American Indian or Alaskan Native Hawaiian or Other Pacific Islander @ Other @ Unknown Primary 86% pt Ce Communication... English _@Spanish_ _™ Other Homelessness 51% Cg m Experienced Homelessness @ Did Not Experience Homelessness Source: Medi-Cal enrollment data from January 2015 to December 2021 and Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Overall enrollee population includes 235,547 enrollees who were enrolled during PY 2 through PY 6 and had Medi-Cal enrollment data. All data except for homelessness are reported using Medi-Cal enrollment data during the 24 months prior to WPC enrollment. Homelessness was based on a Pilot-reported indicator collected at enrollment. E20 Enrollee Demographics, Health Status, and Prior Health Care Utilization | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research nee lpr) Health Economics and Evaluation Research Program Among all WPC enrollees, depression was the most common chronic condition (37%), followed by depressive disorders (34%), anxiety disorders (33%), hypertension (33%), and drug use disorders (32%; Exhibit 105). Other common conditions included schizophrenia and psychotic disorders (26%), bipolar disorder (22%), tobacco use (22%), and alcohol use disorders (21%). Health Status Exhibit 105: Most Frequent Chronic Conditions Among WPC Enrollees, 24 Months Prior to WPC Enrollment HY) DcrleNSi0) L———iiiiiHirCrH Tm” 337, Diabetes LT 17% Hyperlipidenia I 17%, Rheumatoid arthritis/ osteoarthritis LL 17% Chronic Kidney Disease LN 16% Ancnia i 16% Chronic Obstructive Pulmonary Disease A 14% Physical Health Conditions Atha i 14% Depressi 37%, Depressive Disorder ET 34°, Anxiety Disorder Es 33%, Mental Health Conditions Schizophrenia and Other Psychotic Disorders ET 26% Bipolar Disorder TT 22% Drug Use Disorders EE 32% Tobacco Use A, 22% Substance Use Conditions Alcohol Use Disorders EE 21% Source: Medi-Cal enrollment and claims data from January 2015 to December 2021 and Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Enrollee population includes 233,332 enrollees who were enrolled during PY 2 through PY 6 and had Medi- Cal enrollment and claims data. Chronic and disabling conditions were determined using algorithms developed by the CMS Chronic Conditions Data Warehouse (CCW). Conditions with at least 10% prevalence were reported. Whole Person Care Final Evaluation Report | Enrollee Demographics, Health Status, and Prior Health [iBE:}s) Care Utilization UCLA Center for Health Policy Research 1 \1arseyPd . . Health Economics and Evaluation Research Program Utilization Prior to Enrollment Selected Outpatient Service Use Prior to Enrollment Medi-Cal claims data indicated WPC enrollees received 273 primary care services per 1,000 Medi-Cal member months from 1-6 months prior to their WPC enrollment, an increase from 229 from 19-24 months prior to WPC enrollment (Exhibit 106). Specialty services also increased from 123 to 163 from 19-24 months to 1-6 months prior to enrollment. The rates of mental health and substance use disorder services were higher and also increased during this time period as well. Exhibit 106: Selected Ambulatory Care Service Use per 1,000 Medi-Cal Months Among WPC Enrollees in Months Prior to WPC Enrollment 19-24 month 229 13-18 months a 244 712th a 256 16 nonth a 273 19-24 month "123 13-18 months i! 134 7-12 months a 143 1-6 month 163 19-24 So TT S37 Qa. A 565 FE 517 6 771 AD —E>—_—_—_—_—_—_—_—_>_>>_—__———————===# 611 13-18 0nthS LT C2? 2-22 1h 533 16 thos 656 Services Services Services Mental Health Specialty Care Primary Care SUD Services Source: Medi-Cal enrollment and claims data from January 2015 to December 2021 and Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Note: Enrollee population includes 233,332 enrollees who were enrolled during PY 2 through PY 6 and had Medi- Cal enrollment and claims data. HEIN Enrollee Demographics, Health Status, and Prior Health Care Utilization | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research nee lpr) Health Economics and Evaluation Research Program Emergency Department Visits Prior to Enrollment Medi-Cal claims data showed that the rate of overall ED visits followed by discharge per 1,000 Medi-Cal member months increased 19-24 months to 1-6 months before WPC enrollment, from 162 to 212 (Exhibit 107). Examining ED visit rates by condition also showed increasing rates before WPC enrollment for all conditions examined. ED visits with a primary or secondary diagnosis of a mental health condition were most common at 65 visits per 1,000 Medi-Cal member months in 1-6 months prior to WPC enrollment, while ED visit rates for substance use disorder, diabetes, and hypertension in the same time period were 42, 10 and 14, respectively. Exhibit 107: Emergency Department (ED) Visits Followed by Discharge per 1,000 Medi-Cal Member Months Among WPC Enrollees in Months Prior to WPC Enrollment, Overall and by Specific Conditions GE OT 1b 7 13-18 Months LS 17) e 3 OT EF 125 1-6 1 onthS Ls 271 19-24 months ml 25 33 13-18 months MZ 30 Ze 7-12 months Hi 35 1-6 months a 42 19-24 months 40 13-18 months EA 47 7-12 months A 54 Any Mental Health Diagnosis 1-6nonth a 65 19-24months Ml 7 13-18 months Ml 8 7-12 months mi 9 Diabetes Diagnosis 1-6months Ml 10 19-24 months Ml 10 13-18 months Ml 11 7-12 months Mall 13 Hypertension Diagnosis 1-6months Mm 14 Source: Medi-Cal enrollment and claims data from January 2015 to December 2021, Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: “Overall” includes 271,227 individuals identified as enrolled during PY 2 through PY 6 and with sufficient Medi-Cal enrollment and claims data. Conditions were based on the related primary or secondary diagnoses at the time of visit. SUD is substance use disorder. Whole Person Care Final Evaluation Report | Enrollee Demographics, Health Status, and Prior Health Care Utilization Serr) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Hospitalization Prior to Enrollment Medi-Cal claims data showed that the rate of overall hospitalizations per 1,000 Medi-Cal member months increased before WPC enrollment, from 32 to 52 (Exhibit 108). Examining hospitalization rates by condition also showed increasing rates before WPC enrollment for all conditions examined. Hospitalizations with a primary or secondary diagnosis of a mental health condition were most common at 19 stays per 1,000 Medi-Cal member months in 1-6 months prior to WPC enrollment. Exhibit 108: Number of Hospitalization per 1,000 Medi-Cal Member Months Among WPC Enrollees in Months Prior to WPC Enrollment, Overall and by Specific Conditions 111° ——————————————s! 37 BS 13-18 ths TT 35 é 7212 So 4] 16 cthS LT 52 19-24 months i! 4 3 3 13-18 months 5 Zs 7-12months mam 5 < e 16months —7 = , 19:24months A 11 5s 3 1318 months 13 28 & 7-12 months A 15 €& a 1-6 noth a 19 19-24 months mm 2 13-18 months Mm 2 7-12 months mmm 3 1-6months mmm 3 19-24 months mm 2 13-18 months mm 2 7-12 months mam 3 1-6months mmm 4 Diabetes Diagnosis Hypertension Diagnosis Source: Medi-Cal enrollment and claims data from January 2015 to December 2021, Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: “Overall” includes 271,227 individuals identified as enrolled during PY 2 through PY 6 and with sufficient Medi-Cal enrollment and claims data. Diagnosis was based on the primary or secondary diagnosis of stay. SUD is substance use disorder. HE:E3) Enrollee Demographics, Health Status, and Prior Health Care Utilization | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ely) Health Economics and Evaluation Research Program ie Chapter 10: Better Care WPC Pilots aimed to increase “appropriate access to care for the most vulnerable Medi-Cal beneficiaries.” This chapter addresses the following evaluation question: “To what extent did the Pilots increase appropriate access to care and improve beneficiary care outcomes?” Data sources for this chapter included Quarterly Enrollment and Utilization Reports from PY 2 to PY 6 and Medi-Cal enrollment and claims data. UCLA used the Quarterly Enrollment and Utilization Reports to identify enrollees and dates of enrollment. UCLA also used Medi-Cal claims data, which included both managed care and fee-for-service encounters, to construct WPC metrics per the WPC Technical Specifications to create two universal metrics (Follow-Up After Hospitalization for Mental Illness and Initiation and Engagement of Alcohol and Other Drug Dependence Treatment). In addition, UCLA measured the utilization rates of outpatient services (primary care, specialty care, mental health and substance use disorder services) to further examine how access to care was impacted by WPC. UCLA measured trends before and during WPC for each metric based on the date of an individual WPC enrollee’s enrollment. UCLA examined changes in trends before and during WPC using a difference-in-difference (DD) analysis by modeling the changes in yearly increments up to 2 years (Pre-Year 1 and Pre-Year 2) before WPC enrollment and up to 5 years (Year 1, 2, 3, 4, and 5) during WPC. For these, the DD analysis measured the trends or change in yearly rates from Pre-Year 2 vs. Pre-Year 1 for both WPC enrollees and the control group; the change in the yearly rate during WPC from Year 1 to Year 5 for both WPC enrollees and the control group; and the difference between the changes in WPC enrollees vs. the control group from before to during WPC. These estimates were adjusted for beneficiary demographics as well as health status and use of services pre-WPC. Further details can be found in Appendix A. To better understand WPC outcomes, UCLA examined the program impact on enrollees with serious mental illness (SMI), substance use disorders (SUD), or experiencing homelessness (SMI/SUD/HML enrollees) compared to enrollees without these complicating conditions. The latter group was composed of enrollees who were medically complex including those with multiple chronic conditions and those at high risk for various reasons (MC/HR enrollees). UCLA used the Annual WPC Variant and Universal Metric Reports submitted by Pilots to DHCS from baseline to PY 6 to report on one universal (2.5 - Comprehensive Care Plan) and one variant (3.1.7 - Major Depressive Disorder Suicide Risk Assessment) metric, calculated by Pilots based on electronic medical records or chart review and therefore not replicable by UCLA. Pilot- reported metrics on follow-up after hospitalization for mental illness and initiation and engagement of alcohol and other drug dependence treatment were not included in this report Whole Person Care Final Evaluation Report | Better Care [BEE) eee 6 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program because they were found to be heavily dependent on data sharing agreements and data sharing capacity during the first three years of WPC and were therefore incomplete. UCLA reported a weighted average rate for the available metrics across all Pilots that reported each metric. For additional detail on data sources and methodology please see Appendices A and B. Utilization of Outpatient Services UCLA created four measures of health care utilization and examined the trends on an annual basis. These measures were not required by WPC as performance metrics and did not have ana priori intended or desired direction. UCLA used these measures to illustrate potential changes in delivery of care under WPC. Primary Care Services UCLA calculated the number of primary care services per 1,000 beneficiaries per year to show patterns of change in primary care service use. Primary care services are likely to increase to address unmet need but also to decline as unmet needs are addressed or other appropriate services are used. Therefore, the anticipated direction of this measure and DD is decrease. nile Better Care | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Exhibit 109 shows an increase of 727 and 668 primary care services per 1,000 beneficiaries per year for WPC enrollees and the control group before WPC, respectively. After an increase in utilization of primary care services in the first year of WPC for WPC enrollees, this rate decreases during WPC by 208 services per year for WPC enrollees and increases by 63 services per year for controls. The decline from before to during WPC was significantly greater for WPC enrollees than the control group by 330 services (DD). The declining rate from before to during WPC for enrollees compared to their controls was found for both SMI/SUD/HML enrollees (- 255) and for MC/HR enrollees (-535; data not shown). These data showed a greater decline among MC/HR enrollees than the SMI/SUD/HML group. Whole Person Care Final Evaluation Report | Better Care Ea Reaper 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 109: Trends in Primary Care Services per 1,000 Beneficiaries per Year Before and During WPC, PY 2- PY6 t 8 5 a. 3 a Q a a Nn Xx 8 A Re gS a BR RR & nt oy ot om Uh Lt tS oo aa o # ” 2 2 ” a oS | | Pre-Year 2 Pre-Year 1 Year 1 Year 2 Year 3 Year 4 Year S Before WPC During WPC ™ WPCEnrollees ™ Control Group Peta Pye Tat) pierre} Cee (DD) WPC Enrollees 727* -208* -935* Control Group 668* 63* -605* -330* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes p<0.05, a statistically significant difference. Primary care services were identified as services with a primary care physician, physician assistant, or nurse practitioner per NUCC’s Taxonomy code set, and services provided by a Federally Qualified Health Center (FQHC). Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in- difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). NTA 4-} Nae Clemo TTT ae Better Care | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ee err) Health Economics and Evaluation Research Program UCLA calculated the number of specialty care services per 1,000 beneficiaries per year to show Specialty Care Services patterns of change for specialty service use. Specialty care utilization may have increased due to care coordination efforts by Pilots. Therefore, the anticipated direction of the measure and DD is increase. Exhibit 110 shows an increase of 343 more specialty care services before WPC per 1,000 beneficiaries per year and a slower rate or an increase of 131 more services per year during WPC for WPC enrollees. While a similar pattern was observed for the control group, the decline in the rate from before and during WPC was significantly smaller for WPC enrollees vs. controls by 133 services (DD). A similar increasing rate from before to during WPC for enrollees compared to their controls was found for both SMI/SUD/HML enrollees (133 services) and for MC/RR enrollees (132 services; data not shown). Exhibit 110: Trends in Specialty Services per 1,000 Beneficiaries Months Before and During WPC, PY 2-PY6 ~ a = ed g 8 a B @ & Ro Re 3 Ba a 8 9 oN a 8 a og a a o © a a ee a a a 2 cy ge 8 L Pre-Year 2 Pre-Year 1 Year1 Year 2 Year 3 Year 4 Year S Before WPC During WPC @ WPCEnrollees ™ Control Group NTL 1 NTS <9 Pye} PhaCicese ig Ta fole mae TTA ae Between Changes __ Difference (DD) WPC Enrollees 343* 131* -212* Control Group 439* 94* -345* 133* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes ps0.05, a statistically significant difference. Specialty care services were identified as services with a specialty physician, physician assistant, or nurse practitioner per NUCC’s Taxonomy code set. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC — Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). Whole Person Care Final Evaluation Report | Better Care [BIE Reaper 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Mental Health Services UCLA calculated the number of mental health services per 1,000 beneficiaries per year as a measure of mental health service use. Mental health services are likely to increase to address unmet need as a result of care coordination but also to decline as patients are better managed. Therefore, the anticipated direction of this measure and DD is decrease. Exhibit 111 shows that WPC enrollees’ mental health service use was increasing prior to enrollment by 1,566 services per 1,000 beneficiaries per year, but it declined by 957 per year during WPC after initially increasing in the first year of the program. The pattern for the control group was somewhat similar but WPC enrollees did have a significantly greater decline from before to during WPC compared to the control group (-813 services, DD) and the control group’s mental health use did not increase in the first year of the program. Exhibit 111: Trends in Mental Health Services per 1,000 Beneficiaries per Year Before and During WPC, PY 2 - PY6 s 3 a 0 8 ea & a s a oo a S + 0 > oO 5 x 9 Co) > 8 & 2 x a ow g ~ ey e 2 a + wy n a a oa + o oO a oo aL R = « 5 R [ U 7 Pre-Year 2 Pre-Year 1 Year 1 Year 2 Year 3 Year 4 Year S Before WPC During WPC @ WPCEnrollees ™ Control Group Ala aera -i-) baer la) 4-9 Difference Difference-in- Before WPC DTT a Le Between Changes _ Difference (DD) WPC Enrollees 1,566 * -957* -2,523* Control Group 1,050* -661* -1,710* -813* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes ps0.05, a statistically significant difference. Mental health services were identified as services with a mental health procedure code. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). uke) Better Care | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ee err) Health Economics and Evaluation Research Program The declining rates from before to during WPC among WPC enrollees compared to their controls was restricted to SMI/SUD/HML enrollees (-1,125 services; Exhibit 112). For MC/HR enrollees, there was a significant increase in utilization of mental health services in the first year of WPC compared to the year prior to enrollment (increase from 848 to 2,508 services per 1,000 beneficiaries per year). Compared to controls, these enrollees had a slightly increasing rate compared to controls (43 services). Exhibit 112: Trends in Mental Health Services per 1,000 Beneficiaries per Year Before and During WPC, PY 2 - PY 6, by Subpopulations © y B g a 2 a + K a & 5 8 8 ¢ Ss = “ Ss q & 2 S a = ON a Le oo o 2 + ¢ 8 6 NS a 2 nw nw th o g 2 i io) ° st ¢ a © a +t Ss Bt Go Ran Sa o xe Q % ty dM ID COON ” X we en NT WGN AN NX NN av Sa aa a a a a a ( j ee | ee ee i nc 4 4 N ” s in N a 4 nN ” + in 5 S S G 5 G 5 S 5 G 5 G S G a o o o o o o o o a a a a a > > > > > > > > > > > > > > 2 2 2 2 a a a a Before WPC During WPC Before WPC During WPC Medically Complex or High Risk SMI/SUD or Experiencing Homelessness mWPCEnrollees Control Group Sle] Nr Ta ig Difference Liat Change Change rina 4 LTS aT) f Toe) g-) pyaar 4 chances yess} Wiss Wiss 8 (DD) Medically Complex or High_| WPC Enrollees | 162* -66* -228* | Risk Control Group | 216* -55* -271* 43* SMI/SUD or Experiencing | WPC Enrollees | 2,077* -1,281* -3,358* Homelessness Control Group | 1,352* -881* -2,233* | -1,125* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes ps0.05, a statistically significant difference. Mental health services were identified as services with a mental health procedure code. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). SMI/SUD is serious mental illness or substance use disorder. Whole Person Care Final Evaluation Report | Better Care [RIES Reaper 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Substance Use Disorder Services UCLA calculated the number of substance use disorder (SUD) treatment services per 1,000 beneficiaries per year. Substance use services are likely to increase to address unmet need and continuous assessment. Therefore, the anticipated direction of this measure and DD is increase. Exhibit 113 shows that trends in SUD treatment service use were increasing prior to enrollment for WPC enrollees by 614 services per 1,000 beneficiaries per year. After an initial increase in the first year of WPC, these rates declined during WPC by 607 services, though overall rates remained high. In contrast, the rate of use of these services was declining for the control group by 758 services per year during WPC. This led to a significant differential between the two groups of 56 more services per 1,000 members per year for WPC enrollees vs. the control group (DD). Exhibit 113: Trends in Substance Use Disorder Services per 1,000 Beneficiaries per Year Before and During WPC, PY 2- PY6 e a g e g 5 mn @ ° 3. 8 8 eo) wo Cs wn o co a uw bail N a wn g 2. w Ww oy 2 ir 2, + < n 9 w st % oo 00 B mS. | I Pre-Year 2 Pre-Year 1 Year1 Year 2 Year 3 Year 4 Year S Before WPC During WPC @ WPCEnrollees ™ Control Group NTL 1 NTS <9 yucca} Difference-in- Ta fole mae TTA ae Between Changes __ Difference (DD) WPC Enrollees 614* -607* -1,221* Control Group 519* -758* -1,277* 56* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes ps0.05, a statistically significant difference. SUD services were identified as services with a SUD treatment procedure code or an NDC for pharmacotherapy. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). silo) Better Care | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ee err) Health Economics and Evaluation Research Program The increasing rates from before to during WPC among WPC enrollees compared to their controls was restricted to MC/HR enrollees (357 services; Exhibit 114). For SMI/SUD/HML enrollees, there was a significant decline compared to controls of 53 services. The MC/HR enrollees saw a significant increase in utilization of substance use disorder services in the first year of WPC compared to the year prior to enrollment (increase from 171 to 1,010 services per 1,000 beneficiaries per year). Exhibit 114: Trends in Substance Use Disorder Services per 1,000 Beneficiaries per Year Before and During WPC, PY 2 - PY 6, by Subpopulations Ss & 2 5 a x = st oO Ra OO a 2 S Ro Om 8 rn 2 oy 4 nS a g q go8 a o So iN. x S a o o ~ ® 0 w Be + z & ° mA 2. yr of we a Nn S oe & Nn % 8 8% SS Ss ge Be BA fa at ad aH ag ge 4 a “ae "eo sf on oe oe Ee nC 4 4 N ” s in N a 4 nN ” + in 5 S S G 5 G 5 S 5 G 5 G S G a o o o o o o o o a a a a a > > > > > > > > > > > > > > 2 2 2 2 a a a a Before WPC During WPC Before WPC During WPC Medically Complex or High Risk SMI/SUD or Experiencing Homelessness mWPCEnrollees Control Group Sle] Nr Ta ig ee LIne Change Change rina 4 LTS aT) f Toe) g-) pyaar 4 chances yess} Wiss Wiss 8 (DD) Medically Complex or High | WPC Enrollees | 17* 129* 113* Risk Control Group | 103* -141* -244* | 357* SMI/SUD or Experiencing WPC Enrollees | 831* -874* -1,705* Homelessness Control Group | 670* -982* -1,652* | -53* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes ps0.05, a statistically significant difference. Mental health services were identified as services with a mental health procedure code. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). SMI/SUD is serious mental illness or substance use disorder. Whole Person Care Final Evaluation Report | Better Care Ea Reaper 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Follow-Up After Hospitalization for Mental Illness Follow-Up After Hospitalization for Mental Illness was a WPC universal metric that measures the percentage of discharges for beneficiaries 6 years of age and older hospitalized for treatment of selected mental illness diagnoses who had a follow-up visit with a mental health practitioner at (1) 7-days or (2) 30-days. The intended direction of the metric and DD is increase. Exhibit 115 shows that the 7-day follow-up rate did not change for both WPC enrollees and controls before WPC. After enrollment, the WPC enrollees continued to have a high rate (59% in Year 5), which did not change per year. However, this rate declined for controls significantly by 1.7% per year. These differences in patterns led to a 2.7% yearly increase in likelihood of 7- day visits for WPC enrollees compared to controls (DD). Exhibit 115: Trends in Follow-Up After Hospitalization for Mental Illness within 7 Days Before and During WPC for WPC Enrollees and the Control group, PY 2 - PY6 x 3s 2 BS a . 2 8 8 § & 72 22 ot 2s €2 v3 v 2 = = S x x = a a | | | | } Pre-Year 2 Pre-Year 1 Year 1 Year 2 Year 3 Year 4 Year S Before WPC During WPC mWPCEnrollees m Control Group Ala Ouray i-) Yearly Change Difference iis eX fele Ne Taya ae Between Changes Difference (DD) WPC Enrollees -0.1% 1.0% 1.1% Control Group -0.1% -1.7%* -1.2% 2.7%* | Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes p<0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). ui:}3) Better Care | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ee err) Health Economics and Evaluation Research Program Exhibit 116 shows trends for 30-day follow-up. Trends were similar to those seen at 7-days except that there were no significant differences in the change in yearly rates between WPC enrollees and controls. The rate of this follow-up per year remained high for WPC enrollees during WPC with 83% having had a 30-day follow-up visit in Year 5. Exhibit 116: Trends in Follow-Up After Hospitalization for Mental Illness within 30 Days Before and During WPC for WPC Enrollees and the Control group, PY 2 - PY6 R x R x Rx Rx ey = = Roo No a SA oS ae x Rg Rg g S x8 e oS S Pre-Year 2 Pre-Year 1 Year1 Year 2 Year 3 Year 4 Year 5 Before WPC During WPC mWPCEnrollees m Control Group Arar Yearly Change Difference yee Cole el TTT Ae Between Changes Difference (DD) WPC Enrollees -0.1% -0.6% -0.5% Control Group -0.1% -3.3%* -3.2%* 2.7% Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes ps0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). Whole Person Care Final Evaluation Report | Better Care [BIEE) Reaper 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Initiation and Engagement of Alcohol and Other Drug Dependence Treatment Initiation of Alcohol and Other Drug (AOD) Dependence Treatment was a WPC universal metric measuring the percentage of adolescent and adult beneficiaries with a new episode of AOD dependence who initiated treatment through an inpatient AOD admission, outpatient visit, intensive outpatient encounter or partial hospitalization within 14 days of the diagnosis. Engagement of AOD Dependence Treatment is a WPC universal metric that measures the percentage of adolescent and adult beneficiaries who initiated treatment and who had two or more additional services with a diagnosis of AOD within 30 days of the visit initiation. The intended direction of this metric and DD is increase. Exhibit 117 shows that the rate of initiation of AOD treatment increased significantly before WPC for WPC enrollees by 1.9% but this rate decline by 1.1% per year during WPC. The same pattern was observed among the control group and the two trends were similar (DD). However, these data showed that WPC enrollees had higher rates of initiation than controls during WPC even when the rates of change were similar. Exhibit 117: Trends in Initiation of Alcohol and Other Drug Dependence Treatment Before and During WPC for WPC Enrollees and the Control Group, PY 2 - PY 6 ss 5 < 2 x X xe x x ” x x x x 5 & a 8 x nN a an in Nn @ *m x x x N N N a N ss | | | | ] Pre-Year 2 Pre-Year 1 Year 1 Year 2 Year 3 Year 4 Year 5 Before WPC During WPC @WPCEnrollees Control Group Yearly Change Se NAO <3 Pye} yao ST fele Mae Yaya ae Between Changes Difference (DD) WPC Enrollees 1.9%* -1.1%* -3.0%* Control Group 1.9%* -1.6%* -2.9%* -0.2% Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes ps0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). vil Better Care | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ee err) Health Economics and Evaluation Research Program Exhibit 118 shows that trends in engagement in AOD treatment following initiation did not change for WPC enrollees either before WPC or during WPC. Comparatively, the rates of engagement for controls declined significantly per year during WPC, resulting in a significant difference between WPC enrollees and the control group by 1.9% (DD). These data also showed that the rate of engagement for WPC enrollees during WPC was as high as 49% for most years compared to lower rates for controls. Exhibit 118: Trends in Engagement of Alcohol and Other Drug Dependence Treatment Before and During WPC for WPC Enrollees and the Control Group, PY 2 - PY 6 s 2 2 s xs e 2 s s R x R x Fx 5 5 g ¢ = 9 = 3 x 8 x | | | | Pre-Year 2 Pre-Year 1 Year 1 Year 2 Year 3 Year 4 YearS Before WPC During WPC ™WPC Enrollees @ Control Group Ata erly) Bete NAL 9 Pies) Difference-i Before WPC TTT aa Between Changes Difference (DD) WPC Enrollees -0.1% -0.1% 0.1% Control Group -0.1% -2.0%* -1.9%* 1.9%" | Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes ps0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). Whole Person Care Final Evaluation Report | Better Care Trends in WPC Pilot-Reported Metrics UCLA Center for Health Policy Research Health Economics and Evaluation Research Program UCLA calculated the weighted average values for one universal and one variant metric using Pilot-reported data (Exhibit 119). Some Pilots did not report planned metrics every year for reasons such as no enrollment or program activities during the reporting time period or lack of data in that time period. See Appendix B for further details on reporting for each metric, including which Pilots reported on each metric during each measurement year. Exhibit 119: Pilot-Reported Universal and Variant Metrics That Indicate Better Care Universal | Metric Name Description Baseline | Reporting | Numbers | Improvement vs. and Number Year Years of Pilots Measured by Variant Reporting | Increase or by Year Decrease Universal | 2.5 CCP-E: Percent of enrollees PY2 PY 3, PY 20 in PY 2 | Increase Comprehensive | who received a CCP 4,PY5, Care Plan (CCP) | (accessible by their entire PY6 24 in PY3 care team), within 30 days of enrollment CCP-A: Percent of enrollees | PY3 PY 4, PY 19 in PY3 | Increase who received a CCP 5,PY6 (accessible by their entire care team) within 30 days of the enrollee’s anniversary of enrollment in WPC Variant 3.1.7: Major MDD: Percentage of PY1 PY 2, PY 19 in PY1 | Increase Depressive enrollees aged 18 and older | (2016) 3, PY4, Disorder with a diagnosis of MDD PY5,PY6 | 18inPY2 Suicide Risk with a suicide risk Assessment assessment completed 22 in PY3 (MDD) during the visit in which a new diagnosis or recurrent episode was identified Source: Baseline, PY 2, PY 3, PY 4, PY 5, and PY 6 Annual WPC Variant and Universal Metric Reports and Whole Person Care Universal and Variant Metrics Technical Specifications (March 22, 2019). Better Care | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program All Pilots were required to report on the percent of enrollees who received a comprehensive care plan, accessible by their entire care team, (1) within 30 days of enrollment (CCP-E) and (2) within 30 days of the enrollee’s anniversary of enrollment in WPC (CCP-A). Exhibit 120 shows that the overall CCP-E rate for WPC increased from 12% in PY 2 to 54% in PY 5 before declining slightly to 46% in PY 6. There was substantial variation in CCP-E rates by individual Pilots, ranging from a low of 0% to a high of 100% during most years The rates for CCP-E were influenced by two large Pilots. Without these influential Pilots, the trends remain the same, but annual rates varied from 33% to 86% (data not shown). Comprehensive Care Plan Exhibit 120: Percent of Enrollees Who Received a Comprehensive Care Plan Within 30 Days of Enrollment, by Program Year 100% 100% 100% 100% 100% > * % ° ¢ 54% 46% ‘ 46% 27% 12% 0% ae 0% 0% Baseline PY 3 (n=42,618) PY 4 (n=43,471) PY 5 (n=34,794) PY 6 (n=21,356) (PY 2, n=36,021) Overall WPC —@ Pilot-Specific Minimum @ Pilot-Specific Maximum Source: WPC Annual Universal and Variant Metric Reports, baseline through PY 6 Notes: The comprehensive care plan was to be accessible by the entire care team. Only Pilots that reported on this metric were included in the analysis. The number of Pilots reporting varied by year. The denominator size is shown as sample size per year. Appendix B, Exhibit 13 provides details on which Pilots reported in each year. Bars represent the range reported by Pilots, with minimum being the lowest rate reported by a Pilot and maximum being the highest rate reported by a Pilot. The rate of 0% indicates that no enrollees received a comprehensive care plan within 30 days of enrollment. Whole Person Care Final Evaluation Report | Better Care ery UCLA Center for Health Policy Research Health Economics and Evaluation Research Program CCP-A was reported starting in PY 3 once enrollees had the opportunity to be enrolled for one year. Exhibit 121 shows that CCP-A rates increased from 43% in PY 3 to 72% in PY 6 and were consistently higher than CCP-E rates. Similar to CCP-A, there was large variation in the Pilot- specific rates, ranging from 0% to 100%. One Pilot did not report this universal metric. Exhibit 121: Percent of Enrollees Who Received a Comprehensive Care Plan Within 30 Days of the Anniversary of their Enrollment, by Program Year 100% 100% 100% 100% , 72% 61% 63% a il 7% 14% 0% 0% Baseline (PY 3, n=23,027) PY 4 (n=33,339) PY 5 (n=39,022) PY 6 (n=36,214) Overall WPC — @ Pilot-Specific Minimum ® Pilot-Specific Maximum Source: WPC Annual Universal and Variant Metric Reports, baseline through PY 6 Notes: The comprehensive care plan was to be accessible by the entire care team. Only Pilots that reported on this metric were included in the analysis. The number of Pilots reporting varied by year. The denominator size is shown as sample size per year. Appendix B, Exhibit 14 provides details on which Pilots reported in each year. Bars represent the range reported by Pilots, with minimum being the lowest rate reported by a Pilot and maximum being the highest rate reported by a Pilot. The rate of 0% indicates that no enrollees received a comprehensive care plan within 30 days of enrollment. ve Better Care | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program Major Depressive Disorder: Suicide Risk Assessment A subset of 23 WPC Pilots elected to report the percent of enrollees age 18 or older with a diagnosis of major depressive disorder (MDD) who had a suicide risk assessment completed during the visit in which a new diagnosis or recurrent episode was identified. The overall MDD rate increased from 10% in baseline to 32% in PY 6, with consistent growth from year to year ( Exhibit 122). There was variation in MDD by Pilot, ranging from a low of 0% in all measurement years to a high of 100% in all years apart from baseline. Many Pilots had less than ten enrollees with a diagnosis of major depressive disorder during each measurement year, which led to high variation in this metric. One Pilot with 47% to 68% of all enrollees with a diagnosis of major depressive disorder each year had consistently low rates of 2% or lower. Without this Pilot, the MDD rate increased from 30% to 48% from baseline to PY 3 and then fell to 43% by PY 6 (data not shown). Exhibit 122: Percent of Adult Enrollees with a Diagnosis of Major Depressive Disorder That Received a Suicide Risk Assessment During the Visit in Which a New Diagnosis or Episode was Identified, by Program Year 100% 100% 100% 100% 100% a% + * * % + 32% 28% 23% 17% 21% 0% 0% 0% 0% 0% 0% Baseline PY 2 (n=6,882) PY 3 (n=8,864) PY 4(n=12,026) — PY 5 (n=11,347) PY 6 (n=9,843) (PY 1, n=12,924) Overall WPC = # Pilot-Specific Minimum —_@ Pilot-Specific Maximum Source: WPC Annual Universal and Variant Metric Reports, baseline through PY 6 Notes: Only Pilots that reported on this metric were included in the analysis. The number of Pilots reporting varied by year. Appendix B, Exhibit 9 provides details on which Pilots reported in each year. The denominator size is shown as sample size per year and the overall WPC rate is weighted based on denominator size. Bars represent the range reported by Pilots, with minimum being the lowest rate reported by a Pilot and maximum being the highest rate reported by a Pilot. Whole Person Care Final Evaluation Report | Better Care eee 6 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Pilot Assessment of Challenges to and Impact of WPC on Better Care Pilots reported on challenges to achieving better care, factors that promoted better care, and their overall their perceptions of aspects of care delivery that were impacted by WPC. In PY 6 follow-up interviews and bi-annual narrative reports, Pilots identified a lack of primary care capacity as a barrier to connecting enrollees to primary care. In particular, inability to secure same-day or next-day appointments for enrollees was a challenge. Another challenge that arose during PY 5 was the COVID-19 pandemic, which required providers to shift to telehealth services, particularly for delivery of primary care. WPC Pilots noted that this transition was challenging for many enrollees who often did not have reliable access to the resources needed to participate in telehealth services (e.g., phone, internet). WPC Pilots strove to provide these resources, but were often limited in their capacity to do so. Primary care provided via telehealth also limited the ability of care coordinators to accompany enrollees during their appointments. “The largest challenge faced by CommunityConnect is the lack of capacity within the overburdened safety-net system (housing, primary and specialty care, substance abuse, mental health, and social services). Linking thousands of high-risk patients to resources creates an enormous downstream impact and adds stress on the already-strained safety net system. Many of the existing health centers are physically out of space and capital funds are often limited in availability. The inherent capacity issues must be addressed across the health system, social services, and community to realize the long-term benefits and system change possible in Whole Person Care.” -Contra Costa “The decrease in psychiatric hospital days suggest that these individuals are being connected to appropriate mental health services to avoid additional hospitalizations.” -San Joaquin In contrast, factors that promoted better care included targeted use of financial incentives to motivate meeting set goals, particularly for partner organizations. For example, eight Pilots had financial incentives linked to improvements in follow-up after hospitalization for mental illness. In attempt to meet these incentives, several Pilots developed teams dedicated to behavioral health crisis response, and improved linkage of enrollees to ongoing behavioral health services in the community. Additionally, ten Pilots had financial incentives specifically focused on improving initiation and engagement of enrollees in alcohol and other drug dependence treatment. In attempt to meet these incentives, multiple Pilots were focused on ensuring patients with opioid use disorder (OUD) in the ED were administered or prescribed buprenorphine and then assisted with engagement in outpatient SUD treatment. viel Better Care | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program In PY 5 surveys, Pilots indicated relatively high impact of WPC on overall care quality, with average rating of 7.6 of 10, where 0 is “very low impact” and 10 is “very high impact” (data not shown). Pilots were also asked about aspects of care delivery that improved for WPC enrollees attributed to WPC (Exhibit 123). Pilots indicated highest impact of WPC on enrollee access to needed services (8.3 of 10), followed by impact on comprehensiveness (7.6) and timeliness of services provided (7.3). Exhibit 123: WPC Pilot Perceptions of Impact on Aspects of Better Care, PY5 Comprehensvenessofavaiable vices i > VS Timeliness of services provided Source: PY 5 Lead Entity (LE) Survey (n=25), June-August 2020. Note: Ratings of impact on a scale of 0-10, where 0 = “very low” and 10 = “very high”. Whole Person Care Final Evaluation Report | Better Care UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Chapter 11: Better Health WPC Pilots aimed to “reduce inappropriate emergency and inpatient utilization” and “improve health outcomes for the WPC population.” This chapter addresses the following evaluation question: “To what extent did the Pilot: a)improve beneficiary care and health outcomes, including reduction of avoidable utilization of emergency and inpatient services; and b) improve outcomes such as controlled blood pressure and Hemoglobin Alc (HbA1c)?” Data sources for this chapter included Quarterly Enrollment and Utilization Reports from PY 2 to PY 6 and Medi-Cal enrollment and claims data. The Quarterly Enrollment and Utilization Reports were used to identify enrollees and dates of enrollment. UCLA used Medi-Cal claims data, which included both managed care and fee-for-service encounters, to construct WPC metrics per the WPC Technical Specifications to create two universal metrics (ambulatory care: emergency department visits and inpatient utilization) and three variant metrics (controlled blood pressure, comprehensive diabetes care, and all cause readmissions) to further examine how enrollee health and acute care use was impacted by WPC. UCLA further constructed a measure of use of long-term care for a clearer understanding of changes in patterns of care. UCLA measured trends before and during WPC for each metric and measure based on the date of an individual WPC enrollee’s enrollment. UCLA examined changes in trends before and during WPC using a difference-in-difference (DD) analysis by modeling the changes in yearly increments up to 2 years (Pre-Year 1 and Pre-Year 2) before WPC enrollment and up to 5 years (Year 1, 2, 3, 4, and 5) during WPC. For these, the DD analysis measured the trends or change in yearly rates from Pre-Year 2 vs. Pre-Year 1 for both WPC enrollees and the control group; the change in the yearly rate during WPC from Year 1 to Year 5 for both WPC enrollees and the control group; and the difference between the changes in WPC enrollees vs. the control group from before to during WPC. These estimates were adjusted for beneficiary demographics as well as health status and use of services pre-WPC. Further details can be found in Appendix A. To better understand WPC outcomes, UCLA examined the program impact on enrollees with serious mental illness (SMI), substance use disorders (SUD), or experiencing homelessness (SMI/SUD/HML enrollees) compared to enrollees without these complicating conditions. The latter group was composed of enrollees who were medically complex including those with multiple chronic conditions and those at high risk for various reasons (MC/HR enrollees). UCLA used the Annual WPC Variant and Universal Metric Reports submitted by Pilots to DHCS from baseline to PY 6 to report on five variant metrics (decreased jail incarceration, overall beneficiary health, controlled blood pressure, comprehensive diabetes care, and depression remission at 12 months), calculated by Pilots based on electronic medical records, chart review, visi: Better Health | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program or other administrative data and therefore not replicable by UCLA. UCLA reported a weighted average rate for the available metrics across all Pilots that reported each metric. For additional detail on data sources and methodology please see Appendix B. Utilization of Acute and Long-Term Care Services UCLA created three measures of acute and long-term health care utilization and examined the trends on an annual basis. Two of these measures, emergency department visits and hospitalizations, were required by WPC and the program aimed to reduce the inappropriate use of these services. The measure of long-term care stays was not required by WPC. UCLA used these measures to illustrate potential changes in patterns of delivery of care under WPC. Ambulatory Care: Emergency Department Visits Ambulatory Care: Emergency Department Visits is a WPC universal metric that measures the rate of emergency department (ED) visits that do not result in hospitalization. UCLA reported this metric per 1,000 beneficiaries per year. The intended direction of the metric and DD is decrease. Whole Person Care Final Evaluation Report | Better Health a Exhibit 124 shows an increase in the number of ED visits before WPC by 365 visits per 1,000 beneficiaries per year for WPC enrollees and by 314 visits for the controls. During WPC, this rate declined by 196 and 118 visits per year for WPC enrollees and controls, respectively. The declining change from before to during WPC was significantly greater for WPC enrollees compared to the control group by 130 visits (DD). ysl Better Health | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ee err) Health Economics and Evaluation Research Program Exhibit 124: Trends in Ambulatory Care: Emergency Department Visits per 1,000 Beneficiaries per Year Before and During WPC, PY 2 - PY6 ° © ta N R a 8 a Q eal av 8 Nv se 4 a nN a 2 o S N am g a » g Q ot st a aio “i t a a 2 ao x = 3 S ae “ a ; + | [ Pre-Year 2 Pre-Year 1 Year 1 Year 2 Year 3 Year 4 Year 5 Before WPC During WPC ™ WPCEnrollees ™ Control Group NTO bla ASE <9 yey Ties Tole ae TTA ae Between Changes Difference (DD) WPC Enrollees 365* -196* -561* Control Group 314* -118* -431* -130* | Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: Includes ED visits that do not result in hospitalization. * Denotes ps0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC — Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). Whole Person Care Final Evaluation Report | Better Health erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program When examining the MC/HR subpopulation, the declining change in yearly ED visits from before to during WPC was significantly different from the control group by only 11 fewer visits per 1,000 beneficiaries per year (Exhibit 125). Comparatively, SMI/SUD/HML enrollees had a declining rate that was greater than their controls by 173 visits per 1,000 beneficiaries per year. Exhibit 125: Trends in Ambulatory Care: Emergency Department Visits per 1,000 Beneficiaries per Year Before and During WPC, PY 2 - PY 6, by Subpopulations Mm 1,158 ME 1,052 1,293 me 1,174 Mam 1,076 1,053 M941 «887 2 es | «2,460 2,182 2,282 (—Ns1,706 1931 (1,410 1787 («1,316 1,614 1222 «1,442 MM «1,159 Ma 868 808 S74 M724 Mmm 696 Mm 634 nC 4 4 nN ” s wn Nn a a n ” t wn S S S S 5 S S S S S S S S S @ o o o @ a a a 3 a o a 2 a > > > > > > > > > > > > > > 2 2 2 2 a a a a Before WPC During WPC Before WPC During WPC Medically Complex or High Risk SMI/SUD or Experiencing Homelessness ™WPCEnrollees Control Group a TaN] ara yt e stile -os Peat bettate (ol Tat 4 Change ln Lefer Ta) Toole) orate 4 i Pye} wec wec EUS (Cy) Medically Complex or High_| WPC Enrollees | 209* -82* -291* | Risk Control Group | 188* -92* -280* -11* SMI/SUD or Experiencing WPC Enrollees | 422* -237* -659* Homelessness Control Group | 359* -127* -487* -173* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: Includes ED visits that do not result in hospitalization. * Denotes ps0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). SMI/SUD is serious mental illness or substance use disorder. Better Health | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Inpatient Utilization is a WPC universal metric that measures the rate of acute inpatient care and services. UCLA reported this metric per 1,000 beneficiaries per year. The intended direction of the metric and DD is decrease. Exhibit 126 shows an increase in the number of hospitalizations before WPC by 163 and 145 stays per 1,000 beneficiaries per year for WPC enrollees and controls, respectively. During WPC, this rate declined by 57 stays per year, while it only declined by 30 stays per year for controls. Comparing the changes from before to during WPC, WPC enrollees declining rate was greater by 45 stays compared to controls (DD). SMI/SUD/HML enrollees had a larger declining rate (53 fewer stays per 1,000 beneficiaries per year), but the decline was also present for MC/HR enrollees (21 fewer stays; data not shown). Inpatient Utilization Exhibit 126: Trends in Inpatient Utilization per 1,000 Beneficiaries per Year Before and During WPC, PY 2- PY6 a 2 el o 48 a S 2 zo & 2 es = $ a a an) a > 2 B 2 | | | | | i | | | l ] Pre-Year 2 Pre-Year 1 Year 1 Year 2 Year 3 Year 4 Year S Before WPC During WPC @ WPCEnrollees ™ Control Group amet} Bla Orla) 4-9 Difference Difference-in- Le folg MWe Ae Between Changes Difference (DD) WPC Enrollees 163* -57* -220* Control Group 145* -30* -176* -45* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes p<0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in- difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). Whole Person Care Final Evaluation Report | Better Health eater 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Long-Term Care Stays UCLA calculated the number of long-term care stays per 1,000 beneficiaries per year to show patterns of change in utilization of all services. Long-term care stays are likely to increase as beneficiaries age or their health deteriorates. Therefore, the anticipated direction of this measure and DD is increase. Exhibit 127 shows an increase of 55 long-term care stays per 1,000 members per year for WPC enrollees and the control group before WPC. The increasing trend continues during WPC for both groups, with WPC enrollees having 131 more stays per 1,000 beneficiaries per year and the controls having 53. The change in trends from before to during WPC was significantly greater for WPC compared to controls by 78 stays per 1,000 beneficiaries per year (DD). While both SMI/SUD/HML and MC/HR enrollees had increasing rates of long- term care stays from before to during WPC compared to controls, it was higher among the SMI/SUD/HML enrollees (95 vs. 32 stays; data not shown). Exhibit 127: Trends in Long-Term Care Stays per 1,000 Beneficiaries per Year Before and During WPC, PY 2- PY6 ry zy © a8 Ee 2 8 g 8 3 oe 9 a 6 z + + 0 a n 2° 9 a nN a ai a N ° N a : [| . | L l rR a i Pre-Year 2 Pre-Year 1 Year1 Year 2 Year 3 Year 4 Year 5 Before WPC During WPC ™ WPCEnrollees ™ Control Group Arla mer Lai) Yearly Change Difference Difference-in- eX elena TTT aa Between Changes Difference (DD) WPC Enrollees 55* 131* 76* Control Group 55* 53* -2 78* | Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes ps0.05, a statistically significant difference. SUD services were identified as services with a SUD treatment procedure code or an NDC for pharmacotherapy. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). yx) Better Health | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research eer) Health Economics and Evaluation Research Program e Better Health Outcomes Controlling High Blood Pressure Controlling High Blood Pressure is a WPC variant metric that measures the percentage of beneficiaries aged 18 to 85 who had a diagnosis of hypertension and whose blood pressure was adequately controlled during the measurement year. The intended direction of the measure and DD is increase. Exhibit 128 shows that both WPC enrollees and controls have increasing rates of controlled blood pressure during WPC (3.2% for WPC enrollees and 3.8% for controls), but the change from before to during WPC was slightly smaller among WPC enrollees by 0.6%. Exhibit 128: Trends in Controlling High Blood Pressure Before and During WPC for WPC Enrollees and the Control Group, PY 2 - PY 6 a x oN x a eu 2 aoa x 3 = © xe 3s : : i i Pre-Year 2 Pre-Year 1 Year1 Year 2 Year 3 Year 4 Year S Before WPC During WPC MWPCEnrollees Control Group bla Cray} bela ATL -9 Pye) Difference-in- arlene During WPC IsTotal a1 LLLP) | WPC Enrollees 0.8%* 3.2%* 2.4%* Control Group 0.8%* 3.8%* 3.0%* -0.6%* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes ps0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in- difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). Whole Person Care Final Evaluation Report | Better Health Reaper 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Comprehensive Diabetes Care Comprehensive Diabetes Care is a WPC variant metric that measures the percentage of beneficiaries aged 18 to 75 with either Type 1 or Type 2 diabetes, who had controlled Hemoglobin A1c (HbA1c), with a value of less than 8%. UCLA was unable to reconstruct this metric using Medi-Cal claims data due insufficient reporting of resulting HCA1c values after a test. As an alternative, UCLA constructed a metric that examined the percentage of beneficiaries aged 18 to 75 with either Type 1 or Type 2 diabetes that had a HbA1c test during the measurement year. The intended direction of the measure and DD is increase. Exhibit 129 shows that after increasing rates before WPC, both WPC enrollees and controls had no significantly yearly change in diabetes testing during WPC. However, WPC enrollees did have higher rates of HbA1c testing during WPC overall compared to controls. Exhibit 129: Trends in HbA1c Testing Rates Before and During WPC for WPC Enrollees and the Control Group, PY 2 - PY 6 “ x 8 x x x 5 &S 6 x & e © S 2 Ry 2 8 5 mn Px va) oOo WH wn s x wo ©. 9. x 3 3S x S ] I Pre-Year 2 Pre-Year 1 Year 1 Year 2 Year 3 Year 4 Year 5S Before WPC During WPC mWPCEnrollees m Control Group Naar Lax) bla NACL) i-9 Difference Thies Before WPC TTT aoe Between Changes Difference (DD) WPC Enrollees 1.4%* -0.1% -1.6%* Control Group 1.5%* 0.1% -1.3%* -0.2% | Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes ps0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). paisa Better Health | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program All-Cause Readmission is a WPC variant metric that measures the number of acute inpatient stays during the measurement year that were followed by an unplanned acute readmission for any diagnosis within 30 days for beneficiaries ages 21 and older. The intended direction of the metric and DD is decrease. Exhibit 130 shows that readmission rates slightly increased before WPC for both WPC enrollees and controls (0.8%) and then declined during WPC by 1.1% and 1.0%, respectively. There was no significant difference in the changing yearly rates from before to during WPC between WPC enrollees and controls. All-Cause Readmission Exhibit 130: Trends in All-Cause Readmission following an Acute Inpatient Admission, Before and During WPC for WPC Enrollees and the Control Group, PY 2 - PY 6 © x & © se Bm = a 5 5 x aoa a3 N & x aos a x x x ss aS a 8 lI | | | | I Pre-Year 2 Pre-Year 1 Year 1 Year 2 Year 3 Year 4 Year 5 Before WPC During WPC ™WPC Enrollees @ Control Group AraC} NT NOEL 4-9 yess} Phas oi stole Mae TTT ae Between Changes Difference (DD) WPC Enrollees 0.8%* -1.1%* -1.8%* Control Group 0.8%* -1.0%* -1.8%* 0% Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes p<0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). Whole Person Care Final Evaluation Report | Better Health UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Trends in Better Health Based on WPC Pilot-Reported Metrics UCLA calculated the weighted average values for five variant metrics using Pilot-reported data (Exhibit 131). Some Pilots did not report planned metrics every year for reasons such as no enrollment or program activities during the reporting time period or lack of data in that time period. See Appendix B for further details on reporting for each metric, including which Pilots reported on each metric during each measurement year. Exhibit 131: Pilot-Reported Variant Metrics That Indicate Better Health Universal | Metric Name | Description Baseline | Reporting | Numbers | Improvement vs. Year Years of Pilots | Measured by Variant Reporting | Increase or by Year Decrease Variant Decrease Jail | DJI: PY1 PY 2, PY 6in PY1 Decrease Incarceration | Incarcerations | (2016) 3, PY4, Sin PY2 (DJI) per 1,000 PY5, PY6 | 7in PY3 member Tin PY4 months of 6in PYS enrollees 14 6inPY6 years of age and older Variant Overall OBH-O: Self- PY2 PY 3, PY 4in PY2 Increase Beneficiary reported 4,PY5, 6in PY3 Health (OBH) | rating for PY6 7inPY4 enrollee’s 7inPYS overall health 6 in PY6 OBH-E: Self- PY2 PY 3, PY 4inPY2 | Increase reported 4,PY5, 5 in PY3 rating for PY6 7inPY4 enrollee’s 7inPYS mental or 6 in PY6 emotional health yxk:3 Better Health | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 Universal | Metric Name | Description Baseline | Reporting | Numbers | Improvement vs. Year Years of Pilots | Measured by Variant Reporting | Increase or by Year Decrease Variant Controlled CBP-18-59: PY1 PY 2, PY 8inPY1 Increase Blood Percent of (2016) 3, PY4, 6in PY2 Pressure enrollees 18- PY5, PY6 | 7inPY3 (CBP) 59 years of 8inPY4 age whose BP 8inPYS was <140/90 8inPY6 mmHg CBP-60-85-D: | PY1 PY 2, PY 8inPY1 Increase Percent of (2016) 3, PY4, 6in PY2 enrollees 60- PY5,PY6 | 7inPY3 85 years of 8inPY4 age witha 8inPYS diagnosis of 8in PY6 diabetes whose BP was <140/90 mmHg CBP-60-85- PY1 PY 2, PY 8inPY1 Increase ND: Percent (2016) 3, PY4, 6in PY2 of enrollees PY 5, PY6 | 7in PY3 60-85 years of 8inPY4 age without a 8inPYS diagnosis of 8in PY6 diabetes whose BP was <150/90 mmHg Variant Comprehen- | CDC: PY1 PY 2, PY 11 in PY 1 | Increase sive Diabetes | Percentage of | (2016) 3, PY4, 11 in PY 2 Care (CDC) enrollees 18- PY 5, PY6 | 11 in PY3 75 years of 12 in PY 4 age with 12 in PYS diabetes 12 in PY6 Whole Person Care Final Evaluation Report | Better Health eee 6 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Universal | Metric Name | Description Baseline | Reporting | Numbers | Improvement vs. Year Years of Pilots | Measured by Variant Reporting | Increase or by Year Decrease (type 1 and type 2) who had HbA1c control (<8%) Variant PHQ NQF 0719: PY1 PY!22PY QinPY1 | Increase 9/Depression | Percentage of | (2016) 3, PY 4, 9in PY2 Remission at | enrollees 18 PY5, PY6 | 11 in PY3 12 Months years of age 14 in PY 4 (NQF 0719) and older 15 in PYS with Major 14 in PY6 Depression or Dysthymia who reached remission 12 months (+/- 30 days) after an index visit Source: Baseline, PY 2, PY 3, PY 4, PY 5, and PY 6 Annual WPC Variant and Universal Metric Reports and Whole Person Care Universal and Variant Metrics Technical Specifications (March 22, 2019). Notes: BP is blood pressure. HbA1c is the hemoglobin Alc test that measures the average level of blood sugar. yell Better Health | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research . December 2022 Health Economics and Evaluation Research Program Variant Metric: Decrease Jail Incarcerations (DJI) Seven WPC Pilots elected to report the number of incarcerations that occurred per 1,000 member months for those ages 14 or older as of December 31 of the measurement year (DJI). The overall DJI rate increased from 18 incarcerations per 1,000 member months during baseline to 24 in PY 2, but declined to 6 in PY 6 (Exhibit 132). There was variation in DJI by Pilot, for example, ranging from a low of 11 in PY 1 to a high of 358 in PY 2. One large Pilot accounted for between 72% and 83% of the denominator each year for this metric and this Pilot reported the lowest DJI rate among all Pilots for five out of six reporting years. Without this influential Pilot, the DJI rate remained steady from baseline to PY 2 at 48 and declines to 20 in PY 6 (data not shown). Exhibit 132: Number of Incarcerations per 1,000 WPC Member Months, by Program Year 358 303 & ¢ 195 85 77 82 & 18 24 20 21 . * 11 ue 11 3 4 3 Baseline PY 2 (n=160,889) PY 3 (n=372,580) PY 4 (n=473,124) PY 5 (n=507,358) PY 6 (n=515,371) (PY 1, n=241,154) @ Overall WPC @ Pilot-Specific Minimum Pilot-Specific Maximum Source: WPC Annual Universal and Variant Metric Reports, baseline through PY 6 Notes: Only Pilots that reported on this metric were included in the analysis. The number of Pilots reporting varied by year. Appendix B, Exhibit 4 provides details on which Pilots reported in each year. The denominator size is shown as sample size per year. Bars represent the range reported by Pilots, with minimum being the lowest rate reported by a Pilot and maximum being the highest rate reported by a Pilot. Whole Person Care Final Evaluation Report | Better Health err UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Variant Metric: Overall Beneficiary Health Seven WPC Pilots elected to report the percent of enrollees reporting “Excellent” or “Very Good” overall health (OBH-O) and the percent of enrollees reporting “Excellent” or “Very Good” emotional health (OBH-E) as part of the overall beneficiary health metric. Overall OBH-O increased from 11% during baseline to 22% in PY 3 and then after a small decline to 19% in PY 4, it increased to 28% in PY 6 (Exhibit 133). There was variation by Pilot in percent reporting good overall health, ranging from a low of 5% to a high of 44%. Exhibit 133: Percent of Enrollees Who Reported “Excellent” or “Very Good” Overall Health (OBH-O), by Year 42% ae 39% 32% 31% 7 ¢ & 28% 22% ] 19% 21% 11% + Bo, A 5% 4% Baseline PY 3 (n=12,937) PY 4 (n=11,834) PY 5 (n=10,756) PY 6 (n=7,208) (PY 2, n=4,919) WOverall WPC — @Pilot-Specific Minimum __@ Pilot-Specific Maximum Source: WPC Annual Universal and Variant Metric Reports, baseline through PY 6 Notes: Only Pilots that reported on this metric were included in the analysis. The number of Pilots reporting varied by year. Appendix B, Exhibit 5 provides details on which Pilots reported in each year. The denominator size is shown as sample size per year. Bars represent the range reported by Pilots, with minimum being the lowest rate reported by a Pilot and maximum being the highest rate reported by a Pilot. Better Health | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program Overall OBH-E increased from 17% in baseline to 27% in PY 6 (Exhibit 134). Similar to OBH-O, variation exited between Pilots with a range of 5% in baseline to 36% in PY 6. Exhibit 134: Percent of Enrollees Who Reported “Excellent” or “Very Good” Emotional Health (OBH-E), by Year 35% 36% 28% + 25% 24% 27% 20% 17% 8% fe 4% Baseline PY 3 (n=10,152) PY 4 (n=11,785) PY 5 (n=10,736) PY 6 (n=7,145) (PY 2, n=4,829) Overall WPC @Pilot-Specific Minimum —_@ Pilot-Specific Maximum Source: WPC Annual Universal and Variant Metric Reports, baseline through PY 6 Notes: Only Pilots that reported on this metric were included in the analysis. The number of Pilots reporting varied by year. Appendix B, Exhibit 6 provides details on which Pilots reported in each year. The denominator size is shown as sample size per year. Bars represent the range reported by Pilots, with minimum being the lowest rate reported by a Pilot and maximum being the highest rate reported by a Pilot. Whole Person Care Final Evaluation Report | Better Health eee 6 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Variant Metric: Controlling High Blood Pressure Eight WPC Pilots elected to report on the percent of three groups of enrollees (individuals age 18-59, individuals age 60-85 with diabetes, and individuals age 60-85 without diabetes) whose blood pressure was adequately controlled during the measurement year. The blood pressure control rate for all three groups increased from baseline to PY 4 before declining in PY 5 and PY 6 (Exhibit 135, Exhibit 136, Exhibit 137). Rates of blood pressure control remained above baseline in PY 6 for all three groups. There was variation by Pilot in the percent of enrollees who had controlled blood pressure in all measurement years. Many Pilots had denominators less than 10 during all measurement year, resulting in substantial variation in the rates by Pilots. Exhibit 135: Percent of WPC Enrollees 18 to 59 years old with Controlled Blood Pressure, by Program Year 100% 100% 89% 0% © ’ , 70% 63% ay e oa 36% | es ers Evi, 31% 9% Baseline PY 2 (n=144) PY3(n=438) © PY4(n=1,497) — PY5 (n=2,392) PY 6 (n=2,401) (PY 1, n=413) Overall WPC Pilot-Specific Minimum _@ Pilot-Specific Maximum Source: WPC Annual Universal and Variant Metric Reports, baseline through PY 6 Notes: Only Pilots that reported on this metric were included in the analysis. The number of Pilots reporting varied by year. Appendix B, Exhibit 1 provides details on which Pilots reported in each year. The denominator size is shown as sample size per year. Bars represent the range reported by Pilots, with minimum being the lowest rate reported by a Pilot and maximum being the highest rate reported by a Pilot. Controlled blood pressure was defined as less than 140/90 mmHg for those age 18 to 59. yy Better Health | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research rer Health Economics and Evaluation Research Program Exhibit 136: Percent of WPC Enrollees 60 to 85 years old and Diabetic with Controlled Blood Pressure, by Program Year 100% 100% 100% 100% ’ 67% ax , 56% a I 1% 47% 35% 0% Baseline PY 2 (n=30) PY3 “nse) PY4 (n237) PYS (nn628) PY6 inn896) (PY 1, n=43) 76% Overall WPC Pilot-Specific Minimum _—_@ Pilot-Specific Maximum Source: WPC Annual Universal and Variant Metric Reports, baseline through PY 6 Notes: Only Pilots that reported on this metric were included in the analysis. The number of Pilots reporting varied by year. Appendix B, Exhibit 2 provides details on which Pilots reported in each year. The denominator size is shown as sample size per year. Bars represent the range reported by Pilots, with minimum being the lowest rate reported by a Pilot and maximum being the highest rate reported by a Pilot. Controlled blood pressure was defined as less than 140/90 mmHg for those age 60 to 85 with a diagnosis of diabetes. A rate of 0% indicated that no enrollees had controlled blood pressure in the measurement year. Exhibit 137: Percent of WPC Enrollees 60 to 85 years old and not Diabetic with Controlled Blood Pressure, by Program Year 100% 100% 100% 86% 89% 83% 81% 829 57% 52% a6 48% Pa 0% Baseline PY 2 (n=28) PY 3 (n=96) PY 4(n=398) PY5(n=788) —_—PY 6 (n=990) (PY 1, n=57) WOverallWPC @Pilot-Specific Minimum —@ Pilot-Specific Maximum Source: WPC Annual Universal and Variant Metric Reports, baseline through PY 6 Whole Person Care Final Evaluation Report | Better Health eee 6 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Notes: Only Pilots that reported on this metric were included in the analysis. The number of Pilots reporting varied by year. Appendix B, Exhibit 3 provides details on which Pilots reported in each year. The denominator size is shown as sample size per year. Bars represent the range reported by Pilots, with minimum being the lowest rate reported by a Pilot and maximum being the highest rate reported by a Pilot. Controlled blood pressure was defined as less than 150/90 mmHg for those age 60 to 85 without a diagnosis of diabetes. A rate of 0% indicated that no enrollees had controlled blood pressure in the measurement year. Variant Metric: Comprehensive Diabetes Care (CDC) Twelve WPC Pilots elected to report the percent of enrollees age 18 to 75 with either Type 1 or Type 2 diabetes, who had controlled Hemoglobin A1c (HbA1c), with a value of less than 8% (CDC). The overall CDC rate increased from 52% in baseline, to 58% in PY 3, and ended at 54% in PY 6 (Exhibit 138). There was variation by Pilot, ranging from a low of 0% in baseline to a high of 100% in PY 2. Exhibit 138: Percent of Adult Enrollees with Diabetes Who Had Controlled HbA1c, by Program Year 100% 80% 84% 83% 85% ® © rs 69% ¢ 52% 53% a 56% * 54% | 23% i i 0% Baseline PY2(n=917) PY3(n=1,102) PY 4(n=1,700) PY (n=2,033) PY 6 (n=1,807) (PY 1, n=710) WOverall WPC — @Pilot-Specific Minimum _@ Pilot-Specific Maximum Source: WPC Annual Universal and Variant Metric Reports, baseline through PY 6 Notes: Only Pilots that reported on this metric were included in the analysis. The number of Pilots reporting varied by year. Appendix B, Exhibit 7 provides details on which Pilots reported in each year. The denominator size is shown as sample size per year. Bars represent the range reported by Pilots, with minimum being the lowest rate reported by a Pilot and maximum being the highest rate reported by a Pilot. A rate of 0% indicated that no enrollees had controlled HbA1c scores in the measurement year. HbA1c is the hemoglobin Alc test that measures the average level of blood sugar. yi Better Health | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research . December 2022 Health Economics and Evaluation Research Program Variant Metric: PHQ-9/Depression Remission at 12 Months (NQF 0719) Fifteen WPC Pilots elected to report the percent of enrollees age 18 or older with major depression or dysthymia who reached remission measured at 12 months, plus or minus 30 days, after an index visit (NQF 0719). There was some increase in the overall NQF 0719 rate, but it remained low all years of the program, at 4% or less (Exhibit 139). There was variation by Pilot, ranging from a low of 0% in all measurement years to a high of 100% in PY 3. Variation was largely due to small denominators. Exhibit 139: Percent of Enrollees Age 18 or Older with Major Depression or Dysthymia Who Reached Remission at 12 Months, by Program Year 100% ¢ 9 50% 44% 50% ¢ e — 25% ¢ ¢ e 2% 3% 1% 3% 3% 4% o- eee 0% 0% 0% 0% 0% 0% Baseline PY2(n=660) PY3(n=1,339) PY4(n=2,878) PY5 (n=3,126) PY 6 (n=2,175) (PY 1, n=738) Overall WPC Pilot-Specific Minimum —_@ Pilot-Specific Maximum Source: WPC Annual Universal and Variant Metric Reports, baseline through PY 6 Notes: Only Pilots that reported on this metric were included in the analysis. The number of Pilots reporting varied by year. Appendix B, Exhibit 8 provides details on which Pilots reported in each year. The denominator size is shown as sample size per year. Bars represent the range reported by Pilots, with minimum being the lowest rate reported by a Pilot and maximum being the highest rate reported by a Pilot. A rate of 0% indicated that no enrollees reached remission in the timeframe. Pilot Assessment of Challenges to and Impact of WPC on Better Health Pilots reported on challenges to achieving better health, factors that promoted better health, and their overall their perceptions of aspects of care delivery that were impacted by WPC. In PY 6 follow-up interviews and bi-annual narrative reports, Pilots described their challenges to control of high blood pressure and provision of comprehensive diabetes care were closely related to the shift to telehealth during the earlier phases of the COVID-19 pandemic and limited availability of primary care appointments, which led to enrollees who were concerned with contracting COVID-19 to forgo or delay care. Furthermore, a small group of Pilots had Whole Person Care Final Evaluation Report | Better Health eee 6 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program financial incentives tied to these metrics or reported activities focused specifically on diabetes or blood pressure control. Instead, most focused on health education (e.g., nutrition class, access to a dietitian, providing information on diabetes) to impact these metrics. Pilots also described conducting quality improvement studies to divert patients from the ED to more appropriate settings. These studies aimed to understand enrollee behavior and motivation for ED visits, as well as best practice methods for diverting patients from the ED, including use of mobile crisis teams and real-time notifications of ED visits to primary care providers. These studies were complemented with care coordinator efforts to build trust with enrollees and help navigate enrollees to more appropriate settings. “Understanding what leads people to utilize the Crisis System as their primary source of care will be an ongoing process; early exploration indicates the reasons are much more varied than expected. We are developing approaches to talk with consumers and families to better understand their needs so we can better work with them to design the crisis continuum of care and interventions that are optimized to meet their needs.” -Alameda “WPC practitioners report difficulty breaking ER visit habits when office visits are less accessible due to a shortage of physicians in the community, especially when medicine is urgently needed after normal business hours.” -Shasta In PY 5 surveys, Pilots perceived rated the impact of WPC on improved enrollee health and well- being at 8.3 out of 10, where 0 is “very low impact” and 10 is “very high impact” (Exhibit 140). Pilots also indicated a moderately high impact of WPC on reducing inappropriate emergency department visits and hospitalization (7.7). Exhibit 140: WPC Pilot Perceptions of Impact on Aspects of Better Health, PY 5 Inoproved enrollee Wealth and.well-peing es BS Reduced inappropriate emergency department visits ce esis 77 and hospitalization Source: PY 5 Lead Entity (LE) Survey (n=25), June-August 2020. Note: Ratings of impact on a scale of 0-10, where 0 = “very low” and 10 = “very high”. Better Health | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program i Chapter 12: Lower Cost This chapter addresses the following evaluation question: “To what extent did WPC Pilots reduce costs of health care for WPC enrollees compared to the control group and were total Medi-Cal expenditures reduced during the WPC program?” Data sources for this chapter included Quarterly Enrollment and Utilization Reports from PY 2 to PY 6 and Medi-Cal enrollment and claims data. UCLA used the Quarterly Enrollment and Utilization Reports to identify enrollees and dates of enrollment. UCLA calculated estimated payments for all services provided to WPC enrollees and the control group before WPC and during WPC using Medi-Cal claims and encounter data. Dental claims were not included as part of this analysis. Medi-Cal payments were estimated by creating unique categories of service and attributing a fee to each Medi-Cal claim in that category (Appendix A: Attributing Estimated Medi-Cal Payments to Claims). The resulting measure estimates the annual average payment per beneficiary. This methodology allowed UCLA to estimate payments for WPC enrollees and the control group before each enrollee’s WPC enrollment and during WPC and assess if payments for WPC enrollees declined more than for the control group using the DD methodology. UCLA developed DD models to measure changes in total estimated payments and in specific categories of services including outpatient services, outpatient medications, ED visits, hospitalizations, and long-term care stays. These estimates were adjusted for beneficiary demographics, health status, and use of services pre-WPC. Further details can be found in Appendix A. The findings were not subject to potential seasonality in service utilization due to rolling enrollment throughout the year and measuring change following the date of enrollment per beneficiary. The payment amounts reported in this section are estimates and are not equivalent to actual Medi-Cal expenditures for multiple reasons, including significant differences between this attribution methodology vs. per member per month payments to managed care plans for enrolled beneficiaries. These estimated payments are primarily intended to compare change in trends between WPC enrollees and the control group. See Appendix A for further detail and limitations. UCLA measured trends before and during WPC for each metric based on the date of an individual WPC enrollee’s enrollment. UCLA examined changes in trends before and during WPC using a difference-in-difference (DD) analysis by modeling the changes in yearly increments up to two years (Pre-Year 1 and Pre-Year 2) before WPC enrollment and up to five year increment (Year 1, 2, 3, 4, and 5) during WPC. For these, the DD analysis measured the Whole Person Care Final Evaluation Report | Lower Cost err UCLA Center for Health Policy Research Health Economics and Evaluation Research Program trends or change in yearly rates from Pre-Year 2 vs. Pre-Year 1 for both WPC enrollees and the control group; the change in the yearly rate during WPC from Year 1 to Year 5 for both WPC enrollees and the control group; and the difference between the changes in WPC enrollees vs. the control group from before to during WPC. These estimates were adjusted for beneficiary demographics as well as health status and use of services pre-WPC. To better understand WPC outcomes, UCLA examined the program impact on enrollees with serious mental illness (SMI), substance use disorders (SUD), or experiencing homelessness (SMI/SUD/HML enrollees) compared to enrollees without these complicating conditions. The latter group was composed of enrollees who were medically complex including those with multiple chronic conditions and those at high risk for various reasons (MC/HR enrollees). UCLA created seven measures of health care costs and examined the trends on an annual basis. These measures were not required by WPC as performance metrics. UCLA used these measures to illustrate potential changes in health care costs associated with better care and better health measures under WPC. The estimated changes in costs by category of service do not sum to the overall costs because each change was modeled separately. Total Estimated Medi-Cal Payments UCLA measured total estimated Medi-Cal payments before and during WPC as described above. These estimates include payments for all health and behavioral services used by beneficiaries such as outpatient services, hospitalizations, outpatient pharmaceuticals, imaging and laboratory services, behavioral health services, and long-term care stays. WPC was expected to lead to a decline in total costs. v<{0 Lower Cost | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Exhibit 141 shows that total estimated payments per beneficiary per year were significantly increasing before WPC for both WPC enrollees and the controls by $3,205 and $2,943, respectively. The total estimated payments decreased during WPC by $955 and $834 for WPC enrollees and controls, respectively. The declines in total estimated payments from before WPC to during WPC per beneficiary per year were significantly greater for WPC enrollees compared to the control groups by $383 (DD). Whole Person Care Final Evaluation Report | Lower Cost December 2022 Exhibit 141: Trends in Total Estimated Medi-Cal Payments Before and During WPC, PY 2 - PY6 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program 28 gg g 3 z Bo aS RA a g 2 on 4 aa << a a 2 a a ° av an a 4 s a8 am 8 Son as a a s Bos 8 ad no an Hn a a, a | B g R Pre-Year 2 Pre-Year 1 Year1 Year 2 Year 3 YearS Before WPC During WPC MWPCEnrollees Control Group Differe ETc gs » PA ange ence (DD WPC Enrollees $3,205* -$955* -$4,160* Control Group $2,943* -$834* -$3,777* -$383* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes ps0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). Lower Cost |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ee err) Health Economics and Evaluation Research Program Additional analyses showed that difference in the change in total payment per year from before to during WPC between enrollees and controls differed between SMI/SUD/HML enrollees and MC/HR enrollees. Compared to controls, MC/HR enrollees saw declining rates in total cost per beneficiary per year from before to during WPC that was $581 less than controls ( Exhibit 142). Comparatively, SMI/SUD/HML enrollees saw a decline of $311 compared to controls. Exhibit 142: Trends in Total Estimated Medi-Cal Payments Before and During WPC, PY 2 - PY 6, by Subpopulations ° ot an a st nN a Ra ANS om Fh RG 3 o BQ am SH NS OF IQ B a 0 ve ww om WS ut SQ OK 8, So 2, B & 2. | Se Be Bo @S ge as 38 XZ. te SS am Ae Feo RL] Va mo BA ge oe aN DN BA ah DA ws Yo ee Nn 2. an Yi - o at AY FQ wm ro a é ~ ~ HS Oi an os uw an @ We nN a 4 N -” + in n_ 4 4 nN ” + wn S 5 S S S S S S S S S S 5 S o o o o o o o o 3 o o o @ o > > > > > > > > > > > > > > 2 2 2 2 a a a a Before WPC During WPC Before WPC During WPC Medically Complex or High Risk SMI/SUD or Experiencing Homelessness MWPCEnrollees Control Group ala] Nea ig eee Pieces Change Change ln 5 LTS Ta) - To ole) erty Eee yes} wPCc wec u (DD) Medically Complex or High | WPC Enrollees | $2,108* -§502* -$2,611* Risk Control Group | $1,618* | -$411* -$2,030* -$581* SMI/SUD or Experiencing | WPC Enrollees | $3,604* -$1,120* | -$4,724* Homelessness Control Group | $3,425* -$988* -$4,413* | -$311* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: Includes ED visits that do not result in hospitalization. * Denotes ps0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). SMI/SUD is serious mental illness or substance use disorder. Whole Person Care Final Evaluation Report | Lower Cost ery) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Estimated Payments for Outpatient Services UCLA estimated Medi-Cal payments for outpatient services. Outpatient services are likely to increase due to unmet need and increased access to these services, but payments are likely to decrease once health needs are addressed and service use declines. Exhibit 143 shows that estimated payments for outpatient services were significantly increasing per beneficiary per year before WPC for both WPC enrollees and the controls by $690 and $632, respectively. Both groups had declines in estimated outpatient payments during WPC by $285 and $247 per beneficiary per year for WPC enrollees and controls, respectively. The declining rates of outpatient costs from before to during WPC was greater among WPC enrollees compared to controls by $96 per beneficiary per year (DD). Exhibit 143: Trends in Estimated Medi-Cal Payments for Outpatient Services Before and During WPC, PY 2-PY6 ° 0 > 2 5 hom rn 3 1 6S $19 2 om © a a Bo nae ms mo oN mo ow oo = aA a ow 9 R a Pee ~s Dn @ al a8 a x & x ak uw of om a a st See uw N Pre-Year 2 Pre-Year 1 Year1 Year 2 Year 3 Year 4 Year 5S Before WPC During WPC MWPCEnrollees Control Group secon Wg ) g WP ange D DD WPC Enrollees $690* -$285* -$975* Control Group $632* -$247* -$880* -$96* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes ps0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). p£U) Lower Cost | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ee err) Health Economics and Evaluation Research Program Both SMI/SUD/HML enrollees and MC/HR enrollees saw declining rates of outpatient services costs compared to controls, but it was greater among MC/HR enrollees ($185 vs. $63; Exhibit 144). Exhibit 144: Trends in Estimated Medi-Cal Payments for Outpatient Services Before and During WPC, PY 2 - PY 6, by Subpopulations $2,653 $3,063 a S$2,277 $3,317 Year] pum $2,508 EEE $3,000 Year 2 gum $7339 «$2,961 lam $2,283 ME $2,867 (mmm $2,233 = (mmm $2,083 EEE $3,159 Pre-Year 2 Sess $3,049 = Pre-Year 1 (pepe re es. Year 1 $3 Byaee $4,168 [a $3,408 EN $3,925 Year3 ummm $3,158 NN $3,791. Year’ pum $298) «$2,487 Pre-Year2 sess $1,849 ENE «$3,553 YearS pum $2,622 4 o s wn n_ G 5 S 5 S © a o o a > > > > > 2 a Before WPC During WPC Before WPC During WPC Medically Complex or High Risk SMI/SUD or Experiencing Homelessness mWPCEnrollees Control Group ACE TaNy ACTA hie} SS CoD Change [ete aT1a) ae Before WPC During WPC Changes ¥ an Medically Complex or | WPC Enrollees | $576* -$166* -$742* High Risk Control Group | $428* -$129* -$557* | -$185* SMI/SUD or WPC Enrollees | $732* -$328* -$1,060* Experiencing Homelessness Control Group | e797 -$290* -$997* -$63* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: Includes ED visits that do not result in hospitalization. * Denotes ps0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). SMI/SUD is serious mental illness or substance use disorder. Whole Person Care Final Evaluation Report | Lower Cost ery) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Estimated Payments for Outpatient Medications UCLA estimated Medi-Cal payments for outpatient medications. Payments for outpatient medications are likely to increase due to unmet need and increased access to these medications, but payments are likely to stabilize or decrease once health needs are addressed. Exhibit 145 shows that estimated outpatient medication payments per beneficiary per year were significantly decreasing before WPC for both WPC enrollees and the controls by $50 and $44, respectively. The estimated payments decreased at a slower rate during WPC by $10 and $63 per beneficiary per year for WPC enrollees and controls, respectively. Therefore, the change in yearly costs of outpatient medication from before WPC to during WPC was significantly more for WPC enrollees compared to the controls by $58 (DD). Exhibit 145: Trends in Estimated Medi-Cal Payments for Outpatient Medications Before and During WPC, PY 2 - PY6 + 4 & Pre-Year 2 Pre-Year 1 Year1 Year 2 Year 3 Year 4 Year 5 MEE 3208s Os 838 ME 20: 1.70 es :: es «:. a :2 a: < $1,695 ie 52 25: EEE 51570 Before WPC During WPC MWPCEnrollees Control Group Para piece Ti Yearly Change Ae AOU aA ae vt RR Te HD ae st 0. 5 a 2 eo 8 a i a a g z x a x oy Me a @ + a 4 a oy a 4 2 age an am Rm 3 ag dg °S "2 “S of Gd gc VS BR OR om en li li | | | I nc a 4 Nn ” s wn Nn a dq n ” + wn S G S S 5 S 5 S 5 S S S S S a o o o a o o o a o o o a o > > > > > > > > > > > > > > 2 2 2 2 a a a a Before WPC During WPC Before WPC During WPC Medically Complex or High Risk SMI/SUD or Experiencing Homelessness @WPCEnrollees Control Group NEI) Seley Pye} Pe Change ie Change Toa t-1) - EIN e eer rt Prieta Wize 2 e (DD) Medically Complex or High | WPC Enrollees | -$25* $145* $171* | Risk Control Group | -$18* $33* $51* $119* SMI/SUD or Experiencing | WPC Enrollees | -$58* -$67* -$8* | Homelessness Control Group | -§53* -$98* -$45* | $36* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: Includes ED visits that do not result in hospitalization. * Denotes ps<0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). SMI/SUD is serious mental illness or substance use disorder. Whole Person Care Final Evaluation Report | Lower Cost ery) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Estimated Payments for Emergency Department Visits UCLA estimated Medi-Cal payments for emergency department (ED) visits followed by discharge. The anticipated direction of the measure and DD under WPC is decrease, consistent with an intended decline in ED visits. Exhibit 147 shows that estimated emergency department visit payments were significantly increasing before WPC for both WPC enrollees and the controls by $193 and $187 per beneficiary per year. The estimated payments decreased during WPC by $60 and $49 for WPC enrollees and controls, respectively. The annual change in trends from before WPC to during WPC declined by $18 more per year for WPC enrollees compared to the control group (DD). Exhibit 147: Trends in Payments for Emergency Department Visit Before and During WPC, PY 2 - PY6 Q ” 8 R 8 2 + wm Pre-Year 2 Pre-Year 1 Year 1 Year 2 Year 3 Year 4 Year 5 ss: DO sss a 7 EN 722 oie ME ss: EE 3327 $456 EE «0 ME s520 ME 5:07 Before WPC During WPC @WPCEnrollees Control Group Fs er F Betwee ) ralen Lp py ay ey D DD WPC Enrollees $193* -$60* -$254* Control Group $187* -$49* -$235* -$18* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes p<0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). Lower Cost |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ee err) Health Economics and Evaluation Research Program Comparing the impact of WPC on the estimated costs of emergency department visits among enrollees with and without the highest need conditions showed that compared to controls the trends in emergency department costs from before to during WPC increased for MC/HR enrollees ($21 per beneficiary per year), but declined for SMI/SUD/HML enrollees (-$32 per beneficiary per year; Exhibit 148). Exhibit 148: Trends in Estimated Emergency Department Payments Before and During WPC, PY 2 - PY 6, by Subpopulations S35 2 cm fe & B es 2B go & 6 n oe ¢ a S 2 OH ee _ ow 3 Ss bad + # Pr 3 2 st x 9 R Pa = a reo 25 GG, 8 cS 6 on s da 8h OS Be Ms man Mm a RE OR MR BQ 22 HS Ba apn on Te x qQ n 4 4 nN ” + ry N a 4 a ” t rr) 5 S 5 S 5 S 5 S 5 S S S S S a 3 @ a @ a o a 3 o 3 o 3 a > > > > > > > > > > > > > > 2 2 2 2 a a a a Before WPC During WPC Before WPC During WPC Medically Complex or High Risk SMI/SUD or Experiencing Homelessness @WPCEnrollees Control Group Aled alin Pn ee Difference- Change Change Tis si LTC) ; a ele oy 4 nes Piece} wec wec B (DD) Medically Complex or High | WPC Enrollees | $94* -$4* -$98* Risk Control Group | $88* -$31* -$119* $21* SMI/SUD or Experiencing | WPC Enrollees | $229* -$81* -$310* Homelessness Control Group | $223* -$55* -$278* -$32* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: Includes ED visits that do not result in hospitalization. * Denotes ps<0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). SMI/SUD is serious mental illness or substance use disorder. Whole Person Care Final Evaluation Report | Lower Cost ery) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Estimated Payments for Hospitalizations UCLA estimated Medi-Cal payments for hospitalizations. The anticipated direction of the measure and DD is decrease consistent with an intended decline in hospital stays. Exhibit 149 shows that estimated hospitalization payments were significantly increasing before WPC for both WPC enrollees and the controls ($752 and $585 per beneficiary per year, respectively). The estimated payments for hospitalizations decreased significantly during WPC by $472 and $329 for WPC enrollees and controls, respectively. The change in trends for estimated hospitalization payments declined significantly more from before WPC to during WPC for WPC enrollees compared to the control group ($310 per beneficiary per year; DD). This significant decline compared to controls was present for both SMI/SUD/HML enrollees (-$360) and MC/HR enrollees (-$172; data not shown). Exhibit 149: Trends in Payments for Hospitalizations Before and During WPC, PY 2 - PY6 ‘8 4 2 e z 2 s Dw Ry 7, gS 3 ~ 6 n wn 4 in ww o& a + 8 nA a at om wn st a an + g wo nm oO 5 x 6 q 7 a on we wn on Ss nN RD DB R 2 : 0. | | | l | Pre-Year 2 Pre-Year 1 Year 1 Year 2 Year 3 Year 4 Year 5 Before WPC During WPC @WPCEnrollees ™ Control Group WPC Enrollees $752* -$472* -$1224* Control Group $585* -$329* -$914* -$310* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes ps0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). PL‘ Lower Cost | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program arSeaitsa tae Estimated Payments for Long-Term Care Stays UCLA estimated Medi-Cal payments for long-term care stays. Payments for long-term care stays are likely to increase over time consistent with an anticipated increase in long-term care stays. Exhibit 150 shows that estimated payments for long-term care stays were decreasing before WPC for both WPC enrollees and the controls by $77 and $128 per beneficiary per year, respectively. The estimated payments significantly increased during WPC by $313 and $249 for WPC enrollees and controls, respectively. The change in annual trends of estimated payments for long-term care stays from before WPC to during WPC did not differ significantly between WPC enrollees and the control group (DD). Exhibit 150: Trends in Estimated Medi-Cal Payments for Long-Term Care Stays Before and During WPC, PY 2 - PY6 2 el g g & aS 5 o © & a 5 s 8a a4 on 8 g BS oo + Pe S 8G a a 5 a a s nn go ae 6 9 5 : [ i tl [ | Pre-Year 2 Pre-Year 1 Year 1 Year 2 Year 3 Year 4 Year 5 Before WPC During WPC @WPCEnrollees Control Group Meee Cr ange cr Play <3 Betwee Meta STW Ly » Pag Ly ue n PD) WPC Enrollees -$77* $313* $391* Control Group -$128* $249* $377* -$13 Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes ps0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). Long-term care includes stays at skilled nursing facilities and intermediate care facilities. Whole Person Care Final Evaluation Report | Lower Cost Renee oly) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program There was a significant difference in trends in estimated payments for long-term care between WPC enrollees and controls when restricting to MC/HR enrollees (Exhibit 151). The increasing estimated costs from long-term care stays was smaller among these WPC enrollees by $79per beneficiary per year compared to controls. Comparatively, SMI/SUD/HML enrollees saw an increase of $47 compared to controls. Exhibit 151: Trends in Estimated Long-Term Care Stays Before and During WPC, PY 2 - PY 6, by Subpopulations + o¢5 ow & Br am g e KR RB aN a oO 2 nog oo wo x Bw oad mo AMR ma 3 in S ag 4 i Si 8a 4 a gs a gi 1 GD — 3 a £ an wn es snow nm a an x n un t+ DOD NH ao a 20.00 no oH OH S 0 q Be Bh Bh ee Si fe & a a I i I l | I I | n 4 4 nN ” + ry N a 4 a ” t rr) 5 S S S 5 S 5 S 5 S S S S S a 3 @ a @ a o a 3 o 3 o 3 a > > > > > > > > > > > > > > 2 2 2 2 a a a a Before WPC During WPC Before WPC During WPC Medically Complex or High Risk SMI/SUD or Experiencing Homelessness ™WPCEnrollees Control Group nian aT Difference pis Change To aT) a EIS oli Pratt se Geaces Pie WPc (DD) Medically Complex or High | WPC Enrollees | -§74* $171* $246* | Risk Control Group | -$99* $225* $325* |___-$79* SMI/SUD or Experiencing | WPC Enrollees | -$79* $365* $444* | Homelessness Control Group | -$139* $258* $397* | $47* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: Includes ED visits that do not result in hospitalization. * Denotes p<0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). SMI/SUD is serious mental illness or substance use disorder. Lower Cost |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research eer) Health Economics and Evaluation Research Program Estimated Payments for Residual Medi-Cal Payments UCLA estimated Medi-Cal payments for all residual services paid by Medi-Cal (apart from dental services) not included in the previous service categories. The residual categories include home health, dialysis, hospice, laboratory, radiology, therapy (e.g., physical, occupational, speech, respiratory), non-institutional residential care (e.g., mental health), among others. The use of such services may have increased due to care coordination and unmet need. Exhibit 152 shows that estimated residual Medi-Cal payments increased during WPC by $157 and $159 for WPC enrollees and controls, respectively. During WPC, the cost of residuals continued to increase for enrollees as slower rate ($12 per beneficiary per year), but declined for controls (-$37). The change in annual estimated payments for residual Medi-Cal payments from before WPC to during WPC declined significantly less for WPC enrollees than the control groups by $50 (DD). While this change in trend compared to controls was present for both groups of WPC enrollees, it was greater among SMI/SUD/HML enrollees ($63 per beneficiary per year) than MC/HR enrollees ($17; data not shown). Exhibit 152: Trends in Estimate Medi-Cal Payments for Residual Medi-Cal Before and During WPC, PY 2- PY6 + 2 + 5 d oO 38 6 + 0 = = = x N, a a a o s 3 8 Bn e 5 no bo 90 0 re PJ 3 n 5.8 aun a x nN x S SR a wn B 5 | [ | Pre-Year 2 Pre-Year 1 Year 1 Year 2 Year 3 Year 4 Year 5 Before WPC During WPC @WPCEnrollees Control Group g Betwee D raelen Lp During WP Ey: Mute PD WPC Enrollees $157* $12* -$145* Control Group $159* -$37* -$196* $50* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes p<0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). The residual categories include home Whole Person Care Final Evaluation Report | Lower Cost err UCLA Center for Health Policy Research Health Economics and Evaluation Research Program health, dialysis, hospice, laboratory, radiology, therapy (e.g., physical, occupational, speech, respiratory), non-institutional residential care (e.g., mental health), among others. UCLA examined at the descriptive breakdown of residual estimated Medi-Cal payment before and during WPC. The proportion of residual payments that resulted from hospice care, community-based adult services, therapy services, and home health services increased from before to during WPC for WPC enrollees. PLU Lower Cost | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Chapter 13: WPC Services and Outcomes for Enrollees Experiencing Homelessness All 25 WPC Pilots provided some form of housing and supportive services to enrollees, either directly, through partner organizations, or through linkages within the community. This chapter addresses the following evaluation question: “To what extent did the Pilot increase access to housing and supportive services and improve housing stability, if applicable?” In addition to addressing this question, this chapter includes data on characteristics of enrollees experiencing homelessness and Pilot-reported metrics relevant to this population. Furthermore, UCLA provides updated information since the interim report on strategies used by Pilots to identify and outreach to individuals experiencing homelessness, track and retain these enrollees, and leverage alternative funding sources to provide them with housing or housing support. This chapter also provides additional data since the interim report on specific types of housing and supportive services offered by WPC Pilot and their partners, with and without WPC funding. Data sources for this chapter include PY 3 and PY 5 LE surveys, as well as PY 6 follow-up interviews with leadership and frontline staff. Additional qualitative data around challenges and solutions was provided in 25 WPC mid-year and annual narrative reports. Characteristics of enrollees experiencing homelessness and housing outcomes were obtained from enrollment and utilization reports from 25 Pilots and Medi-Cal enrollment and claims data. For additional detail on data sources and methodology, please see Appendices C, D, E, and F. Quantitative data sources for this chapter included Quarterly Enrollment and Utilization Reports from PY 2 to PY 6 and Medi-Cal enrollment and claims data. UCLA used the Quarterly Enrollment and Utilization Reports to identify enrollees experiencing homelessness, their dates of enrollment, and patterns of enrollment. UCLA also used Medi-Cal claims data, which included both managed care and fee-for-service encounters, to construct WPC metrics per the WPC Technical Specifications. UCLA used the Annual WPC Variant and Universal Metric Reports submitted by Pilots to DHCS from baseline to PY 6 to report on three variant metrics on housing, calculated by Pilots based on administrative data. UCLA reported a weighted average rate for the available metrics across all Pilots that reported each metric. For additional detail on data sources and methodology please see Appendices A and B. Whole Person Care Final Evaluation Report | WPC Services and Outcomes for Enrollees Experiencing [RZ5) Homelessness eee 6 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Approaches to Enrolling and Delivering Housing Support Services to Individuals Experiencing Homelessness and At-Risk-Of-Homelessness Populations As detailed in the interim report, in PY 3 surveys, Pilots rated increasing enrollee access to housing support services (e.g., housing navigation, tenancy support) as a relatively high priority (8.7 of 10). Although all Pilots reported providing WPC services to at least some individuals experiencing homelessness, 15 Pilots explicitly identified individuals experiencing homelessness as a primary target population. Nine Pilots also chose individuals at-risk-of-homelessness as a primary target population. Monterey and San Francisco solely focused on individuals experiencing homelessness and no other target populations. Identification of Individuals Experiencing Homelessness Pilots utilized various methods for determining if a prospective enrollee was experiencing homelessness or at-risk for homelessness. In PY 5 surveys, Pilots most often reported utilizing a standardized tool, such as the Vulnerability Index - Service Prioritization Decision Assistance Tool (VI-SPDAT), or a definition, such as the United States Department of Housing and Urban Development (HUD), to assess enrollee homelessness or risk of homelessness (14 of 25). Eight Pilots reported receiving data or assessment(s) from another source (e.g., Homeless Management Information System (HMIS), hospitals/EDs, coordinated entry system (CES), continuum of care (COC), partner referrals). Five Pilots reported use of a Pilot modified version of a standardized tool/definition to assess homelessness and risk. Outreach to Individuals Experiencing Homelessness In bi-annual narrative reports and PY 6 follow-up interviews, Pilots discussed their approaches to engaging and maintaining communication with individuals experiencing homelessness. Pilots highlighted significant challenges with outreach and engagement due to outdated or unavailable contact information, the transience associated with homelessness, and an unwillingness to engage with County services due to prior negative experiences. Successful approaches to outreach included in-person communication through visits to homeless shelters or encampments and other areas where these populations gathered. Alameda, Napa, Riverside, Kings, and San Francisco had dedicated homeless outreach teams that worked primarily in the field. Several Pilots noted that efforts to locate individuals often required direct coordination with WPC partners and local organizations such as shelters, pLIcM WPC Services and Outcomes for Enrollees Experiencing Homelessness | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research . “ December 2022 Health Economics and Evaluation Research Program churches, and police departments. Pilots emphasized the importance of consistency and trust building when working with individuals experiencing homelessness; these efforts were key to establishing rapport, which led to successful enrollment and retainment in WPC. Outreach strategies were adjusted to account for COVID-19 response, and some benefits were recognized with individuals receiving short-term housing and supportive resources in a single location with efforts such as Project Roomkey. “| think that one of the things that we do on the Homeless Outreach Team is... take each interaction as a separate interaction, so if Case Manager hasn't been successful building a connection and rapport with a client, he doesn't say, well, | tried five times, it didn't work. He goes out and tries it 50 times and eventually it will almost always work, where you can engage and build trust.” - Marin “Our onsite presence at the shelters has afforded us the opportunity to successfully outreach to, and ultimately enroll in many cases, some of the most vulnerable, transient and hard to reach beneficiaries of our target populations” -Kern Selected examples of WPC outreach and engagement activities for individuals experiencing homelessness are outlined in Exhibit 153. Exhibit 153: Selected Examples of Outreach Approaches for Individuals Experiencing Homelessness in WPC WPC Pilot Selected Examples Alameda “Street Health” outreach teams visited encampments, community partners, and medical providers and referred prospective enrollees to WPC. Prior to enrollment, case managers dedicated time to build trust, identify basic barriers to services that could be addressed (e.g., transportation), and delineate goals. “Street Health” included a street psychiatry outreach program comprised of a psychiatrist, a nurse case manager, and a community outreach worker; who conducted psychiatric evaluations and administered medication and substance use disorder treatment to individuals in homeless encampments. Alameda also utilized their 211 call center as a method for identifying individuals seeking housing resources. Kern Kern maintained a presence in shelters for continuous outreach and engagement. Co-location and the use of a peer support specialist (i.e., ability to build trust and rapport with people experiencing homelessness based on lived experience) were strategies identified as fundamental to successful engagement. Monterey Monterey primarily identified individuals experiencing homelessness through outreach at shelters, encampments, and healthcare facilities, as well as through referrals from partner organizations. Teams of public health and licensed vocational Whole Person Care Final Evaluation Report | WPC Services and Outcomes for Enrollees Experiencing 7 Homelessness UCLA Center for Health Policy Research December 2022 - ; Health Economics and Evaluation Research Program WPC Pilot Selected Examples nurses would actively outreach throughout the county, specifically targeting areas with the highest concentration of individuals experiencing homelessness. Napa Enrollees were identified through referrals from various organizations and partners, including healthcare clinics, police and fire departments, and shelter systems. Outreach was conducted in shelters and through street-engagement by a multi- disciplinary team. Outreach teams performed initial intake assessments, enrolled individuals, and entered them into the county’s coordinated entry system. Riverside Riverside’s homeless outreach teams were responsible for connecting homeless individuals to social support services and acquiring basic documentation needed to apply for Medi-Cal, and subsequently enroll into WPC. Riverside also had WPC Housing Navigators in the coordinated entry system to help with housing access for WPC enrollees. San Francisco San Francisco identified and auto-enrolled beneficiaries using a data-driven approach within their coordinated care management system records. New enrollments and engagement occurred when staff of the county’s Homeless Outreach Team or Street Medicine and Shelter Health programs met with and enrolled previously unidentified individuals experiencing homelessness. WPC staff co-location within the County’s extensive shelter system provided an opportunity for consistent and meaningful engagement of enrollees. | Sources: PY 6 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=26), June-September 2021 and WPC Mid-Year and Annual Narrative Reports, PY 2 (2017) - PY 6 (2021). Housing Support Services In PY 5, all but one Pilot reported providing one or more housing related service either through the LE or through partner organizations (Exhibit 154). Housing support services (e.g., tenancy support, completing applications for the coordinated entry system, supporting housing search, or obtaining housing funds) were most often provided by partner organizations using WPC funds (21 of 25 Pilots) or by partner organizations using alternative funding sources such as Housing and Disability Advocacy Program (HDAP) funds (16). Direct assistance with housing search (e.g., finding available temporary or permanent housing stock) was the most common service provided by partner organizations (19). Ten LEs provided housing support services in-house using WPC funds, with the most common service involving assistance completing applications for the coordinated entry system (8), followed by tenancy support (e.g., counseling and training individuals to move in or remain in temporary or permanent housing; 7). pLESa WPC Services and Outcomes for Enrollees Experiencing Homelessness | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program Exhibit 154: Type of Housing Support Service(s), Provided by Lead Entity or WPC Partner Organization, Using WPC Funds or an Alternative Funding Source, PY 5 Any housing support service(s) Provided assistance with housing search Completed applications for coordinated entry Type of housing support service Provided tenancy support 21 16 10 19 11 6 16 Provided assistance obtaining housing funds 13 6 15 12 8 17 10 7 Provided by WPC partner with WPC funds m Provided by WPC partner with alternative funds Provided by LE with WPC funds Source: PY 5 Lead Entity (LE) Survey (n=25), June-August 2020. Notes: Tenancy support includes counseling and training individuals to move in or remain in temporary or permanent housing; housing search includes finding available temporary or permanent housing stock; assistance with obtaining housing funds includes assistance with housing choice vouchers or rental subsidies. Direct housing resources and services (e.g., funds for security deposit, home items, utilities, or housing improvements; landlord incentives, medical respite, motel vouchers, short- or long- term housing) were provided by nearly all Pilots using WPC (22) and alternate (21) funds. Most LEs relied on partner organizations to provide these services, although over half of LEs also provided at least some of these services in-house (14; Exhibit 155). Partner organizations most often used WPC funds to provide ongoing assistance with enrollee- landlord relationships after enrollees were housed (18). LEs most often directly provided motel vouchers (8), medical respite (7), and short-term housing stays (7). Whole Person Care Final Evaluation Report | WPC Services and Outcomes for Enrollees Experiencing 9 Homelessness UCLA Center for Health Policy Research D ber 2022 . 2 Ls Health Economics and Evaluation Research Program Exhibit 155: Type of Direct Housing Services and Resources Provided by Lead Entity or WPC Partner Organization, Using WPC Funds or an Alternative Funding Source, PY 5 | Pe 8 Any direct housing service(s) and/or resource(s) 14 12 Funds for furniture, appliances, or other home items 16 15 10 Funds for security deposit 16 3 a5 5 a 8 Funds for utilities 16 q I5 6 Funds for housing improvements for specific health needs 17 15 18 a Ongoing assistance with enrollee-landlord relationships 9 = 3 € S 7 ‘B__ Legal support for issues related to housing/tenancy issues 13 2 I 3 S s 5 e 5 Landlord incentives 12 5 4 10 wo Provide medical respite to homeless 2 Z £ 8 £ Provide motel vouchers or equivalent 14 2 8 2 vf “ Provide short-term housing in a shelter 18 17 pe wes | “ es53 Provide permanent, long-term housing TS 16 3838 2 Provided by WPC partner with WPC funds ™@ Provided by WPC partner with alternative funds m= Provided by LE with WPC funds Source: PY 5 Lead Entity (LE) Survey (n=25), June-August 2020. Notes: Funds for housing improvements for specific health needs (e.g., accessibility ramp); landlord incentives (i.e., prior to enrollee move-in to encouraging renting to WPC enrollees). WPC funds could not be used for direct housing/to provide permanent, long-term housing (e.g., pay rent). pAst0 WPC Services and Outcomes for Enrollees Experiencing Homelessness | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program “lf we're going to be working with a client after they get housed... we try to get a release of information. So that we can work with that landlord and figure out what's going on, what's working, what's not working, if they're not paying their rent, the landlord can usually notify us, and we (WPC) can help with that... And... it can [help] avoid them failing out of housing.” -Placer “The recuperative care program ... provides a safe place for clients, the homeless clients who are transitioning from hospitalization... they would be discharged to the street, but they need a safe place to recuperate... [With recuperative care] these clients have a place, at least for 30 days, to recuperate after they have been discharged from hospital so that they are not on the street post hospitalization. And... they have a case manager that checks on them to ensure that they are able to recover safely.” -San Mateo Whole Person Care Final Evaluation Report | WPC Services and Outcomes for Enrollees Experiencing Homelessness err UCLA Center for Health Policy Research Health Economics and Evaluation Research Program In PY 5, nearly all Pilots (23) promoted a "Housing First" approach in which provision of permanent housing was prioritized (i.e., persons experiencing homelessness were not required to address behavioral health problems or graduate from other service programs before accessing housing; Exhibit 156). Over half of Pilots (15) participated in streamlining processes or program restructuring around delivery of housing services, while slightly fewer (12) participated in streamlining processes or programs that affected financing of housing services and/or promoting policy and legislation to increase housing availability. Eight Pilots engaged in activities related to workforce training of housing navigation and/or co-location of housing services with other service programs. Exhibit 156: Pilot Participation in Activities to Promote Community, Policy, and/or Systems Change Related to Homeless Assistance, PY 5 Utilized a "housing first" approach 23 Streamlined services around delivery of housing related Ld ‘is services Steammeemencnyeronne cerns Eso Promoted policy and legislation to increase housing Ll 2D availability Offered workforce training in housing navigation Co-located housing services with other services a 8 Source: PY 5 Lead Entity (LE) Survey (n=25), June-August 2020. viva WPC Services and Outcomes for Enrollees Experiencing Homelessness | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program “In order to really achieve health and wellness, you do have to have the base of Maslow's hierarchy in place... we've had housing programs for a long time, but really the health programs and the housing programs had never really been in the same sandbox... So [now] looking at how some of the medical services are delivered... they really have embraced a housing first approach... There's more understanding about the barriers that inhibit or prohibit people from accessing or keeping appointments... the nature of what people are experiencing when they're living unsheltered or without a stable home.” -Shasta Tracking and Retention Given the transience associated with homelessness and difficulty in maintaining contact post- WPC enrollment, tracking and retention efforts required collaboration with partners. In PY 3 surveys, LEs reported on the degree of buy-in for data sharing among partners on a scale of zero (very low) to ten (very high). Out of all partner types (e.g., health plans, hospitals, mental health providers), LEs identified housing providers as having the highest buy-in at a mean of 7.7 of 10 (data not shown). In PY 5 surveys, 20 LEs reported participation in direct collaboration activities with a housing agency as a part of WPC (Exhibit 157). Over half of LEs (13 of 25) had established universal consent forms or other data sharing agreements with housing agencies (e.g., MOUs, BAAs). Ten LEs participated in a coordinated assessment system with a housing agency to identify and prioritize high-risk/high-need patients for receipt of housing services. Whole Person Care Final Evaluation Report | WPC Services and Outcomes for Enrollees Experiencing PS} Homelessness err UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 157: Participation of Lead Entity with Housing Agency in Select Collaboration Activities, PY'S Utilized a universal consent form for data sharing with Po 2 housing agency Established data sharingesreements with housing agency i 2; Participated in a housing coordinated entry/assessment Ld 10 system shared care coordinators with showing cen I Physical co-location of housing agency staff to facilitate ae 5 access to services and/or resources Source: PY 5 Lead Entity (LE) Survey (n=25), June-August 2020. “And that (flexible housing) pool does not pay for rent, but it does pay for application fees, furniture, deposits, which really help get the enrollee into housing and not like just alone. And it's not a lot, most often the funds pay for, again, a deposit, an application fee, first month's rent, a mattress, and some toilet paper, but it's something. And I think that's a huge part of retention from my perspective. ... Since we increased it in October, some housing partners are saying, well, can we go back and actually apply those funds to retention purposes? So let's go back and see our folks who were housed, do they need some cooking utensils, can we do that to help keep them in their housing?” —Sacramento Specialized Housing Staff in Care Coordination Teams In PY 5 surveys, 20 Pilots reported use of housing navigators to provide care coordination (16), clinical consultation (13), and/or enrollee outreach (10). Eight Pilots also used housing navigators in a supervisory role (data not shown). In follow-up interviews, Pilots indicated that inclusion of dedicated housing staff and particularly peer support staff as part of the care coordination team was essential to effectively engaging enrollees experiencing homelessness in care. In PY 5 surveys, nearly all (22) LEs yA) WPC Services and Outcomes for Enrollees Experiencing Homelessness | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program reported the use of housing support specialists, many of whom had previous lived experience of homelessness or risk of homelessness to provide housing and supportive services for WPC enrollees. Selected examples of approaches to inclusion of specialized housing staff in WPC are provided in the interim report, in “Chapter 13: Homeless WPC Enrollee Services and Outcomes”. “The staff, they have to be a good listener. They have to be aware of their surroundings. They have to be empathetic. If someone said, ‘I don't want to be bothered today.’ They had to take that and say, ‘Okay, | understand, can we try again tomorrow?’ Back away from them. Give them a chance to get to know you and trust you and that's the basis of working with this population. And you find out that they start to call you and depend on you more and more and more if you want to treat them like you want to be treated, whether they have alcohol and drug problems or whether they're mentally ill, you still want to treat them with respect. That's the biggest thing is treating them a respect and like human beings and so this way you're going to be successful .” - Monterey Whole Person Care Final Evaluation Report | WPC Services and Outcomes for Enrollees Experiencing Pes) Homelessness err UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Enrollment Patterns and Characteristics of WPC Enrollees Experiencing homelessness Under WPC, Pilots were required to identify enrollees experiencing homelessness in their quarterly WPC Enrollment and Utilization Reports, regardless of whether or not they were a target population. UCLA used the homeless indicator to provide a profile of these enrollees. Of the 247,887 enrollees in WPC, 124,414 (50 %) were identified as experiencing homelessness. However, some Pilots reported difficulties in obtaining this data and therefore the number of these enrollees may be under reported. Enrollment Patterns and Size Exhibit 158 shows the unduplicated enrollment of WPC enrollees experiencing homelessness by month. The cumulative enrollment of these enrollees increased from 25,752 at the end of PY 2 to 124,414 at the end of PY 6. Total enrolled as of December 2021 was 50,610. Exhibit 158: Unduplicated Monthly and Cumulative Total WPC Enrollment among Enrollees Experiencing Homelessness, January 2017 to December 2021 8 Monthly Current Enrollment 82,180 124,414 ——= Cumulative Enrollment 105,366 45,645 50,610 | | AISA SISISIAISISIB AIS SI S/S/S Sl 48/RIB8 & g 4 a PY2 PY3 py4 PYS PY6 Source: Whole Person Care Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 124,414 unique individuals. Excludes individuals who received outreach or other WPC services but did not enroll. yA) WPC Services and Outcomes for Enrollees Experiencing Homelessness | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program Exhibit 45 shows the total, unduplicated WPC enrollment of enrollees experiencing homelessness through PY 6 by Pilot, indicating none in Sonoma and a high of 56,413 enrollees in Los Angeles. Three Pilots had counts over 10,000 and eight had counts over 1,000. Exhibit 159: Total Unduplicated Enrollment in WPC by Pilot among Enrollees Experiencing Homelessness, December 2021 Los AngeleS Same 5, |‘) 3 San Francisco A 27,749 Orange A 13,861 Alameda EEE «8,258 Riverside mmm 3,847 SanJoaquin mm 3,016 Contra Costa mmm 2,950 Sacramento mmm 2,314 Santa Clara mm 2,195 San Mateo mm 1,527 Marin mm 1,326 Ventura m@ 975 Kern m 881 Monterey = 801 Napa = 733 Kings #591 San Diego = 568 Shasta = 567 Placer » 489 Santa Cruz 1 319 Mendocino 1 290 Solano | 162 SCWPCC 90 San Bernardino 56 Sonoma 0 Source: Whole Person Care Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 124,414 unique individuals. Excludes individuals who received outreach or other WPC services but did not enroll. SCWPCC is the Small County Whole Person Care Collaborative. Whole Person Care Final Evaluation Report | WPC Services and Outcomes for Enrollees Experiencing Homelessness eee 6 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 160 shows the percent of total WPC enrollees experiencing homeless by Pilot. Among Pilots that had selected homelessness or at-risk-of-homelessness as their only primary target population, all or most (96% in Monterey and 95% in Napa) were experiencing homelessness. However, there was significant variation among Pilots with homelessness as one of their primary target populations and those that had not selected this population as a target. Exhibit 160: Percent of WPC Enrollees Experiencing Homelessness by Pilot, January 2017 to December 2021 ES yo we KK LK vo S ss 8 SB & 0 asangartoan a x 2 g © 2 XS 8 x S 8x xy 2 8 ® Sa 8 8 “ nw fh x gx 5 x x mR ma ON 2 Ss 3 x x i Pine: -__= ory 8 wos Ee YEU OE BD BRP OBROARHK OR HY GO eee SEER PST ES EEE ESE EEE E e224 e¢¢2 8 MP SSO~*~ EC FoF S*e FSsUYUEGCE & § Oo § S& Ge 8g)" aoe £ gS os & is 5 ca a 8 a4 5 cegcezEE 5 8 8 4 = ona 5 2 8 Og S a Onl Targeted Homeless and Other Groups Did Not Target Homeless Specificall Y y Targeted Homeless Source: Whole Person Care Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 124,414 unique individuals. Excludes individuals who received outreach or other WPC services but did not enroll. SCWPCC is the Small County Whole Person Care Collaborative. Sonoma County did not report on homelessness but did identify 14% of their enrollees in the homeless target population. yA) WPC Services and Outcomes for Enrollees Experiencing Homelessness | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program Exhibit 47 displays the length of enrollment among WPC enrollees experiencing homelessness through PY 6. Enrollees experiencing homelessness were most commonly enrolled for 1-6 months (37%). The mean, median, and mode length of enrollment in the program for enrollees experiencing homelessness was 15, 10, and 1 months, respectively (data not shown). Exhibit 161: Length of Enrollment in WPC Among Enrollees Experiencing Homelessness, January 2017 to December 2021 37% 20% 14% 8% 6% 5% 4% im 2% 2% 2% a O | = i 16 7-12 13-18 19-24-2530 31-36) 37-42 43-48 = 49-54 55-60 months months months months months months months months months months Source: Whole Person Care Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 124,414 unique individuals. Excludes individuals who received outreach or other WPC services but did not enroll. Includes enrollees who enrolled at two Pilots without cross enrollment. Whole Person Care Final Evaluation Report | WPC Services and Outcomes for Enrollees Experiencing 9 Homelessness UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Demographics Of the 124,414 total enrollees experiencing homelessness, 119,912 (96%) were Medi-Cal enrollees during their two years prior to WPC enrollment and described in Exhibit 162. The majority of these enrollees were male (64%), ages 50-64 (34%), White or Black (28%), and primarily communicated in English (92%). Exhibit 162: Demographics of WPC Enrollee Experiencing Homelessness 1% Age at enrollment 28% 30% 34% (S9 0-17 18-34 35-49 50-64 M65+ Gender Cry ELS = Male & Female 1% 7% 1% Race/Ethnicity @ White B Hispanic m Black M Asian @ American Indian or Alaska Native @ Hawaiian or Other Pacific Islander Other m Unknown Primary Communication [547275 EPs Language @Spanish Other M English Source: Medi-Cal enrollment data from January 2015 to December 2021 and Quarterly Whole Person Care Enrollment and Utilization Reports from PY 2 to PY 6. Notes: Overall enrollee population includes 125,331 enrollees who were enrolled during PY 2 through PY 6 and had Medi-Cal enrollment data. All data are reported using Medi-Cal enrollment data during the 24 months prior to WPC enrollment. pA WPC Services and Outcomes for Enrollees Experiencing Homelessness |Whole Person Care Final Evaluation Report Health Status UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 Analyses of Medi-Cal claims show that enrollees experiencing homelessness most often had hypertension (34%), depression (41%), and drug use disorders (41%; Exhibit 163). Other mental health conditions such as depressive disorders (38%), anxiety disorders (35%), and schizophrenia and psychotic disorders (32%) were also common Exhibit 163: Proportion of WPC Enrollees Experiencing Homelessness with Chronic Conditions Substance Mental Health Use Conditions Physical Health Conditions Conditions Hypertens) 34% Rheumatoid arthritis/ osteoarthritis Diabetes Chronic Kidney Disease Anemia Chronic Obstructive Pulmonary Disease Hyperlipidemia Asthma Pneumonia Depression Depressive Disorders Anxiety Disorders Schizophrenia and Other Psychotic Disorders Bipolar Disorder Drug Use Disorders Tobacco Use Alcohol Use Disorders 19% 16% 16% 16% 16% 15%, es 14% ME «10% a 41% es 38% es 35% Se 32% 27% a 41% es 27% Ss 27% Source: Medi-Cal enrollment and claims data from January 2015 to December 2021 and Quarterly Whole Person Care Enrollment and Utilization Reports from PY 2 to PY 6. Notes: Enrollee population includes 119,911 enrollees who were enrolled during PY 2 through PY 6 and had Medi- Cal enrollment and claims data. Chronic and disabling conditions were determined using algorithms developed by the CMS Chronic Conditions Data Warehouse (CCW). Conditions with at least 10% prevalence were reported. Whole Person Care Final Evaluation Report | WPC Services and Outcomes for Enrollees Experiencing Homelessness eee 6 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Estimated WPC Service Use and Cost Using WPC Quarterly Enrollment and Utilization Reports, Exhibit 164 shows the proportion of WPC enrollees experiencing homelessness and not experiencing homelessness that received different specific WPC services. The rates of receipt of outreach (75% vs 70%), care coordination (88% vs. 91%), housing support (68% vs. 72%), benefit assistance (81% vs 76%), transportation (61% vs 64%), and legal services (69% vs 68%) was similar between enrollees experiencing homelessness and not experiencing homelessness. However, enrollees experiencing homelessness more frequently received re-entry services and medical respite and less frequently received employment assistance and health education. Exhibit 164: Proportion of WPC Enrollees Experiencing Homelessness and Not Experiencing Homelessness That Received WPC Services, PY 2 to PY 6 RR x x 8a 2 ax i x8 % © Rx rs =a aed e B & oe x Ss s os & & & a ss g x = x 7 N Rye aS aa a 8 n x z s s @ @ s & Ss J & & Ss ss SS se S (co e 3 é @ 3 é & s aS < Pas s ¢ S ES < g ge ee ee & Ss e & so & Ss Experiencing Homelessness _ ml Not Experiencing Homelessness Source: WPC Quarterly Enrollment and Utilization Reports (n=25), PY 2 to PY 6. Notes: Includes 132,925 individuals with enrollment in WPC identified as experiencing homelessness and 115,674 individuals with enrollment in WPC not identified as experiencing homelessness. Service estimates indicates that the enrollee received a fee-for-service intervention or per-member per-month intervention bundle that included the service, but does not guarantee individual use of that service. The average cost of services received by enrollees experiencing homelessness was $8,481 and higher than $3,798 estimated for enrollees not experiencing homelessness (data not shown). Furthermore, the average cost of services per month was $407 for enrollees experiencing homelessness compared to $267 for enrollees not experiencing homelessness. viva) WPC Services and Outcomes for Enrollees Experiencing Homelessness | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Trends in Pilot-Reported Housing Metrics To assess housing services UCLA calculated the weighted average rates across Pilots for three housing services variant metrics (Exhibit 165). These metrics were not available for Pilots that lacked sufficient data due to data sharing issues did not enroll individuals experiencing homelessness, or did not deliver services to those enrolled in a given reporting period. See Appendix B for further details on reporting for each metric. Exhibit 165: Housing Metrics Selected by WPC Pilots Universal Metric Name Description Baseline | Reporting | Numbers | Improvement vs. Variant | and Number Year Years of Pilots | measured by Reporting | Increase or by Year Decrease Variant Permanent PH: Percent of Pr PY 3,.PY 4inPY2 Increase Housing (PH) homeless who were 4,PY5, 9in PY3 permanently housed PY6 11linPY4 longer than 6 12in PYS consecutive months’ 11in PY6 experience of permanently housed 8Variant Housing Services | HS: Percent of PY2 PY.3,:P¥ 12 in PY2 | Increase (HS) homeless who received 4,PYS, 13 in PY3 housing services after PY6 15 in PY4 being referred for 16 in PYS housing services 14in PY6 Variant Supportive SH: Percent of PY2 PY 3, PY 6 in PY2 Increase Housing (SH) homeless who received 4,PY5, 6in PY3 supportive housing PY6 7inPY4 after being referred for 8inPYS supportive housing Gin PY6 Source: PY 1 (baseline), PY 2, and PY 3 Annual WPC Variant and Universal Metric Reports and Whole Person Care Universal and Variant Metrics Technical Specifications (March 22, 2019). Variant Metric: Permanent Housing Twelve WPC Pilots elected to report the percentage of enrollees experiencing homelessness who were permanently housed and reached seven months of permanent housing (PH) during the measurement year. The overall PH rate decreased slightly from 99% in PY 2 to 94% in PY 3 before increasing to back to 99% in PY 5 (Exhibit 166). The PH rates varied by Pilot with differences as low as 5% and as high as 100% in PY 3. One large Pilot represented between 82% and 95% of the enrollees in the denominator each year and had a very high success rate. The PH rate was lower for the remaining Pilots. Without this influential Pilot, the PH rates were Whole Person Care Final Evaluation Report | WPC Services and Outcomes for Enrollees Experiencing } Homelessness err UCLA Center for Health Policy Research Health Economics and Evaluation Research Program lower during PY 3 at 50% and between 85% and 89% during the other reporting years (data not shown). Exhibit 166: Proportion of Enrollees Formerly Experiencing Homelessness in Permanent Housing Who Reached the Seventh-Month, by Program Year 100% 100% 100% 100% 100% 99% obs 98% 99% 98% oi) an er) 5% 6% Baseline PY 3 (n=4,991) PY 4 (n=8,727) PY 5 (n=12,202) PY 6 (n=14,741) (PY 2, n=2,053) @ Overall WPC @ Pilot-Specific Minimum @ Pilot-Specific Maximum Sources: WPC Annual Universal and Variant Metric Reports, baseline through PY 6 Notes: Only Pilots that reported on this metric were included in the analysis. The number of Pilots reporting varied by year. Appendix B, Exhibit 10 provides details on which Pilots reported in each year. The denominator size is shown as sample size per year. Bars represent the range reported by Pilots, with minimum being the lowest rate reported by a Pilot and maximum being the highest rate reported by a Pilot. pL WPC Services and Outcomes for Enrollees Experiencing Homelessness | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program Variant Metric: Housing Services A subset of 16 WPC Pilots elected to report the metric that measured proportion of enrollees experiencing homelessness who received housing services after being referred for housing services (HS). One Pilot was excluded from the analysis due to differences in their denominator methodology. The overall HS rate increased from 47% in PY 2 to 78% in PY 5 before declining to 61% in PY 6 (Exhibit 167). There was large variation in HS rates by Pilot, ranging from a low of 0% to a high of 100% in PY 5. Overall, the number of individuals receiving housing services each year ranged from 525 in PY 2 to 7,032 in PY 5 (including data from the Pilot that was excluded from the rate analysis; data not shown). Exhibit 167: Proportion of Homeless Enrollees Who Received Housing Services After Being Referred for Housing Services, by Program Year 100% 100% 100% 100% 100% | ehh er) Xe Ga 47% Bt) 36% 24% 12% 0% Baseline PY 3 (n=3,347) PY 4 (n=5,026) PY 5 (n=5,663) PY 6 (n=4,423) (PY 2, n=952) Overall WPC = @Pilot-Specific Minimum _@ Pilot-Specific Maximum Sources: WPC Annual Universal and Variant Metric Reports, baseline through PY 6 Notes: Only Pilots that reported on this metric were included in the analysis. The number of Pilots reporting varied by year. Appendix B, Exhibit 11 provides details on which Pilots reported in each year. The denominator size is shown as sample size per year. Bars represent the range reported by Pilots, with minimum being the lowest rate reported by a Pilot and maximum being the highest rate reported by a Pilot. These data exclude one large Pilot that included all enrollees in the denominator rather than only those referred for housing services, leading to reported rates of 1% to 22%. The inclusion of this Pilot would have led to a WPC rates of 6% in PY 2 and 36% in PY Se Whole Person Care Final Evaluation Report | WPC Services and Outcomes for Enrollees Experiencing Ss Homelessness eee 6 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Variant Metric: Supportive Housing A subset of 8 WPC Pilots elected to report the percentage of homeless enrollees who received supportive housing after being referred for supportive housing (SH). One Pilot was excluded from the rate analysis due to differences in their denominator methodology. The overall SH rate varied from year to year, with rates consistently below the baseline rate of 42% in PY 2 (Exhibit 168). There was variation in SH rates by Pilot, ranging from a low of 0% to a high of 100% in some years. One Pilot represented between 63% and 87% of the enrollees in the denominator each year and had a very low success rate. The SH rate was higher for the remaining Pilots. Without this influential Pilot, the SH rates started at 51% in PY 2 and increased to 85% in PY 5 before declining to 28% in PY 6 (data not shown). Overall, the number of individuals receiving housing services each year ranged from 399 in PY 2 to 2,756 in PY 5 (including data from the Pilot that was excluded from the rate analysis; data not shown). Exhibit 168: Proportion of Homeless Enrollees Who Received Supportive Housing after Being Referred, by Program Year 100% 100% 100% 100% 100% ° % ° ¢ ¢ a2 20% av_ RO Ba ae 0% 4% 4% 3% 0% Baseline PY 3 (n=1,967) PY 4 (n=3,540) PY 5 (n=10,160) PY 6 (n=9,987) (PY 2, n=201) Overall WPC @Pilot-Specific Minimum —_@ Pilot-Specific Maximum Source: PY 2 Annual, and PY 3 Annual WPC Variant and Universal Metric Reports. Notes: Only Pilots that reported on this metric were included in the analysis. The number of Pilots reporting varied by year. Appendix B, Exhibit 12 provides details on which Pilots reported in each year. Bars represent the range reported by Pilots, with minimum being the lowest rate reported by a Pilot and maximum being the highest rate reported by a Pilot. These data exclude one large Pilot that included all enrollees in the denominator rather than only those referred for housing services during PY 2 and PY 3, leading to reported rates of 4% and 7%, respectively. The inclusion of this Pilot would have led to overall WPC rates of 5% in PY 2 and 37% in PY 5. pl-tci WPC Services and Outcomes for Enrollees Experiencing Homelessness | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ee err) Health Economics and Evaluation Research Program Comparison of Adjusted Trends Between WPC Enrollees Experiencing Homelessness and their Controls, Before and After WPC Implementation UCLA measured trends in metrics before and during WPC for WPC enrollees that were experiencing homelessness and their matched controls to assess the impact of WPC on individuals experiencing homelessness. Because controls did not have reported homelessness by the Pilots, UCLA matched enrollees and their controls using a propensity score methodology that included a UCLA created indicator of homelessness. This indicator used both address-based and claims-based methods to identify individuals likely to be homeless. Metrics were based on the date of an individual WPC enrollee’s enrollment. UCLA examined changes in trends before and during WPC using a difference-in-difference (DD) analysis by modeling the changes in yearly increments up to 2 years (Pre-Year 1 and Pre-Year 2) before WPC enrollment and up to 5 years (Year 1, 2, 3, 4, and 5) during WPC. For these, the DD analysis measured the annual change from Pre-Year 2 vs. Pre-Year 1 for both WPC enrollees and the control group; the annual change during WPC from Year 1 to Year 5 for both WPC enrollees and the control group; and the difference between the changes in WPC enrollees vs. the control group from before to during WPC. Further details can be found in Appendix A. Whole Person Care Final Evaluation Report | WPC Services and Outcomes for Enrollees Experiencing Homelessness Reaper 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Health Service Utilization Ambulatory Care: Emergency Department Visits Ambulatory Care: Emergency Department Visits is a WPC universal metric that measures the rate of emergency department (ED) visits that do not result in hospitalization. UCLA reported this metric per 1,000 beneficiaries per year. The intended direction of the metric and DD is decrease. Exhibit 169 shows an increase in the number of ED visits before WPC by 384 visits per 1,000 beneficiaries per year for WPC enrollees experiencing homelessness and by 322 visits for their controls. During WPC, this rate declined by 264 and 130 visits per year for enrollees and controls, respectively. The declining trend from before to during WPC was significantly greater for enrollees compared to the control group by 196 visits (DD). Exhibit 169: Trends in Ambulatory Care: Emergency Department Visits per 1,000 Beneficiaries per Year among WPC Enrollees Experiencing Homelessness Before and During WPC, PY 2 - PY 6 S ” © S a 8 8. 3 o is es eA nt 3 + a vn © a st Nv D. a a 2g a Nn io 5 = a g 5 Tony Sas a R a ag a a | = 3 “A a Pre-Year 2 Pre-Year 1 Year 1 Year 2 Year 3 Year 4 Year S Before WPC During WPC m WPCEnrollees = Control Group amet) Alam Orla) 1-9 Difference yt eece li see oR Tae Between Changes Difference (DD) WPC Enrollees 384* -264* -649* Control Group 322* -130* -453* -196* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: Includes ED visits that do not result in hospitalization. * Denotes ps0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). PAS WPC Services and Outcomes for Enrollees Experiencing Homelessness | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ee err) Health Economics and Evaluation Research Program Inpatient Utilization is a WPC universal metric that measures the rate of acute inpatient care and services. UCLA reported this metric per 1,000 beneficiaries per year. The intended direction of the metric and DD is decrease. Exhibit 170 shows an increase in the number of hospitalizations before WPC by 184 and 173 stays per 1,000 beneficiaries per year for enrollees experiencing homelessness and their controls, respectively. During WPC, this rate declined by 71 stays for enrollees, while it declined by 34 stays for controls. The declining trend from before Inpatient Utilization to during WPC was significantly greater for enrollees compared to the control group by 48 stays (DD). Exhibit 170: Trends in Inpatient Utilization per 1,000 Beneficiaries per Year among WPC Enrollees Experiencing Homelessness Before and During WPC, PY 2 - PY 6 8 BE = 8 3 ° g a B 5 33 ‘ 8 + 8a + 2 es a. + I Pre-Year 2 Pre-Year 1 Year 1 Year 2 Year 3 Year 4 Year S Before WPC During WPC @ WPCEnrollees ™ Control Group Arla aera} Arle NOL) 4-9 Pye} yao fle ae oy ae Between Changes Difference (DD) WPC Enrollees 184* -71* -254* Control Group 173* -34* -206* -48* | Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes p<0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). Whole Person Care Final Evaluation Report | WPC Services and Outcomes for Enrollees Experiencing 9 Homelessness Reaper 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Mental Health Services UCLA calculated the number of mental health services per 1,000 beneficiaries per year as an optional measure of service utilization under HHP. There is no intended direction for this measure. Mental health services are likely to increase due to unmet need and increased access, but this use is likely to decrease once health needs are addressed. Exhibit 171 shows that mental health services were increasing prior to enrollment for WPC enrollees experiencing homelessness and their controls by 1,941 and 1,358 services per 1,000 beneficiaries per year, respectively. After enrollment, both groups had declining rates of mental health services by 1,096 and 806 services, respectively. The declining trend from before to during WPC was significantly greater for enrollees compared to the control group by 873 services (DD). Exhibit 171: Trends in Mental Health Services per 1,000 Beneficiaries per Year among WPC Enrollees Experiencing Homelessness Before and During WPC, PY 2 - PY 6 g 2 zg un © ga ° S a 9 = 4 o a 2 N Dw a A ra Q oo nt & eS 8 2 = g 8 w uw - oO x + 5 5 8 a “ 7 4 v ! I l i Pre-Year 2 Pre-Year 1 Year 1 Year 2 Year 3 Year 4 YearS Before WPC During WPC ™@ WPCEnrollees ™ Control Group Alam eur ly} Sele NAS Lay 4-3 yes} Taso LTS fle -M el DTT ae Between Changes Difference (DD) WPC Enrollees 1,941* -1,096* -3,037* Control Group 1,358* -806* -2,164* -873* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes p<0.05, a statistically significant difference. Mental health services were identified as services with a mental health procedure code. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). px( WPC Services and Outcomes for Enrollees Experiencing Homelessness | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ee err) Health Economics and Evaluation Research Program Substance Use Disorder Services UCLA calculated the number of substance use disorder (SUD) services per 1,000 beneficiaries per year as an optional measure of service utilization under WPC. There is no intended direction for this measure. Exhibit 172 shows SUD service use was increasing prior to enrollment for both WPC enrollees experiencing homelessness and their controls by 885 and 704 services per 1,000 beneficiaries per year, respectively, and then rates declined after enrollment by 160 and 246 services, respectively. Overall, the declining change in trend from before to during WPC was not significantly different for WPC enrollees compared to controls (DD). Exhibit 172: Trends in Substance Use Disorder Services per 1,000 Beneficiaries per Year among WPC Enrollees Experiencing Homelessness Before and During WPC, PY 2 - PY 6 5 2 D + a t+ o © a g a a X ” 5 q st og Bn Sy ow On 2 oo Ry ® ¢ 8 8 2 a ~ ot + + n = - a x o eS eS © © 5 wn nw Pre-Year 2 Pre-Year 1 Year 1 Year 2 Year 3 Year 4 Year S Before WPC During WPC @ WPCEnrollees m™ Control Group Nera} ble ASU) Pye) Phase seem ae DTT aa Between Changes Difference (DD) WPC Enrollees 885* -160* -1,044* Control Group 704* -246* -949* -95 | Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes ps0.05, a statistically significant difference. SUD services were identified as services with a SUD treatment procedure code or an NDC for pharmacotherapy. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). Whole Person Care Final Evaluation Report | WPC Services and Outcomes for Enrollees Experiencing [PZAl Homelessness eater 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Follow-Up After Hospitalization for Mental Illness Follow-Up After Hospitalization for Mental Illness is a WPC universal metric that measures the percentage of discharges for beneficiaries 6 years of age and older hospitalized for treatment of selected mental illness diagnoses who had a follow-up visit with a mental health practitioner at (1) 7-days or (2) 30-days. The intended direction of the metric and DD is increase. Exhibit 173 shows that the trends for 7-day follow-up was not changing before WPC for individuals experiencing homelessness. After enrollment, the WPC enrollees had higher rates of 7-day follow-up. However, there was no significant yearly change in 7-day follow-up during WPC and no significant difference in the yearly change from before to during when comparing enrollees and controls (DD). Exhibit 173: Trends in Follow-Up After Hospitalization for Mental Illness within 7 Days among Enrollees Experiencing Homelessness Before and During WPC for WPC Enrollees and the Control group, PY 2 - PY6 ¥ oy ¥ » a x 5 5 x 5S x S x a x 2 2 2 5 + 9 + 9 a x x x = v ¢ 2 q x z | L [ | L | | nn a a nN ” s in s s S S S s S o o o o o o o > > > > > > > 2 2 a a Before WPC During WPC @WPCEnrollees Comparison Group Nae) Yearly Change Difference Pytareece lies eX fee Taya aes Between Changes Difference (DD) WPC Enrollees 0.0% 0.1% 0.1% Control Group 0.0% 71.2% 71.3% 1.3% Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes ps0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). pupa WPC Services and Outcomes for Enrollees Experiencing Homelessness | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Exhibit 174 shows that trends for 30-day follow-up. Trends were similar to those seen at 7-days expect that controls had a significant declining yearly change during WPC. Exhibit 174: Trends in Follow-Up After Hospitalization for Mental Illness within 30 Days among Enrollees Experiencing Homelessness Before and During WPC for WPC Enrollees and the Control group, PY 2 - PY 6 < 5 x Fd x Be & Be ky & & ® ° 3 5 x | i | . a a 0 ” st in o S G G G S GS $ 3 $ $ $ Ss S > > > > > > > v v < < Before WPC During WPC @WPCEnrollees Comparison Group Arla AOU) i-} AT AOL 49 Difference Pye ee st foe ey yaaa Between Changes Difference (DD) WPC Enrollees -0.7% -0.9% -0.2% Control Group -0.6% -3.8%* -3.2% 3.0% Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes p<0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). Whole Person Care Final Evaluation Report | WPC Services and Outcomes for Enrollees Experiencing [JPZé! Homelessness Rete 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Initiation and Engagement of Alcohol and Other Drug Dependence Treatment Initiation of Alcohol and Other Drug (AOD) Dependence Treatment is a WPC universal metric measuring the percentage of adolescent and adult beneficiaries with a new episode of AOD dependence who initiated treatment through an inpatient AOD admission, outpatient visit, intensive outpatient encounter or partial hospitalization within 14 days of the diagnosis. The intended direction of this metric and DD is increase. For rates of initiation of AOD treatment among WPC enrollees experiencing homelessness and their controls, both enrollees and controls saw a significant increasing rate before WPC by 1.9% and significant declining rates during WPC by 0.9% and 0.7%, respectively (Exhibit 175). There was no significant difference between WPC enrollees and controls in their trends from before to during WPC (DD). Exhibit 175: Trends in Initiation of Alcohol and Other Drug Dependence Treatment among WPC Enrollees Experiencing Homelessness Before and During WPC, PY 2 - PY 6 x o 2 x ° x x x e we x x m x Ss a 9 s ROR a 8 N 8 N 8 m x Sx NN s a r = + NS L | | | L . N 4 a x ” + in fa fa a . rt ra . S S S S S S S oO vo oO ov vo ov vo > > > > > > > 2 2 a a Before WPC During WPC m WPC Enrollees Comparison Group Ameri) Tara) 1-9 Difference Difference-in- role ae TTT ase UTet ta LLLP) WPC Enrollees 1.9%* -0.9%* -2.7%* Control Group 1.9%* -0.7%* -2.6%* -0.2% Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes p0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). PYLE) WPC Services and Outcomes for Enrollees Experiencing Homelessness | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Engagement of AOD Dependence Treatment is a WPC universal metric that measures the percentage of adolescent and adult beneficiaries who initiated treatment and who had two or more additional services with a diagnosis of AOD within 30 days of the initiation visit. The intended direction of this metric and DD is increase. WPC enrollees had an increase in their rate of engagement of AOD dependent treatment during WPC. Exhibit 176 shows that trends in yearly rates of engagement in AOD treatment did not change for WPC enrollees either before WPC or during WPC. Comparatively, the controls had significantly declining rates year-to-year during WPC. WPC enrollees had a significantly greater change in year-to-year rates from before WPC to during WPC compared to the controls (2.8%; DD). Exhibit 176: Trends in Engagement of Alcohol and Other Drug Dependence Treatment among HHP Enrollees Experiencing Homelessness Before and During HHP by SPA, PY 2 - PY 6 ¥ 3x x x Bx Bs B xs 5 x 5 39s + 2 Sm +S +S + & a vt [ + + o = : qa | | | I N a a N om t+ uw . as i . é L ris s q o & c 5 S oO vu vu ov vo vo vo > > > > > > > 2 2 a a Before WPC During WPC m WPC Enrollees Comparison Group Ala NAOT) bla AOE 43 Pye) Ties sto (ole MV ae yy Between Changes Difference (DD) WPC Enrollees 0.6% 0.1% -0.5% Control Group 0.6% -2.7%* -3.3%* 2.8%* Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes ps0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). Whole Person Care Final Evaluation Report | WPC Services and Outcomes for Enrollees Experiencing [JPA Homelessness ery) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program All-Cause Readmission All-Cause Readmission is a WPC variant metric that measures the number of acute inpatient stays during the measurement year that were followed by an unplanned acute readmission for any diagnosis within 30 days for beneficiaries ages 21 and older. The intended direction of the metric and DD is decrease. Both WPC enrollees and controls experiencing homelessness had lower rates of all-cause readmissions during WPC. Exhibit 177 shows that the yearly change in readmission rates did not significantly change before WPC and then significantly declined during WPC. However, WPC enrollees and controls did not significantly differ in their changing rates from before to during WPC (DD). Exhibit 177: Trends in All-Cause Readmission following an Acute Inpatient Admission, Before and During WPC for WPC Enrollees and the Control Group, PY 2 - PY 6 as ss Wy Ss Ss s rf ao S53 SS Ss ss a5 = 5 =a of ms x a a9 a8 x | | | | | | | | | s n a 4 nN ” + rr) S S S S S S S o o a a o o a > > > > > > > 2 2 a a Before WPC During WPC mWPCEnrollees = Comparison Group ATA Oar Lay) ATA erly Pye) Paes eT ole eine TTT aa Between Changes Difference (DD) WPC Enrollees 1.1%* -1.0%* -2.1%* Control Group 1.1%* -0.6%* -1.7%* -0.4% Source: Medi-Cal claims data from January 1, 2015 through December 31, 2021. Notes: * Denotes p<0.05, a statistically significant difference. Change Before WPC is calculated as: (1 year before WPC minus 2 years before WPC divided). Change During WPC is calculated as: (5 years of WPC minus 1 year of WPC)/4. Difference between changes is calculated as: (Change During WPC —Change Before WPC). Difference-in-difference is calculated as: (Difference between changes for WPC enrollees — Difference between changes for control group). py( WPC Services and Outcomes for Enrollees Experiencing Homelessness | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research . December 2022 Health Economics and Evaluation Research Program Challenges and Successes In PY 6 follow-up interviews and narrative reports, the most common challenges Pilots faced in serving enrollees at-risk of or experiencing homelessness included: lack of affordable housing stock, difficulty obtaining data on housing outcomes, and successfully linking enrollees to appropriate supportive services once housed. Pilots emphasized that access to secure and stable housing was key for enrollees to improve their overall health. Pilots also recognized the importance of supportive and sustained services once enrollees were housed to stay successfully housed long-term. “Housing is a challenge. There is not a lot of housing stock... In the last year, we have seen rents increased so greatly, and access to housing has become even tighter than it was previously... It's not just about paying rent, it's also the expenses that it takes to get into housing. A lot of our enrollees, maybe their credit score isn't up to par for certain landlords. And in response to that, a mechanism will be like, they pay a double deposit or maybe they pay first and last month's rent at the same time. And they have to apply to multiple different apartments... all of these expenses really start to add up.” - Sacramento Approaches to Address Housing Challenges Pilots attempted to work with local partners to secure access to low-income housing. Several Pilots reported that relationships with local housing agencies or authorities enabled the prioritization of services for WPC enrollees and emphasized the importance of convening committees with representation from multiple sectors to share data and strategies to identify, engage, and prioritize vulnerable clients for health, housing, and social services. Pilots provided information on how they leveraged other funding sources within the county to pay for rent and other costs that were not eligible expenditures under WPC. Over half of WPC Pilots used their flexible housing subsidy pools housing funds to provide financial assistance to individuals facing challenges in accepting or maintaining placement for housing. This funding was used for a variety of purposes including security deposits, rent payments, and incentives to landlords. Some Pilots used other funding sources, such as federal and local grants. Partnerships offered opportunities for expanded housing. For example, in Placer, donations from Sutter Health assisted with the procurement of multiple properties for use by WPC enrollees. Whole Person Care Final Evaluation Report | WPC Services and Outcomes for Enrollees Experiencing [PZz4 Homelessness err UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Additionally, many Pilots found more targeted outreach and engagement with individuals experiencing homelessness as a result of integrating WPC with COVID-19 response. More specifically, COVID-19 emergency housing projects expanded short-term housing availability for many WPC enrollees and facilitated care coordination through co-located medical, behavioral, and social services. Pilots reported collaborative efforts to transition short-term emergency COVID-19 housing projects to long-term supportive housing programs. For example, in Alameda, the County purchased two Project Roomkey hotel sites in Oakland, with the intention of converting the 240 rooms into permanent supportive housing. While many housing challenges persisted, the effectiveness of housing and provision of supportive services to homeless enrollees was viewed as moderately successful by Pilots and many had intentions of continuing these efforts through Cal-AIM. “The pandemic has provided opportunities for Care Connect to coordinate and collaborate with a range of housing partners at a much deeper level and has also led to new opportunities to collaborate and support consumers. Additional funding through the CARES Act and FEMA, as well as the additional flexibility in WPC PY 5 (2020) funding is helpful, however coordinating all these funding sources within short and changing timelines has been challenging.” -Alameda “Care coordination staff have become increasingly proficient in their ability to address the housing needs for WPC patients through system protocols developed which identify homelessness or at risk of homelessness, being able to see the patient’s housing status in the HMIS system, developing relationships with housing agencies, and gaining familiarity with eligibility criteria and types of housing available.” -Santa Clara P¥E3) WPC Services and Outcomes for Enrollees Experiencing Homelessness | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Chapter 14: WPC Transition to CalAIM This chapter describes transition of WPC to CalAIM after Medi-Cal 2020 waiver funding ended. UCLA examined whether Pilots were contracted by Medi-Cal managed care plans to provide Enhanced Care Management (ECM) and Community Supports (CS) services as part of CalAIM, as well as the infrastructure and support that facilitated the transition from WPC to CalAIM. Consistent with evaluation goals, UCLA also assessed the extent to which Pilots maintained: (1) inter-organizational collaboration between WPC partners, (2) data sharing infrastructure needed to support integration of care, and (3) care coordination protocols under CalAIM or independently. Data sources for this chapter include DHCS administrative data on ECM and CS providers as of May 2022 and after conclusion of negotiations between Medi-Cal managed care plans. These data indicated whether LEs or their partners were going to serve as ECM or CS providers. Further data on challenges and successes of transition were obtained from PY 6 mid-year and annual narrative reports. PY 6 (2021) LE surveys and follow-up interviews with leadership and frontline staff provided perspective on Pilot readiness and transition intentions, as well as Pilot- reported CalAIM transition planning efforts. The PY 5 (2020) surveys were used to obtain the most recent information on specific services Pilots provided under WPC. For additional detail on data sources and methodology please see Appendices C, D, E, and F. Planning and Preparation for Transition Transition of WPC to ECM and CS under CalAIM was originally planned for January 2021, but these plans were delayed due to the advent of the COVID-19 pandemic. DHCS received a one- year extension for WPC to continue providing services through the end of 2021 to minimize disruptions in care for enrollees. In January 2021, DHCS embarked on a yearlong transition planning process. DHCS allowed WPC Pilots to utilize one of two different methods to support WPC enrollee transitions: (1) WPC Pilots could work directly with MCPs to identify members that qualified for transition through utilization and enrollment data, or (2) WPC Pilots could use DHCS as an intermediary and share member utilization and enrollment data with DHCS to develop a transition plan. For the latter, LEs submitted a list of the CINs of WPC enrollees whom they identified as eligible to transition to ECM/CS; DHCS checked the members’ plan assignment and sent the list to each MCP respectively. As part of the WPC closeout requirements, each WPC Pilot had to provide a model of care, detailing CalAIM services and activities, as well as confirmation of their contract(s) with MCP(s). Whole Person Care Final Evaluation Report | WPC Transition to CalAIM Renee 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 178 shows a timeline of key dates and activities related to the WPC transition under CalAIM. Exhibit 178: Timeline of Key Dates and Activities for WPC Transition to CalAIM January 2022 Enhanced Care Management (ECM) benefit available for select populations of focus in WPC* MCPs offer preapproved Community July 2022 Supports ECM benefit available for ' select populations of focus* December 31, 2020 December 31, 2021 (non-wPc) Original WPC WPC waiver January 2023 waiver ends extension ends ECM benefit available for all populations of focus (all) WPC waiver extension September 2021 Lead Entities and January 2021 DHCS finalize WPC Learning Collaborative WPC close out plans begins providing technical assistance to support transition to CalAIM DHCS technical supports offered for transition Notes: CalAIM “Select populations of focus” includes: individuals and families experiencing homelessness; high utilizer adults; adults with serious mental illness or substance use disorder (SMI/SUD); and adults and children/youth transitioning from incarceration. “All populations of focus” includes: adults at risk for institutionalization and eligible for long-term care; nursing facility residents who want to transition to community; and children and youth. “WPC close out plans” detailed Pilots’ transition plans for their WPC enrollees. MCPs is Medi-Cal Managed Care Plans. DHCS is California Department of Healthcare Services. Technical Support for Transition In 2021, the WPC Learning Collaborative, which had provided LEs with technical assistance (TA) on key elements of WPC implementation since the beginning of the Pilot, turned its attention to primarily supporting the transition to new Medi-Cal benefits and services under CalAIM. The Learning Collaborative, led by Aurrera Health Group, provided TA to LEs by sharing new and revised DHCS policies and guidance, providing LEs with the opportunity to discuss operationalization of the policies, and offering a forum for Pilots to ask DHCS target questions. Aurrera Health Group, in partnership with the California Safety Net Institute, also entered into a WPC Transition to CalAIM | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program new contract with the California Healthcare Foundation to run a parallel “Peer to Peer” group, which focused solely on transitioning eligible WPC enrollees to ECM and CS. DHCS held monthly CalAIM transition meetings to review DHCS-issued transition documents, as well as bi-weekly technical advisory meetings for MCPs and WPC programs to discuss common barriers and issues encountered during the transitioning process. When needed, DHCS facilitated ad-hoc meetings with WPC Pilots and/or MCPs to discuss and resolve complex issues unique to a specific county. Additionally, the WPC Services and Transition to Managed Care Mitigation Initiative provides direct funding for former WPC Pilot Les that meet specific criteria to pay for existing WPC services that map to ECM/CS services before they transition to CalAIM. Ten LEs were approved for a total of $137 million in sustaining services until 2024. Pilot Participation in Transition Planning Meetings In PY 6 surveys, all LEs reported that they participated in transition planning meetings with DHCS from mid-PY 5 to mid-PY 6 (26 of 26), and most also met with Medi-Cal MCPs (24) and other WPC partners (22; Exhibit 179). The majority of LEs (24) met with MCPs regarding CalAIM planning. Of these LEs, 23 reported discussing specific CS services with MCPs and 91% of LEs felt they had meaningful input in the transition planning process (data not shown). Many LEs (17) also reported discussing CalAIM with other WPC partners (17). Exhibit 179: Lead Entity Participation in Transition Planning Meetings with DHCS, Medi-Cal Managed Care Plans, and Other WPC Partners, August 2020-May 2021 MCPs Transition/ sustainability planning s 8 re x Other WPC partners MCPs 24 se SE os Other WPC partners 17 Source: PY 6 Lead Entity (LE) Survey (n=26), May-June 2021. Notes: DHCS is California Department of Health Care Services. MCPs are Medi-Cal Managed Care Plans. Whole Person Care Final Evaluation Report | WPC Transition to CalAIM Yer ntolee UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Additional detail on transition planning meetings provided in PY 6 annual narrative reports indicated that meetings between MCPs and Pilots were typically tailored to the specific circumstances and environment of each individual Pilot. Meetings varied in the extent to which they focused on transition of WPC enrollees to the ECM benefit within CalAIM or on infrastructure and changes needed for WPC partner(s) to serve as ECM or CS providers. The specific start dates of CalAIM planning efforts varied by county and the available resources at the time. Some counties had geographic access to several neighboring MCPs and initiated transition planning at an earlier stage of their program. “The executive leaders of Health Care Services Agency (Office of the Agency Director, Behavioral Health, and Public Health), the two health plans (Alameda Alliance and Anthem Blue Cross), and the two large safety net provider organizations (Alameda Health System and Community Health Center Network) met on a monthly basis throughout the year. The group discussed evolving plans for transition of services and infrastructure at the end of Whole Person Care, and how to stay in coordination as timelines changed... This regular cadence created a reliable space for communication, problem solving, collaboration, and coordination, primarily for sustainability planning through this evolving landscape... The group of executives has gelled in a friendly and Supportive way that will serve the safety net care system well into the future... together the parties analyzed the alignment of services, the capacity of the current and possible provider networks, the transition processes, and the financial opportunities and risk to lay the foundation for ongoing decision- making for sustaining as many of the AC Care Connect services as possible once the program would come to an end.” -Alameda Participation in Enhanced Care Management ECM is a new Medi-Cal benefit to provide eligible enrollees with intensive care coordination that addresses their clinical and non-clinical needs. ECM began implementation in January 2022, and is aligned with WPC best practices in requiring (1) use of a single, dedicated care manager to coordinate care and various delivery systems and (2) meeting enrollees “where they are at” (e.g., home, shelter, street) through in-person engagement and service delivery. DHCS estimated that approximately 15,000 WPC enrollees across 23 counties were eligible to transition from WPC Pilots to ECM on January 1, 2022. Eligible enrollees include any of the following seven CalAIM “populations of focus” for the program: (1) individuals and families experiencing homelessness, (2) adult high utilizers, (3) WPC Transition to CalAIM | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program adult SMI/SUD, (4) adults transitioning from incarceration, (5) adults at risk for institutionalization and eligible for long-term care, (6) nursing facility residents who want to transition to community, and (7) children and youth. The first four populations correspond to WPC “target populations;” the remaining three are new under ECM. Participating MCPs are required to provide ECM services to all eligible enrollees by January 2023. However, contracted ECM providers can choose which populations of focus to serve. In PY 6 surveys, 18 (of 26) LEs reported plans to serve as ECM providers. As of May 2022, DHCS reported that all 18 LEs were participating as ECM providers. In five counties (Kings, Los Angeles, Marin, Mendocino, and Sacramento), selected partners of the LE, rather than the LE were participating. As of May 2022, Solano and SCWPCC LEs and partners were not participating as ECM providers. These two Pilots also did not participate in the PY 6 extension year (2021). Whole Person Care Final Evaluation Report | WPC Transition to CalAIM Rare olye! UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 180 shows populations within each WPC-participating county that are being served through ECM as of May 2022. The most common target populations for ECM are individuals experiencing homelessness and adults with SMI/SUD (23 of 23 counties, respectively), followed by high utilizers (17) and justice-involved (14). All counties that identified SMI/SUD and individuals experiencing homelessness as a target population in WPC continued to serve adult SMI/SUD and individuals and families experiencing homelessness under ECM. Similarly, all counties that identified high-utilizers and justice involved as a target population in WPC continued to serve adult high utilizers and adults transitioning from incarceration under ECM, except Placer. All WPC-participating counties, except Placer, began serving new populations of focus under ECM, with the biggest increases seen in the percentage of counties serving adults with SMI/SUD (from 35% in WPC to 100% in ECM) and adults transitioning from incarceration (from 17% to 61% in ECM). PL: WPC Transition to CalAIM |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Exhibit 180: Populations of Focus, Served through Enhanced Care Management and Whole Person Care, May 2022 Target Population in WPC Not a Target Population in WPC 8 e fe 5 5 3S = oO — BE 2 €o ’ 2 53 . = €e& g = set S 5 Be = 2 2 ese 2 s 5» 3 a a& Za fF z 2g < $ = fess fz B se » = gs goe 2 © 32 = a ae a3 2 az 9 28 3 3 $5 |s22/ ee] 2 38 Ey z AEC Pilot £5 2 2 Ze | ic $8 6 Alameda vt vt vv - - Contra Costa Vv v* v v v v v Kern v* v* v vt = : © Kings v : v* v - - - Los Angeles vy v* v* v* v v vo Marin v* v* v v v v v Mendocino vf e v* Vv v : Vv Monterey vt v v v = 3 £ Napa vt * We m= - = om Orange v* * v* v wa * = Placer v* “ v* # = 8 : Riverside v v v At mn © : Sacramento ft v* v vo v v v San Bernardino Sf v* af 7 v - Vv San Diego v* v* v v v v v San Francisco vt v fi = “ s = San Joaquin v* v* v* v v v v San Mateo v v* v v v v <. Santa Clara fi v* v vo vo v v Santa Cruz v v v* - = = = Shasta ¥ ial of = é 5 iS Sonoma v* - v* - - - - Ventura Jf v* Vv - = Source: Cal-AIM Transition Spreadsheets by Medi-Cal Managed Care Plan, Submitted to California Department of Healthcare Services, May 2022. Notes: Vv indicates population of focus under Enhanced Care Management. * indicates a target population under Whole Person Care. Whole Person Care Final Evaluation Report | WPC Transition to CalAIM UCLA Center for Health Policy Research December 2022 - . Health Economics and Evaluation Research Program Community Supports Under CS, MCPs are permitted to provide eligible enrollees with 14 pre-approved services designed to address social determinants of health. CS were intended to serve as a cost-effective alternative to traditional services covered by Medi-Cal, and include services such as housing support and day rehabilitation. CS services are not restricted to ECM populations of focus, and eligible enrollees can receive CS in addition to ECM. DHCS estimated that approximately 8,000 WPC enrollees were eligible to transition to various CS services on January 1, 2022. In PY 5 surveys, UCLA collected systematic data from Pilots on six WPC services that were subsequently pre-approved CS services. These included: (1) environmental accessibility adaptations, (2) housing deposits, (3) housing tenancy and sustaining services, (4) housing transition navigation services, (5) recuperative care/medical respite, and (6) sobering centers (Exhibit 181; CS services are defined in the footnote below). Pilots may have elected to provide other CS services as part of WPC (e.g., short-term post-hospitalization housing), but UCLA did not collect systematic data on the extent to which these services were provided. As of May 2022, DHCS reported that all WPC Pilots were providing CS, although specific CS services offered varied by county. The most commonly provided CS services are housing tenancy and sustaining services (20 of 23), housing transition/navigation services (20), and recuperative care/medical respite (18); these are services that were also offered through WPC. Services that were not commonly offered through WPC, were less likely to be offered through CS (see Appendix T: Comprehensive Community Support Offerings by County). When comparing DHCS data from May 2022 to PY 5 survey data, results indicate a high degree of continuity of service provision from WPC to CS, particularly for environmental accessibility adaptations (100% who provided in WPC provide as CS), housing tenancy and sustaining services (85%), and provision of housing deposits (79%). pi: WPC Transition to CalAIM |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Exhibit 181: Participation of WPC Pilots in Selected Community Supports by County, May 2022 g bo ‘2 = o = 5 Es 2 a 3 3 a 3 2 = 3g ie s = 8 G £ 2 < = 3 5 2 3 3 | & 3 : 2 Ey 2 5 5 2 5 £8 a e ec 3 3 53 2 23 23 z = ee) 3 | 32] 32) 2 / 8 County 23 3 35 35 3 3 at =x xr za 4 a Alameda ve ve ve ve ve * Contra Costa # * v* * v Kern * v* v* v* v* Kings * v* v v* v v* Los Angeles * v* v* v* v* v Marin * * he v* * Mendocino % * * % = 2 Monterey v* v* v v* Napa v v v Orange * v* v* v v* Placer v* v* v* Vv v* v Riverside * v* v* v* v v* Sacramento v* v* v* v Vt v San Bernardino v Vv v* * = San Diego v* v v* v* v* San Francisco * * * * v* * San Joaquin % v* v v* v* v* San Mateo v* v* v* v* * Santa Clara * ve vt v v* * Santa Cruz v* v* Vt Vt Shasta * v* v* v v * Sonoma v v* v v * Ventura * v* v* v* ve Number Offering CS Service 5 19 20 20 18 7 Percent Offering Service Through CS Who Offered 100% 79% 85% 65% 67% 71% Through WPC Source: Cal-AIM Transition Spreadsheets by Medi-Cal Managed Care Plan, Submitted to California Department of Healthcare Services, May 2022. Notes: V indicates service under Enhanced Care Management. * indicates a service under Whole Person Care. Whole Person Care Final Evaluation Report | WPC Transition to CalAIM UCLA Center for Health Policy Research December 2022 - ; Health Economics and Evaluation Research Program As defined in DHCS Community Support Policy Guide, Environmental Accessibility Adaptations (e.g., Home Modifications) are physical adaptations to a home that are necessary to ensure the health, welfare, and safety of the individual, or enable the individual to function with greater independence in the home. Housing Deposits assist with identifying, coordinating, securing, or funding one-time services and modifications necessary to enable a person to establish a basic household that do not constitute room and board. Housing Tenancy and Sustaining Services ensure maintaining safe and stable tenancy once housing is secured. Recuperative Care/Medical Respite is short-term residential care for individuals who no longer require hospitalization, but still need to heal from an injury or illness (including behavioral health conditions) and whose condition would be exacerbated by an unstable living environment. Sobering Centers are alternative destinations for individuals who are found to be publicly intoxicated (due to alcohol and/or other drugs) and would otherwise be transported to the emergency department or jail. Transition Challenges and Successes Exhibit 182 shows the most common challenges and successes related to transition under CalAIM as reported in PY 6 mid-year and annual reports. In PY 6, the most frequently mentioned challenge in bi-annual narrative reports was that the scope of services and eligibility requirements for ECM differed from WPC (14 of 23). Pilots were concerned that clients would no longer receive the same intensity of touch that allowed for necessary trust and rapport building. Furthermore, Pilots were able to define their target population eligibility criteria for WPC but the eligibility criteria for ECM was viewed as stricter. For example, the most common definition for high utilizers in WPC was individuals with 3 or more emergency department (ED) visits in the last 12 months. For ECM, individuals with 5 or more ED visits in the last 6 months were considered to be high utilizers. Alameda estimated that their eligible pool for high utilizers would be cut by 90% due to narrowly defined target population definitions. There was also uncertainty around continued use of data sharing infrastructure developed through WPC (12). Due to changing requirements for reporting for CalAIM at the time, these Pilots lacked clarity in whether existing data systems would be sufficient and able to handle the CalAIM requirements. Pilots noted that there was a significant workload required for the transition to CalAIM (10), and that this came in the midst of still providing services for current WPC enrollees in PY 6. Lastly, Pilots noted that dissatisfaction with the proposed rates and contracting processes (8), as reimbursements were significantly lower than those provided under WPC. P2539) WPC Transition to CalAIM |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program “As the WPC Pilots end and services transition to managed care benefits, the flexibility to implement innovative approaches to patient care will decrease as providers are held to rigid regulatory requirements. Opportunities to innovate will be further restricted by funding shortfalls, with insufficient rates to support the scope of services offered under WPC. For example, CCHS WPC will no longer be able to support program provided cell phones, non-medical transportation, and free legal aide. These initiatives made possible by WPC funding have been tangible benefits that provide vital services to patients.” - Contra Costa Despite these challenges, Pilots made significant progress in their sustainability planning and transition to CalAIM. Most often, Pilots noted success in regular planning meetings and workgroups, which brought participating partners together to discuss the necessary next steps (18). Often as a result of these meetings, Pilots emphasized success in the transition/hand-off of qualifying WPC enrollees to ECM (16). Many Pilots utilized their data sharing platforms to facilitate the transition of enrollees to ECM and had concrete plans to utilize this infrastructure in CalAIM, particularly for reporting requirements and partner communication (15). Thirteen Pilots noted success in establishing workflows for ECM and specific CS services. “We successfully negotiated a contract with our local MCP to transition our 70 WPC clients to ECM and have incorporated new policies and procedures for the purpose of reporting timely and accurate member data to the Central California Alliance for Health. Our clients did not experience or notice a change in services due to the collaboration we were able to have with our partners during the closeout process.” -Monterey Whole Person Care Final Evaluation Report | WPC Transition to CalAIM Ree olay UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 182: Commonly Identified Challenges and Successes in Transition to CalAIM among WPC. Pilots, PY 6 CalAIM scope of services and eligibility requirements differed el) from WPC Continued use of data sharing infrastructure developed pe > through WPC Challenges Significant workload associated with transition to CalAIM (i 30 Disatisfaction with rates and contracting processes I s Regular planning meetings and workgroups to discuss pes +; transition WC enrollee transitions and handotts i 3 Successes Data infrastructure improvements and preparation for CalAM [i ; Established workflows and service plans for ECM/CS ry 15 Source: PY 6 (2021) Mid-Year and PY 6 Annual Narrative Reports (n=23). Note: Numbers indicate WPC Pilots that mentioned the thematic challenge at least once across the reporting period. Sustainability of WPC Goals and Pilot Innovations after WPC As of May 2022, all Pilots (either LE and/or their partners) that participated in PY 6 were participating in CalAIM. Key components of WPC that Pilots aimed to sustain to some degree through CalAIM included: (1) inter-organizational collaboration between WPC partners, (2) data sharing infrastructure needed to support integration of care, and (3) care coordination protocols. Inter-organizational Collaboration between WPC Partners As indicated in PY 6 surveys, LEs intended to maintain relationships with WPC partners regardless of CalAIM (21 of 23), with 11 LEs that indicated that CalAIM would be a mechanism to sustain those relationships with their partners. While LEs emphasized that partnerships established through WPC facilitated the transition to CalAIM, uncertainty remained about maintaining strength in those partnerships after WPC and the initial transition. WPC governance structures required participation from specific partner types, encouraging collaboration and communication. Without such formal structures and financial incentives to yl WPC Transition to CalAIM |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research December 2022 - . Health Economics and Evaluation Research Program medical respite for enrollees with chronic physical conditions. Examining the average payment by enrollee as a proxy for service intensity, showed the highest amounts for individuals with SMI/SUD, followed by enrollees with chronic physical conditions and lowest amounts for the COVID-19 population and enrollees at-risk of homelessness. WPC Care Coordination WPC Pilots aimed to “increase coordination and appropriate access to care for the most vulnerable Medi-Cal beneficiaries.” Evidence suggests Pilots were successful in developing diverse and appropriate infrastructure (e.g., staffing, data sharing, standardized protocols) and effectively delivered care coordination services (e.g., needs assessment, care plan, referrals) needed to support effective care coordination. These efforts were particularly innovative and notable in development of multidisciplinary care coordination teams who had access to data across partners, standardized care coordination protocols, financial incentives to WPC partners. Additional innovation included employment of care-coordination staff with “lived experience” (e.g., CHWs) and clinical expertise to address enrollee needs, offered tiered care coordination services and varied caseloads to match the complexity of enrollee need. Further successes in care coordination included regular and comprehensive assessment of medical, behavioral health, and social needs, development of comprehensive care plans, linking enrollees to appropriate service, and promoting accountability among care coordination teams. Pilots used innovative and creative strategies to engage enrollees in care including providing/arranging transportation to and from appointments and offering incentives (e.g., meals, personal care items) and service delivery to enrollees where they lived. WPC Quality Improvement, Program Monitoring, and Stakeholder Engagement WPC aimed to “achieve targeted quality and administrative improvement.” Pilots were required to engage in regular quality improvement activities and submit biannual Plan-Do-Study-Act (PDSA) reports documenting Pilot-led efforts to improve outcomes and metric performance. Evidence indicated substantial effort by Pilots in these quality improvement activities focusing on improving WPC implementation (e.g., ensuring development of a comprehensive care plan within 30 days of enrollment) and improving specific outcomes/metrics (e.g., reducing hospitalizations, diverting patients from the ED to more appropriate settings). Quality improvement and program monitoring activities allowed Pilots to meaningful adjust their implementation approach throughout the course of the Pilot and were perceived as positively contributing to Pilot performance and as helping Pilots identify which elements of their Pilot to prioritize for sustainability after the close of WPC. Conclusions | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research erly) Health Economics and Evaluation Research Program facilitate inter-organizational collaboration within CalAIM, Pilots anticipated challenges in delivery of services by separate ECM and CS entities. “While CalAIM is a good first attempt at incorporating WPC successes into the existing Medi-Cal medical billing model it does miss some of the success found in coordination and collaboration of services. CalAIM acknowledges the need for enhanced or intensive case management and the need for whole person care approach, including some social service and person-centered services. It, however, misses one of the most important needs identified and addressed in the Whole Person Care Program Model... that is coordinating services, collaborating client support, and including the client’s voice in the services that they receive. CalAIM acknowledges the need to address more than just the diagnosed medical or mental health needs of a person and attempts to provide funding for some assistance with basic living. However, it does not facilitate coordination of care among providers... It is up to the providers to reach out and establish relationships with other providers without knowing who that would be... We don't have mechanisms ourselves really, except the relationships and how they become, so nature and organic, that's what we're relying on right now because the funding structure isn't supporting maintenance of those relationships.” -Shasta Data Sharing Infrastructure Needed to Support Integration of Care Through WPC, many LEs established data sharing infrastructure (e.g., formal data sharing agreements with partners, care management platforms, event-based notifications). CalAIM was viewed as a strong mechanism for continuing data sharing infrastructure and processes established through WPC for the majority of Pilots. In PY 6 surveys, 15 of 23 Pilots expressed intentions to maintain data sharing infrastructure established through WPC regardless of CalAIM, whereas 13 had concrete plans to sustain via ECM. Fifteen Pilots had intentions to maintain existing data sharing agreements through CalAIM (data not shown). In PY 5 surveys, almost all Pilots (22 of 23) believed that data platforms and tools established through WPC would facilitate their transition to CalAIM. These tools were critical to ongoing case management, program monitoring, and strategic improvements (data not shown). Pilots described ways in which their data sharing infrastructure would continue through CalAIM as highlighted in Exhibit 183. Whole Person Care Final Evaluation Report | WPC Transition to CalAIM Eg UCLA Center for Health Policy Research D ber 2022 . ; Ls Health Economics and Evaluation Research Program Exhibit 183: Illustrative Examples of Plans to Sustain WPC Data Sharing Infrastructure under CalAIM Pilot Illustrative Example San Diego San Diego developed a “who’s in jail” push notification feature, which alerted case managers through text and e-mail when an enrollee was in jail. This allowed case managers to appropriately respond and organize resources. Due to the success of the feature, it was adopted for CalAIM. San Francisco In preparation for CalAIM, San Francisco assessed capacity of providers to appropriately document services in alignment with Medi-Cal standards across relevant record systems. WPC funded and launched the addition of a comprehensive care coordination module within EPIC called Compass Rose; EPIC will be utilized for CalAIM as it meets the reporting requirements. Santa Clara As learned for WPC reporting, Santa Clara utilized a database design approach within HealthLink. This approach will be utilized for CalAIM reporting to reduce reporting burden as report developers will not need to understand and navigate the vast HealthLink data system. Modifications were made to existing workflows, evaluating what changes were needed for CalAIM’s launch. Marin Marin used lessons learned from their WPC legal/policy framework for data sharing in CalAIM. Sacramento Beginning in mid-PY 6, Sacramento revised their monthly data dashboard to depict month-by-month comparisons of data categories such as total active enrollments, services provided to active enrollees by month (e.g., care coordination, housing, and service supports), housing disposition (permanent, transitional, shelter), clinical and housing hub provider panel size, and MCP assignments. The new transition-centric dashboard provided better understanding of enrollee movement across and out of the program, and facilitated tracking of themes and trends to inform the design and workflow of the transition process. Source: PY 6 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=26), June-September 2021. pPa WPC Transition to CalAIM |Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research ely) Health Economics and Evaluation Research Program ECM will use a single dedicated care coordinator, which in PY 6 interviews, many WPC Pilots identified as a “best practice” approach. Care Coordination Pilots emphasized the importance of ECM was viewed as a strong mechanism for continuing key care coordination elements established through WPC. As indicated in PY 6 surveys, 16 Pilots had intentions of maintaining care coordination processes (e.g., intake/assessments, linkages to services, communication pathways) through ECM. Eighteen Pilots had intentions of sustaining WPC staff through ECM, with 11 of those maintaining peer support staff (data not shown). high- intensity, field-based or in-person contact to meaningful enrollee engagement. When considering the transition to ECM, WPC Pilots had concerns about the intensity of touch possible with ECM defined scope and rates. More specifically, Pilots had concerns about inability to build the necessary trust and rapport to actively engage prospective enrollees in needed services. “The minimal amount of funding that is going to go to this work, will mean that hardly any hands-on, real time spent with their clients... You figure the actual cost that goes into even someone being seen for an hour a week, which is about what we were asking the wellness coaches [to do]. Sometimes, it's a little bit more time, because you can't sit there and like, ‘We have an hour and then your time's up.’ You want to build a trusting relationship, and [there’s] really, really private parts of somebody's life.” -Mendocino Whole Person Care Final Evaluation Report | WPC Transition to CalAIM Rare olye! UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Conclusions This final report presented findings from the comprehensive statewide evaluation of Whole Person Care (WPC) in California during the six years of implementation. The report provides extensive evidence of how the infrastructure for WPC implementation was developed by WPC Pilots, what processes were followed to implement the program, what services were delivered, and whether WPC led to better care, better health, and lower costs. These conclusions are detailed below. Structure of WPC Pilots Available data suggest that WPC Pilots successfully achieved WPC goals of “increased integration among county agencies, health plans, providers, and other entities within the county that serve high-risk and high-utilizing beneficiaries” and “developed infrastructure that would ensure local collaboration among the entities participating in the WPC Pilots over the long term.” Pilots chose Lead Entities (LE) that had the leadership and administrative capacity to effectively implement WPC, with the majority being county health services or public health departments and agencies. Pilots also included other county agencies, health plans, and community providers as partners. Reflecting Pilots’ commitment to improving integration of health and human services, over a third of partners were housing support or other social service providers. LEs invested considerable effort to meaningfully engage partners in WPC (e.g., regular meetings, case conferences, etc.). Partners reported significant impact of WPC on goals such as improved data sharing, integration of care, and care delivery. Health Information Technology and Data Sharing Infrastructure WPC Pilots were required to “improve data collection and sharing amongst local entities to support ongoing case management, monitoring, and strategic program improvements in a sustainable fashion.” All Pilots succeeded in improving their data sharing capacity by investing considerable effort and resources into related activities. Initial progress was slow due to the considerable start-up activities required to support data sharing (e.g., overcoming legal and cultural barriers to data sharing, research into and procurement of appropriate care management platform(s), training and modifying workflows to facilitate uptake by frontline staff). However, by the end of WPC, all Pilots successfully established data sharing agreements with at least some partners and most Pilots expanded, acquired, or developed a care management platform to facilitate tracking of enrollee-level data. Other important data sharing infrastructure established through WPC included universal enrollee consent forms, processes to support real-time data access by frontline staff working in the field, integration of care management platforms with existing electronic health records (EHRs), and real-time notification of emergency department or inpatient hospital visits. Most LEs financially incentivized partners to develop needed data sharing infrastructure and report on required Conclusions | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research " ” [pret el gy APs Health Economics and Evaluation Research Program data elements, and viewed these incentives as important for ensuring partner’s participation in data sharing activities. Although most Pilots reported continued room for improvement (e.g., in functionality of selected data sharing platforms), all Pilots were able to share the most important data needed to support enrollee outreach and engagement, care coordination, monitoring of partner performance, and quality improvement activities. Overall, Pilots viewed WPC as critical for facilitating development of new data sharing infrastructure and in facilitating cross-sector coordination needed to effectively manage enrollee care. WPC Enrollment Size, Patterns, and Trends WPC Pilots were required to identify eligible Medi-Cal beneficiaries using pre-defined inclusion criteria, enroll them in WPC, and engage enrollees in care. Evidence from the evaluation indicated that Pilots succeeded in these activities, with a steady growth in enrollment culminating in 249,378 unique beneficiaries, including the majority who were high utilizers or experiencing homelessness and many who had serious mental illness or substance use disorders (SMI/SUD) conditions or were justice-involved. This level of enrollment was made possible by using innovative and tailored approaches for identifying eligible enrollees including referrals from community-based partners, predictive modeling to identify at-risk beneficiaries, and field-based outreach at medical facilities, streets, or shelters where enrollees lived. Another important innovation was employing staff with lived experience for outreach and engagement of eligible population such as those experiencing homelessness who had higher levels of medical mistrust or those who were justice-involved and required warm-handoffs at county jails and probation offices upon release. These efforts may have contributed to longer enrollment particularly among enrollees with SMI/SUD. WPC Services Offered and Delivered WPC Pilots aimed “increase coordination and appropriate access to care” and “increase access to housing and supportive services.” Analysis of data showed that Pilots not only offered more basic services such as outreach, care coordination, and housing support but many added other supportive services including benefit assistance, health education, legal services, employment services, sobering centers, and medical respite to address social needs and avert recidivism or avoidable use of emergency departments (ED) and hospitals. WPC allowed Pilots to deliver WPC services under bundles of services paid through per-member, per-month (PMPM) payments or individual services paid on a fee-for-service (FFS) basis. Services provided by LEs were frequently bundled and services provided by partners were frequently not bundled. As a result, assessment of receipt of specific services per enrollee overall was not possible. Nevertheless, analyses showed targeted use of some services by enrollee need such as highest rates of Whole Person Care Final Evaluation Report | Conclusions (PEs UCLA Center for Health Policy Research " ” [pret el gy APs Health Economics and Evaluation Research Program The COVID-19 pandemic started in early 2020, during the fourth year of WPC implementation and resulted in the program being extended for an additional year. UCLA investigated the extent to which COVID-19 impacted WPC implementation, enrollment, and enrollees, as well as whether the impact of the pandemic was similar among enrollees and their matched controls. The finding indicated that Pilots were able to respond to the challenges presented by the pandemic quickly and minimize its impact on WPC enrollment and service use. The findings also highlighted the unanticipated value of WPC investments in system-wide integration in responding to emergencies such as COVID-19. Specific findings suggested that Pilots were able to respond to COVID-19 protocols that prevented in-person outreach and delivery of care coordination and created new needs among the targeted populations. These efforts included changing their original workflows, using new tools and strategies, and developing other innovative approaches in response to the challenges presented by the pandemic. Some changes were relatively simple (e.g., ability to collect consent over the phone instead of mandating in- person verbal consent), and others were more complex (e.g., expanded short-term housing opportunities, creating a “one stop shop” centered around COVID-19 isolation housing). WPC and COVID-19 Early in the pandemic, Pilots limited in-person outreach and shifted to primarily telephonic care coordination, but most had reverted to previous practices by the close of the program. The changes were possible due to the of infrastructure and processes established through WPC, including availability of screening protocols, trained and experienced staff, and data sharing agreements and platforms. These efforts likely led to the continued growth of WPC enrollment throughout 2020 and into 2021. As the pandemic continued, many Pilots tailored WPC efforts to align with new COVID-19 initiatives such as Project RoomKey and Project HomeKey. Analysis further indicated that the rate of COVID-19 infections and use of related services were similar for WPC enrollees and controls. The findings also indicated a prolonged reduction in ED visits and hospitalizations but a shorter-term impact on primary care and specialty care utilization most likely due to the increased use of telehealth services. Enrollee Demographics, Health Status, and Prior Health Care Utilization WPC Pilots aimed to enroll the “most vulnerable Medi-Cal beneficiaries” but had flexibility in choosing from seven populations of focus (e.g., high utilizers, individuals with chronic physical or behavioral health conditions, individuals experiencing homelessness). Data showed that all WPC Pilots successfully enrolled the most vulnerable Medi-Cal beneficiaries who were at risk of being or who were high utilizers. Specifically, data showed many enrollees were from communities of color; had high prevalence of multiple chronic physical conditions, mental health conditions, and substance use disorders; and/or had an upwards trajectory in use of emergency department visits and hospitalizations prior to enrollment. Whole Person Care Final Evaluation Report | Conclusions (PEy/ UCLA Center for Health Policy Research December 2022 - . Health Economics and Evaluation Research Program Better Care WPC aimed to use care coordination and WPC services to “increase appropriate access to care and improve beneficiary care outcomes.” Evaluation findings provided support for this WPC goal and further insights on how patterns of care changed over time and for important sub- groups of high utilizer Medi-Cal beneficiaries. Specifically, data showed that enrollees use of outpatient services increased in the first year of WPC. Comparing trends from before to during WPGC, enrollees had a reduction in primary care, an increase in specialty care, a decline in mental health care, and an increase in substance use treatment for enrollees overall vs. the control group. These patterns likely indicated that WPC enrollees were overusing primary care services prior to enrollment in lieu of other appropriate care due to limited specialty care access and underdiagnosis and underuse of mental health and substance use treatment prior to enrollment. Following enrollment, care coordination that included assessing need and treating unmet need led to increased access to care early on and more appropriate use of services in the right settings in the following periods. Additional analyses of two important subgroups of enrollees, those with serious mental illness/substance use disorders/experiencing homelessness (SMI/SUD/HML) and those who were medically complex or high risk (MC/HR) showed two somewhat different trajectories and pattern of change for each group. Data showed a greater initial increase in mental health and substance use disorder services for MC/HR enrollees after enrollment; a greater decline in primary care for SMI/SUD/HML than MC/HR enrollees; similar decline in specialty care for both groups; a decline in mental health care for SMI/SUD/HML but an increase for MC/HR group; and an increase in substance use treatment for MC/HR and a decline for SMI/SUD/HML. These findings likely indicated a greater overuse of primary care services for the SMI/SUD/HML, which was addressed by provision of more mental health care rather than substance use treatment. On the other hand, evidence indicated likely presence of undetected and untreated mental health and substance use disorders for the MC/HR group that led to greater use of mental health care and substance use treatment. Further evidence supported delivery of better care under WPC and based on WPC metrics, including the increase in mental health hospitalizations with a follow-up outpatient visit within seven days, engagement in substance use treatment, provision of comprehensive care plans, and suicide risk assessment of enrollees with major depressive disorders. Surveys and interviews with Pilots provided additional insights on how some metrics may have improved such as use of financial incentives to motivate achieving specific metrics. Changes in utilization patterns were also supported by Pilots perceived increases in access and delivery of comprehensiveness and timely care despite challenges such as availability of same or next-day primary care appointments and shifts to telehealth due to the COVID-19 pandemic. Conclusions | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research " ” [pret el gy APs Health Economics and Evaluation Research Program WPC aimed to “reduce inappropriate emergency and inpatient utilization” and “improve health Better Health outcomes for the WPC population.” Evaluation findings provided support for this WPC goal and yielded further insights into how patterns of care changed over time and for important sub- groups of WPC enrollees. Importantly, data showed an overall reduction in ED visits and hospitalizations and an increase in long-term stays for enrollees relative to the control group. Reductions in ED visits could be attributed to changing patterns of outpatient care, described in the Better Care chapter, and to intensive efforts by Pilots to employ more effective ED diversion strategies. Reductions in hospitalizations, coupled with lack of change in all-cause readmissions, could be attributed to a decline in first-time hospitalizations. Increases in long-term stays may have occurred as enrollees were assessed for need and diverted from hospitals to lower intensity settings to receive rehabilitation services. Additional analyses of SMI/SUD/HML and MC/HR subpopulations showed slightly different patterns of change in these groups. Specifically, analyses indicate a larger decline in ED visits for the SMI/SUD/HML than the MC/HR group, a greater decline in hospitalizations for the SMI/SUD/HML than the MC/HR group, and a greater increase in long-term stays for the SMI/SUD/HML than the MC/HR group. The findings further emphasized the concentration of avoidable ED visits and hospitalization among enrollees with SMI/SUD/HML and the likely importance of care coordination in helping navigate these patients to more appropriate care settings. Analyses also revealed positive impacts of WPC on other aspects of health, including better control of blood pressure and Pilot-reported improvements in overall health, comprehensive diabetes care management, and depression remissions. The principal challenge reported by Pilots as limiting their ability to improve enrollee health was the COVID-19 pandemic and enrollee concerns of contracting COVID-19, which limited their willingness to engage in appropriate care. Lower Costs UCLA assessed seven measures of health care costs that corresponded to majority of utilization measures examined in Better Care and Better Health chapters. Together, these measures illustrated potential changes in pattern of care and their associated costs under WPC. The evaluation findings provided support for reduction in overall costs, an estimated $383 per enrollee per year. The examination of costs for relevant categories of service showed that the decline in overall costs was likely accomplished through a decline in hospitalizations, outpatient services, and emergency department visits. This was despite increases in prescription Whole Person Care Final Evaluation Report | Conclusions (REx) UCLA Center for Health Policy Research December 2022 - . Health Economics and Evaluation Research Program medication costs and other residual services and no decline in cost of long-term care stays. These finding likely reflect the potential for savings when avoidable hospitalizations, emergency department visits, and outpatient services are reduced. Evidence further showed a greater decline in overall costs and outpatient costs, a greater increase in outpatient medication costs, an increase in ED costs, and a decline in long-term costs for MC/HR enrollees vs. those with SMI/SUD/HML. At the same time, the findings from the Better Care chapter indicated increased use of mental health services and substance use treatment and findings from Better Health chapter indicated a smaller declines in hospitalizations and ED visits. It is likely that reduction in outpatient costs occurred because these enrollees were better managed with medications and their previously untreated or undiagnosed needs were better addressed. However, it is also likely that when these enrollees had ED visits, they were likely to be for emergent conditions such as alcohol and drug poisonings and required more intensive interventions. For SMI/SUD/HML enrollees, evidence showed a decline in overall, outpatient, ED, and hospitalization costs, an increase in long-term care costs, and a greater decline in hospitalization costs and greater increase in cost of residual services compared to MC/HR enrollees. At the same time, the findings in the Better Health chapter showed a greater decline in ED visits and hospitalization but an increase in long-term stays. It is likely that many of the emergency departments visits that were avoided were non-emergent and these enrollees needed outpatient or social services. It is also likely that reduced hospitalizations were also avoidable and low-cost. WPC Enrollees Experiencing Homelessness Services and Outcomes WPC targeted beneficiaries who were experiencing or at-risk of homelessness and aimed to “increase access to housing and supportive services.” Evaluation findings showed that Pilots succeeded in enrolling mostly beneficiaries who were experiencing homelessness, provided housing support services to them using innovative and effective approaches, and improved their outcomes. Pilots did this through strategic and innovative approaches in outreach and WPC care delivery that matched the needs and living conditions of these enrollees. More specifically, many had higher rates of behavioral health conditions, higher utilization of emergency departments, mental health services and substance use services. Therefore, Pilots provided a higher intensity WPC service utilization and focused on provision of permanent housing following the “housing first” approach. Pilots innovated solutions to address challenges of lack of WPC funding for housing costs and chronic lack of adequate housing supply by leveraging other funding sources and working with external partners. These efforts succeeded in permanent housing for some and retention by other types of financial supports. These efforts and more intensive care coordination likely resulted in increased access to more appropriate Conclusions | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research " ” [pret el gy APs Health Economics and Evaluation Research Program mental health services such as timely follow-up care for mental health hospitalizations and engagement in alcohol and other drug dependence treatment as well as reductions in acute care utilization in emergency department visits and hospitalizations. Sustainability and Transition to CalAIM Sustainability of the goals of WPC was ensured by provision of two new Medi-Cal services called Enhanced Care Management (ECM) and Community Supports (CS) under CalAIM and similarities between the WPC target populations with the CalAIM “populations of focus.” In preparation for CalAIM, DHCS embarked on a one-year effort to provide technical assistance and other supports. Pilot reported transition challenges included need for clarity in scope of services and eligibility requirements for ECM, and these challenges were addressed through facilitation of meetings and provision of policies and guidance to Pilots and managed care plans by DHCS and contractors. Pilots found the regular planning meetings and workgroups brought participating managed care plans and WPC partners together to discuss the necessary next steps. These efforts led to participation of all WPC Pilots, either the LEs or Pilot partners in ECM and CS, with variations by county. This transition insured that the major goals of WPC including promoting development of local public-private partnerships that were supported by data sharing infrastructure in order to provide care coordination to Medicaid beneficiaries who were high utilizers of care were sustained. Specifically, participating WPC Pilots had the needed expertise in provision of care to SMI/SUD, justice-involved, high utilizers, and individuals experiencing homelessness including expertise in providing needed housing services, recuperative care, and medical respite. Implications The evaluation findings stated above described a major and expansive effort by California Department of Health Care Services to address the needs of the most vulnerable Medi-Cal beneficiaries who were at risk of or high utilizers of acute services in emergency departments and hospitals. WPC was specifically focused on care coordination and housing support services in recognition of the most important needs of these beneficiaries. Provision of these services was anticipated to lead to more appropriate use of medical and behavioral health services offered by Medi-Cal and subsequently guide WPC enrollees into more appropriate care settings and reduce avoidable acute care and its associated costs. To achieve these goals, WPC was designed as a localized program that was based on public-private partnerships and therefore could be customized to some degree to fit the existing infrastructure, resources, and population characteristics of each locality. The public-private partnership approach to program implementation required the establishment of data sharing infrastructure and ways to bridge over organizational silos and data confidentiality requirements. Whole Person Care Final Evaluation Report | Conclusions Ea UCLA Center for Health Policy Research December 2022 - . Health Economics and Evaluation Research Program The evaluation findings provided detailed information on what Pilots did to establish partnerships and the other infrastructure and how they succeeded in delivery of WPC services. Evaluation findings further illustrated challenges Pilots faces and innovations they used to overcome them. Ultimately, the findings showed that WPC achieved its goal of guiding patients to more care appropriate settings and receipt of needed services to improve their health. The extensive assessment of two important subgroups of enrollees, including those with serious mental illness, substance use disorders, or experiencing homelessness vs. others who were at high risk or with multiple chronic conditions highlighted that program savings were notably greater for the latter enrollees. Given that savings were not realized for the former group despite significant reductions in their use of potentially avoidable acute care suggest that the high need for continuous care over time overshadowed these cost savings. The early successes of the WPC were instrumental in California’s efforts to sustain several aspects of WPC under CalAIM, including creation of Enhanced Care Management (ECM) and Community Supports (CS) covered services under Medi-Cal managed care. While the coverage of these services became the responsibility of Medi-Cal Managed Care Plans (MCPs), California invested significant effort to retain the infrastructure and processes created by WPC Pilots by facilitating contractual agreements between MCPs and LEs or their partners. In addition, CalAIM’s PATH initiative funding was made available to former WPC Pilot Lead Entities until the services transitioned to managed care coverage under CalAIM. CalAIM seeks to retain best practices at the local level and continuity of care for enrollees. The implications of the WPC evaluation findings are numerous. Broadly, the implementation approach, best practices, and reasoning behind Pilot decisions are helpful for ongoing implementation of ECM and CS, planning the expansion of ECM and CS in new localities where no Pilots were operating, or in other states contemplating similar interventions. The differences in outcomes between beneficiaries who need extensive and continuous services and those whose health profile is less complex is helpful in forming expectations of the outcomes and associated savings of such programs for various beneficiaries. Importantly, the findings implied that navigating very complex beneficiaries to appropriate settings may reduce their health care spending less than those with less complexity but could lead to well-being and other significant system-wide benefits such as reducing congestion in acute care settings. These findings also * ECM is a new statewide Medi-Cal benefit available to select “Populations of Focus” that will address clinical and non-clinical needs of the highest-need enrollees through intensive coordination of health and health-related services; beneficiaries will have a single Lead Care Manager who will coordinate care and services among the physical, behavioral, dental, developmental, and social services delivery systems. CS are new social support services provided by Medi-Cal managed care plans as cost effective alternatives to traditional medical services or settings, including services such as medically supportive foods or housing supports. Conclusions | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research rr Health Economics and Evaluation Research Program indicate the need for a closer look at subgroups of this population such as those who are recently experiencing or have been chronically experiencing homelessness, and those with SMI vs. SUD but no other complications. It is likely that there are multiple categories of complexity among such enrollees. Each requires different tailored interventions, and provision of care could lead to different trajectories in service use and related costs. Whole Person Care Final Evaluation Report | Conclusions Ea December 2022 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Appendix A: Data Sources and Analytic Methods for Quantitative Analysis WPC Quarterly Enrollment and Utilization Reports UCLA used WPC Quarterly Enrollment and Utilization Reports to analyze WPC enrollment and utilization of WPC services. All Pilots submitted quarterly reports during the time they had implemented WPC from January 1, 2017 to December 31, 2021. Analytic Methods Exhibit 184 shows the enrollment data obtained from these reports. If there were conflicting data for individual enrollees between quarterly reports, UCLA used the more recent data. Enrollees that were enrolled in more than one Pilot at the same time were excluded from analysis (n=576). An additional 1,492 individuals were enrolled in more than one Pilot, but not at the same time. These individuals were counted as unique enrollees for each Pilot they enrolled in during the program. Exhibit 184: Beneficiary-Level Variables Data Elements Definitions Pilot Pilot in which enrollee is enrolled. Monthly Enrollment Status Indicator for WPC enrollment status for a particular month. Enrollment Date The date an enrollee starts to enroll in WPC. Disenrollment Date The date an enrollee disenrolled from WPC. Reason for Disenrollment Reason for disenrollment from a standardized list developed by DHCS. Number of Times Disenrolled The number of times each enrollee disenrolled from the MCP throughout their enrollment. Length of Enrollment The differences between disenrollment date and enrollment date. If an enrollee enrolls in and disenrolls from WPC on the same date, the length of enrollment will be one day. Target Population Indicator to inclusion in up to seven target populations. Enrollees were included in a target population if ever reported as part of a given target population. Homeless Indicator Indicator of experiencing homelessness that was separate from homeless target population. Notes: Data from WPC Quarterly Enrollment and Utilization Reports from January 1, 2017 to December 31, 2021. UCLA further used the WPC Quarterly Enrollment and Utilization Reports to identify monthly utilization of Pilot-created WPC service categories. These included per-member, per-month (PMPM) and fee-for-service (FFS) categories. Pilots reported whether enrollees were included EY Appendix A: Data Sources and Analytic Methods for Quantitative Analysis | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research " ” [ptetinlel ayy Health Economics and Evaluation Research Program in each PMPM category each month (yes/no) and how many times they received an FFS category each month (numerical integer). Limitations UCLA analyzed the enrollment data provided by WPC Pilots. Enrollment and utilization data did not always align, with some enrollees having no reported WPC services. In some cases this was the result of services that were not reimbursed through PMPM and FFS, but in other cases it resulted from lack of engagement in the program. Pilot methodology for reporting of target populations differed, with some Pilots reporting on all target populations regardless of whether the target population was a primary target of the Pilot and others only reporting on those that were a primary target. As a result, some enrollees that would meet the criteria of a given target population are not included in that population. One of the standardized disenrollment reasons, “graduated,” was not added until 2018 and as a result some enrollees that successfully left the program are not accurately captured as disenrolling for that reason. Medi-Cal Enrollment and Claims Data UCLA used Medi-Cal eligibility and claims data from January 1, 2015 to December 31, 2021 to create the demographics, health status indicators, health care utilization indicators, WPC performance metrics, and UCLA-created metrics used in this report. Claims data included both managed care and fee-for-service encounters, including Short-Doyle claims. Claims did not include dental claims. Analytic Methods Demographic Indicators Exhibit 185 displays demographic indicators created by UCLA using Medi-Cal monthly eligibility data. UCLA calculated age based on an enrollee’s WPC enrollment date. On the rare occasion enrollment data included more than one birthday for an enrollee, UCLA used the latest birthday reported. While not common, if the Medi-Cal enrollment data contained conflicting data for gender, race, or language for an WPC enrollee, UCLA used the most frequently reported category. Exhibit 185: Demographic Indicators Indicators Definitions Age Enrollee’s final age in years at the time of WPC enrollment. Gender Indicates whether an enrollee is male or female. Race The race label for an enrollee: White, Hispanic, African American, Asian American and Pacific Islander, American Indian and Alaska Native, other, or unknown. Whole Person Care Final Evaluation Report | Appendix A: Data Sources and Analytic Methods for [Reis Quantitative Analysis UCLA Center for Health Policy Research December 2022 - ; Health Economics and Evaluation Research Program Indicators Definitions English as Primary Indicating whether an enrollee’s primary language is English or not. Language Number of Months Full scope coverage is defined as at enrollment in at least one dental MCP and another with Full Scope non-dental MCP during the eligible date period. The number of months that an enrollee Coverage is full scope is reported for the year prior to the enrollee’s initial enrollment in WPC. Health Status Indicators UCLA used Medi-Cal claims data from January 1, 2015 to December 31, 2021 to assess health status of WPC enrollees prior to their enrollment in WPC. UCLA used the criteria set by CMS’s Chronic Condition Warehouse (CCW) to obtain a complete list of chronic condition and potentially chronic or disabling condition categories that were present in the two years prior to an enrollee’s enrollment in WPC (baseline). Additionally, UCLA created two indicators to identify enrollees with serious mental illness and substance use disorders based on ICD codes from the CCW definitions. WPC Metrics and Measures WPC metrics were calculated based on WPC metric specifications. WPC metrics were grouped by whether they measured progress towards better care, better health or lower costs. All metrics were reported in the aggregate and included data for two years prior to and five years following each individual’s enrollment in WPC when possible. UCLA assessed any length of enrollment or required number of months of enrollment on Medi-Cal enrollment rather than WPC enrollment in order to be consistent between WPC enrollees and the control group. All metrics were reported annually in order to assist in interpretation of findings. Exhibit 186 includes descriptions of all WPC metrics and how changes in the metric are to be interpreted. Exhibit 186: WPC Metrics, Definitions, and Intended Direction . — Improvement Measured by Metric Description Increase or Decrease Follow-Up After Percentage of discharges for enrollees age 6 and older Increase Hospitalization for who were hospitalized for treatment of selected mental Mental Illness within illness diagnoses and who had a follow-up visit with a 30 days mental health practitioner within 30 days. Follow-Up After Percentage of discharges for enrollees age 6 and older Increase Hospitalization for who were hospitalized for treatment of selected mental Mental Illness within illness diagnoses and who had a follow-up visit with a 7 days mental health practitioner within 7 days. E{els9) Appendix A: Data Sources and Analytic Methods for Quantitative Analysis | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research " " 1B 2022 Health Economics and Evaluation Research Program seeder Initiation of Alcohol Percentage of enrollees who initiate treatment through | Increase and Other Drug Abuse | within 14 days of the diagnosis. or Dependence Treatment Engagement of Percentage of WPC enrollees who initiate treatment Increase Alcohol and Other and who had two or more additional AOD services or Drug Abuse or MAT within 34 days of the initiation visit. Dependence Treatment Controlling High Blood | Percentage of WPC enrollees ages 18 to 85 who had a Increase Pressure diagnosis of hypertension (HTN) and whose blood pressure (BP) was adequately controlled during the measurement year. Comprehensive Percentage of enrollees with type 1 or type 2 diabetes Increase Diabetes Care that received HgA1c testing during the measurement year.* All-Cause The number of acute inpatient stays during the Decrease Readmissions measurement year that were followed by an unplanned acute readmission for any diagnosis within 30 days and the predicted probability of an acute readmission. Ambulatory Care: The total number emergency department (ED) visits Decrease Emergency resulting in discharge normalized by the total number of Department (ED) Medi-Cal enrolled member months, multiplying the Visits result by 1,000. UCLA multiplied the findings by 12 in order to report rate as per 1,000 beneficiary per year. Inpatient Utilization The total number of inpatient visits normalized by the Decrease total number of Medi-Cal enrolled member months, multiplying the result by 1,000. UCLA multiplied the findings by 12 in order to report rate as per 1,000 beneficiary per year. Source: Detailed information for each metric is available in WPC Metric Specifications. Note: *The WPC metric specified examining rates of controlled diabetes (HgA1c<8%), but reporting rates of tests results were too low in the Medi-Cal claims data. Whole Person Care Final Evaluation Report | Appendix A: Data Sources and Analytic Methods for Quantitative Analysis EO] RY ne olaye UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Additional Healthcare Utilization Measures UCLA also created additional measures of healthcare utilization indicators using Healthcare Effectiveness Data and Information Set (HEDIS) 2019 Volume 2 definitions, National Uniform Claim Committee taxonomy designations, the Chronic Conditions Warehouse, and the American Medical Association’s Current Procedure Terminology (CPT) Codebook. Exhibit 187 displays these indicators. Exhibit 187: Healthcare Utilization Indicators Indicators Definitions Improvement Measured by Increase or Decrease Number of Primary Care Services per The number primary care provider services Decrease 1,000 Beneficiaries per Year during the year for every 1,000 beneficiaries. Number of Specialty Services per 1,000 | The number of specialty services during the Increase Beneficiaries per Year year for every 1,000 beneficiaries. Number of Mental Health Services per | The number of mental health services during Decrease 1,000 Beneficiaries per Year the year for every 1,000 beneficiaries. Number of Substance Use Disorder The number of substance use disorder services | Increase Services per 1,000 Beneficiaries per during the year for every 1,000 beneficiaries. Year Number of Long-Term Care Stays per The number of the long-term care stays during | Increase 1,000 Beneficiaries per Year the year for every 1,000 beneficiaries Control Group Construction In order to construct the control group, UCLA needed to identify a large group of Medi-Cal beneficiaries that were similar to WPC enrollees and had sufficient variability to improve the chance of identifying a match for each enrollee. This was accomplished through a multi-step process. In the first step, UCLA used a very broad set of selection criteria to pull a limited number of variables on possible controls. These selection criteria included Med-Cal beneficiaries that had any of the following during the two years prior to WPC implementation or during the five years of WPC implementation (January 1, 2015 — December 31, 2022): e Any emergency department visit « Any hospitalization e Any claim with a place of service or ICD that indicated homelessness e An address-based keyword that indicated homelessness For these beneficiaries, UCLA obtained annual data on their age, gender, county of residence, number months enrolled in Medi-Cal, homelessness status, and emergency department, hospital and outpatient utilization. E{0}:3) Appendix A: Data Sources and Analytic Methods for Quantitative Analysis | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program For the second step, UCLA used a stratified sampling process to find potential controls for each annual cohort of WPC enrollees. Each annual cohort was matched using data from two years prior to their WPC enrollment and the year of WPC enrollment (for example, 2017 enrollees were matched using data from 2015 through 2017). UCLA selected 10 possible controls for each enrollee that matched based on age group, gender, homelessness status, hospitalization patterns, emergency department visit patterns, outpatient utilizations patterns, and county of residence. If ten possible controls were not identified, UCLA used an urban, suburban, or rural county status instead of exact county or no county indicator to identify potential controls. Once an individual was identified as a potential control, they were removed from the pool available for matches with other annual cohorts. This process identified 2.7 million potential controls. UCLA then obtained complete administrative Medi-Cal monthly enrollment and claims data from January 2015 to December 2021 for 275,840 individuals reported in WPC Quarterly Enrollment and Utilization Reports and for 2.7 million individuals that were potentially eligible for WPC based on the preliminary matching process described above. UCLA used 64 variables indicating demographic, health status, service utilization, and cost to select the control group (Exhibit 188). Demographic variables were constructed from Medi-Cal enrollment data. Health status variables were constructed from claims data and included measures of chronic and behavioral health conditions (e.g., asthma, diabetes, hypertension, chronic kidney disease). Additional variables that measured differential in utilization rates and payments between baseline years were created when possible. Exhibit 188: Variables Used to Select the Control Group Indicator Description Demographics (41 indicators) Age Group (5 indicators) Age at the start of WPC enrollment (0-17, 18-34, 35-49, 50-64, or 65+ years) Gender (1 indicator) Reported Gender in Medi-Cal Enrollment (Male or Female) Race/Ethnicity (5 indicators) Reported Race/Ethnicity in Medi-Cal (White, Hispanic, Black, Asian or Pacific Islander, or Native American/Other/Unknown) Language (1 indicator) English as the preferred language Two years of baseline data (1 Indicator of whether beneficiary had one or two years of baseline data. indicator) Homelessness (2 indicator) Indicator of homelessness during each baseline year. County (26 indicator) County of residence (26 WPC counties) Health Status (12 indicators and variables) CCW chronic conditions (1 Count of the number of CCW chronic and disabling conditions during variable) baseline. Chronic condition category (3 Indicators of chronic condition count (0, 1-2, or 3 or more) during baseline. indicators) Whole Person Care Final Evaluation Report | Appendix A: Data Sources and Analytic Methods for JRE) Quantitative Analysis UCLA Center for Health Policy Research December 2022 - ; Health Economics and Evaluation Research Program Indicator Description Serious Mental Illness (2 Indicators of serious mental illness during baseline years (pre-year 1 and pre- indicators) year 2). Substance Use Disorder (2 Indicators of substance use disorder during baseline years (pre-year 1 and indicators) pre-year 2). Hypertension (1 indicator) Indicator of hypertension during baseline. Diabetes (1 indicator) Indicator of diabetes during baseline. CDPS score (2 variables) CDPS score in each baseline year. Service Utilization and Estimated Medi-Cal Payments (11 variables) Utilization differential (6 Change in emergency department, hospital, mental health services, variables) substance use disorder services, primary care services, and specialty services utilization from pre-year 1 to pre-year 2. Cost differential (5 variables) Change in total, emergency department, hospital, outpatient and outpatient prescription costs from pre-year 1 to pre-year 2. For a limited number of enrollees (n=6,694) that did not have any baseline data, UCLA identified controls based on age group, gender, race, county, and whether they experienced homelessness during the first year of the program. Furthermore, for enrollees with only one year of baseline data (n=26,706), UCLA identified controls based on the total estimated costs and utilization rates rather than the differential between the two baseline years. Due to the phased implementation of WPC, UCLA grouped WPC enrollees into 20 cohorts based on the quarter in which they enrolled and selected a potential pool of control beneficiaries for each cohort. This method ensured that the control group beneficiaries had a similar baseline period to their matched enrollee. To select the final matched control group, UCLA used the Matchit package in R to estimate a propensity score in generalized additive models for modeling non-linear effects and avoiding overfitting using the variables in Exhibit 188 to identify two controls for each enrollee. UCLA used sampling with replacement. The final control group to WPC enrollee ratio was 1.75. To balance the sample, each control group beneficiary that was matched to multiple WPC enrollees was included in the control sample separately for each enrollee, resulting in two matched controls for each enrollee. Exhibit 189 shows the characteristics of enrollees and their matched controls with two years of baseline data and effect of the matching. Data showed that the balance between WPC enrollees and controls improved for nearly all indicators and variables, particularly for measures of utilization and cost. eile Appendix A: Data Sources and Analytic Methods for Quantitative Analysis | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 Exhibit 189: Comparison of Select Characteristics of WPC Enrollees with Two Years of Baseline Data and Matched Control Beneficiaries Before Match After Match WPC Enrollees (n= 200,030) Control Group (n = | Control Group (n= 2 400,060) 400,060) Age (at time of %0-17 2% 4% 4% enroliment} % 18-34 31% 32% 33% % 35-49 27% 24% 25% % 50-64 32% 28% 27% % 65+ 8% 12% 10% Gender % male 54% 52% 54% Race/Ethnicity % White 26% 25% 27% % Latinx 27% 40% 38% % African American 24% 12% 13% % Asian 6% 10% 8% % Other or Unknown 16% 14% 14% Homelessness UCLA-constructed 45% 18% 21% indicator . sa oO 32% 35% 34% conte Condition 12 38% 34% 36% gory 3+ 30% 31% 30% Hypertension 25% 25% 24% Select Chronic Diabetes 14% 16% 15% ae Serious Mental Illness 36% 17% 24% Conditions Substance Use 27% 13% 18% Disorders Emergency Department | -32 4 -18 Hospital Stays -11 2 -9 Utilization Differential | Mental Health Services | -137 -28 -102 in Baseline SUD services -69 -27 -61 Primary Care Services -68 -35 -56 Specialty Services -31 -36 -46 Total costs -222 56 -208 Emergency Department | -14 0 -13 Cost Differential in Hospital Stays -120 10 -110 Baseline Outpatient -56 731 “55 Outpatient Medication -1 6 -1 Long-Term Care Stays _| -12 [ 20 -11 For metrics that focused on specific subpopulations, UCLA developed unique matched control groups based on whether individuals met the denominator criteria (e.g., hospitalized for mental illness) before WPC, during WPC or is both time periods. Whole Person Care Final Evaluation Report | Appendix A: Data Sources and Analytic Methods for JESaE Quantitative Analysis RY ne olaye UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Difference-in-Difference Models UCLA assessed the impact of WPC for the overall WPC population and for enrollees with SMI/SUD or those experiencing homelessness (SMI/SUD/HML enrollees) and enrollees that were medically complex or otherwise high-risk (MC/HR enrollees) separately, using the difference-in-difference (DD) modeling approach. All models were controlled for demographics (gender, age, race/ethnicity, primary language, months of Medi-Cal enrollment), program characteristics (Pilot county, year of enrollment, and enrollment in HHP), acute care utilization indicator (at-risk, low, medium, high and super utilization), and health status indicators (baseline CDPS risk scores, specific baseline chronic conditions, and total count of chronic conditions at baseline). Additionally, models were adjusted for the number of full-scope Medi- Cal enrollment months and the number of months of WPC enrollment during the COVID-19 pandemic. UCLA used logistic regression models for binary metrics (e.g., Controlling High Blood Pressure), and Poisson models for utilization and cost variables (for inpatient and long-term care costs, UCLA used a zero-inflated count model with Poisson distribution). The exposure option within a Generalized Linear Model (GLM) was used to adjust for different number of months of Medi-Cal enrollment and the subsequent different lengths of exposure to WPC. All analyses of individual- level metrics were analyzed based on Medi-Cal member months. UCLA measured trends before and during WPC for each metric or measure based on the date of an individual WPC enrollee’s enrollment. UCLA examined changes in trends before and during WPC by modeling the changes in yearly increments up to 2 years (Pre-Year 1 and Pre-Year 2) before WPC enrollment and up to 5 years (Year 1, 2, 3, 4, and 5) during WPC. For these, the DD analysis measured the trends or change in yearly rates from Pre-Year 2 vs. Pre-Year 1 for both WPC enrollees and the control group; the change in the yearly rate during WPC from Year 1 to Year 5 for both WPC enrollees and the control group; and the difference between the changes in WPC enrollees vs. the control group from before to during WPC. The findings were not subject to potential seasonality in service utilization due to rolling enrollment throughout the year and measuring change following the date of enrollment per beneficiary. Limitations UCLA analysis of Medi-Cal data had limitations. One of the key target populations of WPC was individuals experiencing homelessness. However, Medi-Cal enrollment and claims data do not identify individuals that experience homelessness. As a result, UCLA created an indicator of homelessness based on Medi-Cal eligibility and claims data, which is likely subject to estimation error. The analysis in this report did not include complete claims data for the last four months Es Appendix A: Data Sources and Analytic Methods for Quantitative Analysis | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program of 2021. UCLA received data for those months after the current analyses were completed and further examination showed that DD findings did not change. The identification of chronic conditions may be subject to underreporting because due to use of primary and secondary diagnoses associated with each service.. UCLA was not able to find a control group that had similar levels of utilization or payments AND similar trends in utilization or payment prior to WPC enrollment. Therefore, the control group includes beneficiaries with higher or lower levels of utilization or payments at baseline than the WPC enrollees. Attributing Estimated Medi-Cal Payments to Claims Background The great majority of services under Medi-Cal are provided by managed care plans that receive a specific capitation amount per member per month and do not bill for individual services received by Medi-Cal beneficiaries. While managed care plans are required to submit claims to Medi-Cal, these claims frequently include payment amounts of unclear origin that are different from the Medi-Cal fee schedule. A small and unique subset of Medi-Cal beneficiaries are not enrolled in managed care and receive care under the fee-for-service (FFS) reimbursement methodology and have claims with actual charges and paid values. FFS claims are reimbursed primarily using fee schedules developed by Medi-Cal. The capitation amounts for managed care plans are developed using the same fee schedules by Mercer annually, using complex algorithms and other data not included in claims. To address the gaps in reliable and consistent payment data for all claims, UCLA estimated the amount of payment per Medi-Cal claim under WPC using various Medi-Cal fee schedules for services covered under the program. The methodology included (1) specifying categories of service observed in the claims data, (2) classifying all adjudicated claims into these service categories, (3) attributing a dollar payment value to each claim using available fee schedules and drug costs, and (4) examining differences between these and available external estimates. UCLA estimated payments for both managed care and FFS claims to promote consistency in payments across groups and to avoid discrepancies due to different methodologies. The payment estimates generated using this methodology are not actual Medi-Cal expenditures for health care services delivered during WPC. Rather, they represent the estimated amount of payment for services and are intended for measuring whether WPC led to efficiencies by Whole Person Care Final Evaluation Report | Appendix A: Data Sources and Analytic Methods for ESE} Quantitative Analysis RY ne olaye UCLA Center for Health Policy Research Health Economics and Evaluation Research Program reducing the total payments for WPC enrollees before and after the program, and in comparison, to a group of comparison patients in the same timeframe. Service Category Specifications Data Sources UCLA used definitions from multiple sources to categorize and define different types of services. These sources included Medi-Cal provider manuals, HEDIS value set, DHCS 35C File, American Medical Association’s CPT Codebook, National Uniform Code Committee’s taxonomy code set, and other available sources. DHCS’s Medi-Cal provider manuals included billing and coding guidelines for provider categories and some services. e The HEDIS Value Set by the National Committee for Quality Assurance used procedure codes (CPT and HCPCS), revenue codes (UBREV), place of service codes (POS), and Systematized Nomenclature of Medicine-Clinical Terms (SNOMED CT) to define value sets that measure performance in health care. For example, the HEDIS value set “ED” is a combination of procedure codes that describe emergency department services and revenue codes specifying that services were provided in the emergency room. e DHCS Paid Claims and Encounters Standard 35C File (DHCS 35C File) provided specifications to managed care plans on how claims must be submitted and contained detailed information about claims variables and their meaning and utility, such as vendor codes describing the location of services and taxonomy codes describing the type of provider and their specializations. e The American Medical Association’s Current Procedure Terminology (CPT) Codebook contained a list of all current procedural terminology (CPT) codes and descriptions that are used by providers to bill for services. e The National Uniform Claim Committee’s (NUCC’s) Health Care Provider Taxonomy code set identified provider types such as Allopathic and Osteopathic Physician and medical specialties such as Addiction Medicine defined by taxonomy codes. UCLA also used other resources to address gaps in definitions. For example, hospice codes that were used in claims submitted before 2016 were not included in the Medi-Cal provider manual, but UCLA collected the pre-2016 hospice codes from other DHCS guidelines. ENUM Appendix A: Data Sources and Analytic Methods for Quantitative Analysis | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program UCLA constructed eighteen mutually exclusive categories of service (Exhibit 190). Available Methods claims data included managed care, fee-for-service, and Short-Doyle. Some categories were defined using complementary definitions from more than one source. UCLA assigned claims to only one of the eighteen service categories to avoid duplication when calculating total estimated WPC payments. The outpatient services category may include claims included in other categories and therefore is not included in calculation of the total estimated payment in this report. UCLA assigned claims to the first service category a claim meets the criteria for as ordered in Exhibit 190. All services, apart from primary care visits, provided on the day of an ED visit were grouped as part of the ED visit to represent the total cost of the visit. For example, patients may have received transportation to an emergency department and laboratory tests during the emergency department visit, and these services were included in the ED category rather than the transportation or laboratory services categories. This approach may have included lab or transportation services in the ED category that were not part of the ED visit, and may have undercounted lab and transportation in their respective categories. However, this was necessary because claims data lacked information on the specific time of day when services were rendered. Similarly, all claims for services received during a hospitalization were counted as part of the same stay and were excluded from other categories of service, except for primary care visits on the day of admission. Other categories were identified solely by the procedure code or place of service and were not bundled with other services occurring on the same day, such as long-term care, home health/ home and community-based services, community-based adult services, FQHC services, labs, imaging, outpatient medication, transportation, and urgent care. Some claims lacked the information necessary to be categorized and were classified under an “Other Services” category. These frequently included physician claims without a defined provider taxonomy and durable medical equipment codes that were billed separately and could not be associated with an existing category. Exhibit 190: Description of Mutually Exclusive Categories of Service* folie [tg Service category Prana Descripti Tt 1 Emergency HEDIS Place of service is hospital emergency Department Visits room and procedure code is emergency (ED) service 2 Hospitalizations DHCS 35C File Place of service is inpatient and admission and discharge dates are present and are on different days Whole Person Care Final Evaluation Report | Appendix A: Data Sources and Analytic Methods for JREsEy Quantitative Analysis UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Ord Service category Pree} Todi acola} ts 3 Hospice Care DHCS 35C File, Provider is hospice or procedure code is HEDIS, and hospice service DHCS Medi-Cal Provider Manuals 4 Long-Term Care DHCS 35C File Claim is identified as LTC or provider is (LTC) Stays LTC organization; stays one day apart are counted as one visit, stays two or more days apart are separate stays 5 Home Health and DHCS 35C File Provider is a home health agency or Home and and DHCS Medi- | home and community-based service Community-Based | Cal Provider waiver provider, procedure is home Services (HH/HCBS) | Manuals health or home and community-based service 6 Community-Based | DHCS 35C File Provider is adult day health care center or Adult Services and DHCS Medi- | procedure code is community-based (CBAS) Cal Provider adult service, which are health, Manuals therapeutic and social services ina community-based day health care program 7 Federally Qualified | DHCS 35C File Provider is an FQHC or RHC (FQHC) and Rural Health Center (RHC) Services 8 Laboratory Services | DHCS 35C File Claim is identified as clinical laboratory, laboratory & pathology services, or laboratory tests 9 Imaging Services DHCS 35C File Claim is identified as portable x-ray services or imaging/ nuclear medicine services 10 Outpatient DHCS 35C File Claim is identified as pharmacy Medication 11 Transportation DHCS 35C File Claim is identified as medically required Services transportation 12 Primary Care National Provider is allopathic and osteopathic Services Uniform Claim physician (with specialization in adult Committee medicine, adolescent medicine, or geriatric medicine, family medicine, internal medicine, pediatrics, or general practice), or physician assistant or nurse practitioner (with specialization in ay) Appendix A: Data Sources and Analytic Methods for Quantitative Analysis | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Order Service category Praia Description ets medical, adult health, family, pediatrics, or primary care) 13 Specialty Care National Provider is allopathic and osteopathic Services Uniform Claim physician or physician assistant or nurse Committee practitioner (with all specializations not captured in the Primary Care Services category) 14 Outpatient Facility | DHCS 35C File Claim is identified as outpatient facility Services 15 Dialysis Services DHCS 35C File Provider is a dialysis center and and CPT procedure is dialysis Codebook 16 Therapy Services DHCS Medi-Cal | Procedure code is occupational, physical, Provider Manual | speech, or respiratory therapy 17 Urgent Care National Provider is ambulatory urgent care facility Services Uniform Claim Committee 18 Other Services N/A Provider, procedure, or place of service is not captured above N/A Outpatient Services | HEDIS Claim type is outpatient and procedure code, revenue code, or place of service code is outpatient (including FQHC). Source: UCLA Methodology. Note: * indicates categories are mutually exclusive except for outpatient services category UCLA found that four of the above categories made up the majority (87%) of total payments for WPC claims in 2019 (Exhibit 191). These categories were hospitalizations (37%), outpatient services (28%), outpatient medication (15%), emergency department visits (7%; Exhibit 191). Exhibit 191: Percentage of 2019 Total Estimated Payments by Category of Service for WPC Medi-Cal Claims Percentage of Total Category of Service lester) Pavan All Categories 100% Outpatient Services 28% Outpatient Medication 15% Emergency Department Visits 7% Hospitalizations 37% All other categories 13% Source: UCLA analysis of Medi-Cal Claims data from January 1, 2019 to December 31, 2019 Whole Person Care Final Evaluation Report | Appendix A: Data Sources and Analytic Methods for [ERMA Quantitative Analysis eer ee 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Attributing Payments to Specific Services To attribute payments to each category of service, UCLA developed methods to calculate an estimated payment for each category based on available data. Exhibit 192 displays the categories of service and what is included in the calculation of estimated payments for each category. Exhibit 192: Category of Service and Payment Descriptions Category of Service Emergency Department Visits (ED) Calculation of Estimated Payment Payments for all services taking place in the emergency department of a hospital, including services on the same day of the ED visit, excluding services by PCPs and FQHCs and RHCs. Two sub-categories are reported: ED visits followed by hospitalizations and all other ED visits that are followed by discharge. Hospitalizations Payments for all services that take place during a hospitalization, excluding visits with primary care providers on the first or last day of the stay, FQHC visits on the first or last day of the stay, or ED visits that preceded hospitalization Hospice Care Payments for hospice services in an LTC facility or Home Health setting, excluding hospice services rendered during a hospitalization Long-Term Care (LTC) Stays Institutional fees billed by LTC facilities; the per diem rate includes supplies, drugs, equipment, and services such as therapy Home Health and Home and Community-Based Services (HH/HCBS) Payments for services provided by a home health agency (HHA) and services provided through the home and community-based services (HCBS) waiver Community-Based Adult Services /(CBAS) Payments for community-based adult services and for services rendered at an adult day health care center Federally Qualified (FQHC) and Rural Health Center (RHC) Services Payments for all services provided in an FQHC or RHC Laboratory Services Payments for laboratory services, except those provided during a hospitalization or ED visit Imaging Services Payment for imaging services, except those provided during a hospitalization, ED visit, or LTC stay ENE Appendix A: Data Sources and Analytic Methods for Quantitative Analysis | Whole Person Care Final Evaluation Report (orto) ame) mS] a1(8 1 Outpatient Medication UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 Payments for outpatient drug claims, excluding prescriptions filled on the same day as an ED visit or on the day of discharge from a hospitalization Calculati of Estimated Payment Transportation Services Payments for medically required transportation, excluding transportation on the same day as an inpatient admission or an emergency department visit Primary Care Services Payments for services provided by a primary care physician Specialty Care Services Payments for services provided by a specialist, excluding services provided during an inpatient stay or an emergency department visit, and excluding facility fees Outpatient Facility Services Facility fees paid to hospital outpatient departments and ambulatory surgical centers Dialysis Services Payments for dialysis services rendered in a dialysis center Therapy Services Payments for occupational, speech, physical, and respiratory therapy services Urgent Care Services Payments for services provided in an urgent care setting Other Services Payments for services not captured above Outpatient Services Payments for all services delivered in an outpatient setting Source: UCLA Methodology. UCLA used all available Medi-Cal fee schedules and supplemented this data with other data sources as needed. Payment data sources, brief descriptions, and the related categories of services they were attributed to are provided in Exhibit 193. Exhibit 193: Payment Data Sources Sour Medi-Cal Physician Fee Schedule Annual files 2013 to 2021 inflated/ deflated to 2019 Description Contains rates set by DHCS for all Level | procedure codes that are reimbursable by Medi-Cal for services and procedures rendered by physicians and other providers Applicable Ser (eye Kola -19 ED, Hospitalizations, Hospice, LTC, HH/HCBS, CBAS, Imaging, Transportation, Primary Care, Specialty Care, Dialysis, Urgent Care, Other, and Outpatient Services Durable Medical Equipment (DME) Fee Contains rates set by CMS for Level II procedure codes for durable medical ED, Hospitalizations, Hospice, LTC, HH/HCBS, Whole Person Care Final Evaluation Report | Appendix A: Data Sources and Analytic Methods for Quantitative Analysis Source Schedule Annual files 2017 to 2021 inflated/ deflated to 2019 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Descripti equipment such as hospital beds and accessories, oxygen and related respiratory equipment, and wheelchairs Applicable Service Categories CBAS, Transportation, Primary Care, Specialty Care, Dialysis, Urgent Care, and Other Average Sales Price Data (ASP) for Medicare Part B Drugs Annual files 2014 to 2021 inflated/ deflated to 2019 Contains rates set by CMS for procedure codes for physician-administered drugs covered by Medicare Part B ED, Hospitalizations, Hospice, LTC, Primary Care, Specialty Care, and Other CMS MS-DRG grouping software, DHCS’s APR- DRG Pricing Calculator 9/30/2021 deflated to 2019 Contains Diagnostic Related Grouping (DRG) codes used for hospitalizations (CMS), base rate per DRG (DHCS) and DRG weights (CMS) Hospitalizations, LTC FQHC and RHC Rates 12/19/2018 inflated to 2019 Contains rates set by DHCS for services provided by FQHCs and RHCs FQHC and RHC Rates reimbursement rates set by DHCS for Hospice per diem rates | Contains rates set by DHCS for hospice Hospice Annual files 2020 and stays and services 2021 deflated to 2019 Nursing Facility Level A | Contains per diem rates set by DHCS per | LTC, Hospice per diem rates county for Freestanding Level A Nursing Annual files 2019, and Facilities 2020 and 2021 (deflated to 2019) Distinct Part Nursing Contains per diem rates set by DHCS for | LTC, Hospice Facilities, Level B nursing facilities that are distinct parts Annual files 2019, and of acute care hospitals 2020 and 2021 (deflated to 2019) Home Health Services Contains billing codes and Home health eye Appendix A: Data Sources and Analytic Methods for Quantitative Analysis | Whole Person Care Final Evaluation Report Source Annual files 2020 to 2021 deflated to 2019 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Descripti procedure codes reimbursable by home health agencies December 2022 Applicable Service Categories Home and Community- Based Services Rates 8/1/2020 deflated to 2019 Contains billing codes and reimbursement rates set by DHCS for the home and community-based services program Home and community- based services Community-Based Adult Services Rates 8/1/2020 deflated to 2019 Contains billing codes and reimbursement rates set by DHCS for community-based adult services Community-based adult services National Average Drug Acquisition Cost NADAC) File Annual files 2019, and 2020 and 2021 (deflated to 2019) Contains per unit prices for drugs dispensed through an outpatient pharmacy setting based on the approximate price paid by pharmacies, calculated by CMS Outpatient medication Clinical Laboratory Fee | Contains rates set by CMS for clinical lab | Laboratory Schedule services Annual files 2019, and 2020 and 2021 (deflated to 2019) Therapy Rates Contains billing codes and Therapy 8/1/2020 reimbursement rates set by DHCS for deflated to 2019 physical, occupational, speech, and respiratory therapy Ambulatory Surgical Center (ASC) Fee Schedule Annual files 2019, and 2020 and 2021 (deflated to 2019) Contains billing codes and reimbursement rates set by CMS for facility fees for ASCs ED, Hospitalizations, Outpatient Facility Outpatient Prospective Payment System (OPPS) File Annual files 2019, and Contains billing codes and reimbursement rates set by CMS for facility fees for hospital outpatient departments ED, Hospitalizations, Outpatient Facility Whole Person Care Final Evaluation Report | Appendix A: Data Sources and Analytic Methods for Bal Quantitative Analysis eer ee 2 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Source Description Applicable Service Categories 2020 and 2021 (deflated to 2019) Payments were attributed based on available service and procedures codes included in each claim. A specific visit may have included a physician claim from the providers for their medical services and a facility claim for use of the facility and resources (e.g., medical/ surgical supplies and devices) where service was provided. The Medi-Cal Physician Fee Schedule contained monthly updated rates for all procedures that were reimbursable by Medi-Cal to providers and hospital outpatient departments. Each procedure code had multiple rates that varied based on provider type (e.g. physician, podiatrist, hospital outpatient department, ED, community clinic) and patient age. UCLA distinguished between these rates, but the paid amount for FFS still varied within the same procedure code, likely due to the directly negotiated rates between the providers and DHCS. For the purpose of WPC cost evaluation, UCLA used the procedure code with the most expensive rate when adequate information was lacking. UCLA also included a payment augmentation of 43.44% for claims for physician services provided in county and community hospital outpatient departments following DHCS guidelines. UCLA did not include any other reductions or augmentations that may have been applied by Medi-Cal due to limited information in claims data. Some procedures such as those performed by a qualified physical therapist in the home health or hospice setting did not have a fee in the Medi-Cal physician fee schedule but had fees in the Medi-Cal Provider Manual and UCLA used these fees when applicable. A number of claims lacked procedure codes but had a revenue code such as “Emergency Room- General” or “Freestanding Clinic- Clinic visit by member to RHC/FQHC”. UCLA obtained documentation from DHCS that enabled identification of a price using outpatient revenue codes alone. CMS’s Durable Medical Equipment (DME) Fee Schedule included billing codes that are reimbursable by Medi-Cal for DMEs such as hospital beds and accessories, oxygen and related respiratory equipment, and wheelchairs. Rates for other medical supplies such as needles, bandages, and diabetic test strips were found in DHCS’s Medical Supplies Fee Schedules. Eyy Appendix A: Data Sources and Analytic Methods for Quantitative Analysis | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program FQHCs and RHCs consist of a parent organization with one or more clinic sites and are paid a bundled rate for all services during a visit. DHCS publishes FQHC and RHC Rates for each clinic within the parent organization. Payments for outpatient medication claims were calculated using the national drug acquisition cost (NADAC), which contains unit prices for drugs. UCLA calculated the drug cost by multiplying the unit price by the number of units seen on the claim. Drugs administered by physicians were priced using CMS’s Average Sales Price Data (ASP) for Medicare Part B drugs. Facility fees were priced based on the ambulatory surgical center (ASC) fee schedule or the outpatient prospective payment system (OPPS) depending on whether the billing facility was an ASC or an outpatient department. Medi-Cal paid most LTC institutions such as nursing and intermediate care facilities for the developmentally disabled on a per-diem rate, while long-term care hospital stays were reimbursed via diagnosis related group (DRG) payments. Per diem rates for LTC facilities were obtained directly from DHCS’s long-term care reimbursement webpage, and these rates varied by type of facility. Rates for hospice services were based on DHCS’s hospice care site and hospice room and board rates were based on the Nursing Facility/ Intermediate Care facility fee schedule, UCLA lacked some variables in claims data that were needed to calculate some LTC and hospice payments, such as accommodation code which specifies different rates for each nursing facility depending on the type of program including the “nursing facility level B special treatment program for the mentally disordered” or “nursing facility level B rural swing bed program”. In these cases, UCLA used the rates associated with accommodation code 1: “nursing facility level B regular”, which were higher than other accommodation code rates. Hospitalizations are paid based on diagnosis related groups (DRGs), a bundled prospective payment methodology that is inclusive of all services provided during a hospitalization, except for physician services. Identification and pricing of DRGs varies by payers such as Medi-Cal and Medicare. In California, DHCS uses 3M’s proprietary APR-DRG Core Grouping Software to assign DRGs and 3M’s APR-DRG Pricing Calculator to calculate prices for Medi-Cal DRG hospitals. APR- DRGs have more specific DRGs for Medicaid populations such as pediatric patients and services such as labor and delivery, and incorporate four levels of illness severity. However, UCLA did not have access to this software and used 3M’s publicly available CMS MS- DRG grouping software for the Medicare population, which includes Medicare-Severity DRGs (MS-DRGs) and their corresponding weights. MS-DRGs only include two levels of severity of illness, with complications or without complications. UCLA used this software to assign a DRG to each hospitalization based on procedure code, diagnosis, length of stay, payer type, patient Whole Person Care Final Evaluation Report | Appendix A: Data Sources and Analytic Methods for Quantitative Analysis UCLA Center for Health Policy Research December 2022 - ; Health Economics and Evaluation Research Program discharge status, and patient age and gender. Although CMS uses the Inpatient Prospective Payment System to assign hospital prices based on the MS-DRGs, UCLA used available data and publicly available prices for DHCS’s APR-DRG Pricing Calculator to calculate payments for each DRG. DHCS’s APR-DRG Pricing Calculator used multiple hospital and patient-level variables to calculate the final payment for hospitals, and UCLA incorporated some of these variables into the estimated payment (such as patient age and hospital status of rural vs. urban) but could not incorporate other modifiers due to data limitations (such as other health coverage and whether or not the hospital was an NICU facility). UCLA calculated the estimated payment by starting with the base rate from DHCS’s APR-DRG Calculator, which was $12,832 for rural hospitals and $6,507 for urban hospitals. This base rate was multiplied by the weight assigned to each MS-DRG, which modified the base rate to account for resources needs for a given DRG. For example, more severe hospitalizations such as “Heart Transplant or Implant of Heart Assist System with major complications” had a high weight of 25.4241 but “Poisoning and Toxic Effects of Drugs without major complication” had a lower weight of 0.7502. This rate was further modified by one available policy adjuster, which increased the payment amount by patient age and was higher for those under 21 (1.25) than those 21 and older (1). Overall payment for a hospitalization was calculated by adding the estimated payments for physician specialist services that occurred during the hospitalization. When no fees were found for procedure codes in any payment data sources, UCLA used the most frequent paid amount seen in fee-for-service claims for the procedure code. These included procedures such as tattooing/ intradermal introduction of pigment to correct color defects of skin and excision of excessive skin. When outlying units of service were found on the claim, UCLA used the 90" percentile value of units for the procedure code rather than the observed units. All claims were included in a category of service and were assigned a price. For dual beneficiaries, Medi-Cal is the secondary payer (payer of last resort) and covers a portion of the costs of the service. However, UCLA lacked information on percentage of services paid for by Medi-Cal for dual managed care beneficiaries. Therefore, UCLA used Medi-Cal claims data to calculate payments for these dual beneficiaries using the same methodology as non-dual managed care beneficiaries. Dual beneficiaries made up 15% of the beneficiaries population in 2019. For the purpose of evaluation, all payments were calculated using the 2019 fee schedules when available. In the absence of 2019 data, UCLA inflated or deflated payment amounts using the paid amounts for similar FFS claims in available data. Using the 2019 fees removed the impact of inflation and pricing changes in subsequent analyses. EYL) Appendix A: Data Sources and Analytic Methods for Quantitative Analysis | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Comparison of Estimated Payments with Medi-Cal Paid Amounts UCLA examined the potential bias that may have resulted due to the methodology used to estimate payments by comparing the estimated FFS payments with Medi-Cal paid amounts in FFS claims. Exhibit 194 shows that the estimated FFS payments were 7% lower than paid amounts for all services. There was underlying variation by category of services. For example, outpatient medication payments were 3% higher while estimated payments for hospitalizations were 8% lower. Exhibit 194: Comparison of Estimated Fee-for Service Payments and Paid Amounts for 2019 WPC Medi-Cal Claims Difference Between Estimated Cati f Servi Mat ON Se Payment and Medi-Cal Payment All Categories -7% Outpatient Services -5% Outpatient Medication 3% Emergency Department Visits -7% Hospitalizations -8% All other categories -16% Source: UCLA analysis of Medi-Cal Claims data from January 1, 2019 to December 31, 2019. UCLA further compared the difference in estimated payments for FFS and managed care claims and found that managed care payments were 26% lower than the FFS claims ($226 vs $168; Exhibit 195). Exhibit 195: Comparison of Average Fee-for-Service and Managed Care Payments per Claim for 2019 WPC Medi-Cal Claims Average Estimated Payment per Claim for Managed Care ey $226 $168 Source: UCLA analysis of Medi-Cal Claims data from January 1, 2019 to December 31, 2019. Average Medi-Cal Payment per Claim for FFS Claims Limitations There were limitations associated with UCLA’s payment estimates including the availability of needed data and access to fee schedules and other pricing resources. UCLA did not aim to calculate exactly what DHCS paid for claims, but rather to measure the impact of WPC on cost compared to the control group. The reasons for differences between costs and estimated payments are described below. Whole Person Care Final Evaluation Report | Appendix A: Data Sources and Analytic Methods for Quantitative Analysis UCLA Center for Health Policy Research December 2022 - ; Health Economics and Evaluation Research Program The first limitation was related to using the MS-DRG relative weights for Medicare for hospitalization, which were higher than Medi-Cal. This likely led to higher estimated payments for hospitalization. Second, MS-DRG only identified the levels of severity as with and without complication rather than four level used by APR-DRG. Third, DHCS uses multiple criteria to adjust hospital payments but UCLA was only able to adjust for urban and rural rates. A second limitation was related to availability of fee schedules for accurate pricing. The WPC evaluation required analysis of multiple years of claims data and UCLA used all available fee schedules to price procedures, supplies, and facilities from multiple years and inflated prices to 2019 dollars whenever necessary. UCLA always used the most recent rate for a procedure. The inflation rates used were based on medical care Consumer Price Index provided by US Bureau of Labor Statistics without adjusting for regional-specific inflation rates. Not all procedures that appeared in the claims data had corresponding rates in all the available fee schedules. Procedures that required Treatment Authorization Requests (TARs) lacked a fee-schedule and are frequently more expensive than covered services. Some specific procedures had no fees in the Medi-Cal fee-schedule. When fee schedules were missing, UCLA attributed the most frequently observed price from the paid amount for a similar FFS claim. If the procedure did not appear in any FFS claims, UCLA assigned the median allowed amount from all managed care claims for the given procedure code. A third limitation was related to outlier values for service units, some of which were extremely high. UCLA attributed the 95" percentile value instead of the original value in the claim, potentially underestimating payments for some claims. Finally, UCLA modeled the estimated total payments and payments for each category of service separately. As a result, it was not possible to present the component categories as a proportion of the total payments. Given the differences in approach to costing each category of service and the resulting differences in error and biases, presenting the categories in comparison to one another and as part of the total, may lead to misinterpretations. e¥-m) Appendix A: Data Sources and Analytic Methods for Quantitative Analysis | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Appendix B: Data and Analyses Methods for Pilot- Reported Metrics Overview of Data and Analysis Methods for Self-Reported Metrics Overview of Self-Reported Metrics DHCS required Pilots to regularly report on fifteen DHCS-defined metrics to track progress in better care and better outcomes for WPC enrollees. All Pilots participating in WPC were required to report on a specific subset of five metrics, called “universal metrics” that were collected from all Pilots. The universal metrics were: (1) Ambulatory Care Emergency Department Visits per 1,000 WPC Member months; (2) Inpatient Utilization per 1,000 WPC Member Months; (3) Follow-Up After Hospitalization for Mental Illness; (4) Initiation and Engagement of Alcohol and Other Drug Dependence Treatment, and (5) Comprehensive Care Plan completion. DHCS also required Pilots to select at least four additional metrics out of the remaining ten metrics, called “variant metrics.” Some Pilots changed their variant metrics during WPC implementation due to data collection challenges or changes to strategies or target populations. Under WPC, progress in metrics was compared after enrollment to the baseline period. For quantitative health care utilization metrics, DHCS designated PY 1 as the baseline period and Pilots gathered this data retrospectively for individuals who were enrolled in the first 18 months of WPC enrollment (1/1/2017 to 6/30/2018). For these metrics, progress was measured starting in PY 2. For other quantitative metrics, the baseline period was PY 2 for individuals who were enrolled in the first 18 months of WPC enrollment to allow Pilots to gather this data. For these metrics, progress was measured starting in PY 3. Data Source UCLA analyzed Pilot-reported metrics from the Annual WPC Variant and Universal Metric Reports reported to DHCS. Data included the rate and the numerator and denominator used to calculate that rate, for each metric annually. A limited number of metrics were also reported semi-annually, but these data were not included in the analysis. Additionally, metrics that UCLA was able to recreate using Medi-Cal data (Ambulatory Care Emergency Department Visits per 1,000 WPC Member months, Inpatient Utilization per 1,000 WPC Member Months, Follow-Up Whole Person Care Final Evaluation Report | Appendix B: Data and Analyses Methods for Pilot-Reported Metrics RY ne olaye UCLA Center for Health Policy Research Health Economics and Evaluation Research Program After Hospitalization for Mental Illness, and Initiation and Engagement of Alcohol and Other Drug Dependence Treatment) were not included in this analysis. Methods UCLA calculated the weighted average for each metric by summing the numerators and the denominators separately for all Pilots that reported data, and then dividing the overall numerator by the overall denominator. Pilots may not have reported data if they had limited enrollment during the measurement period or had other constraints on data availability. When the Pilot reported zero or no values, UCLA examined the reports to determine if the Pilot did not report the metric at all, or if the numerator was zero. UCLA excluded Pilots from the analyses who did not report a value. Detailed Methods by Self-Reported Metric This section describes the details of the methods that Pilots used to calculate each of the self- reported metrics, and includes: e An overview of the metric and any sub-metrics. « Measurement specifications, including the numerator and the denominator. « The baseline period, baseline population, and frequency of reporting. « Asummary of whether Pilots reported on this metric in each year. The details in this section are based on the Whole Person Care Universal and Variant Metrics Technical Specifications Guide revised by DHCS on March 22, 2019, and on the WPC Variant and Universal Metrics Report spreadsheet that included instructions for Pilots regarding how to report on the universal and variant self-reported metrics. Variant Metric: Control Blood Pressure Pilots reported the percent of enrollees whose blood pressure was adequately controlled during the measurement year. Three sub-metrics were reported: (1) the percent of enrollees with hypertension age 18-59, whose blood pressure was less than 140/90 mm hg, (2) the percent of enrollees with hypertension age 60-85 with a diagnosis of diabetes, whose blood pressure was less than 140/90 mm Hg, and (3) the percent of enrollees with hypertension age 60-85 without a diagnosis of diabetes, whose blood pressure was less than 150/90 mm Hg. This metric was modeled on the HEDIS Controlling High Blood Pressure metric. However, the official HEDIS measure was revised in 2019, after implementation of data collection for WPC, and no longer distinguishes between the three groups based on age and diabetes status. ey} Appendix B: Data and Analyses Methods for Pilot-Reported Metrics | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program For each of the three sub-metrics, Pilots calculated the percent of enrollees with controlled blood pressure by dividing a numerator (number with controlled blood pressure) by a denominator (number in the group). The denominator consisted of a subset of all individuals enrolled in WPC at any time during the measurement year who were of the appropriate age and diabetes status for each of the three sub-metrics, and had at least one outpatient visit with a diagnosis of hypertension during the first six months of the measurement year. Enrollees were excluded from the denominator if they used hospice services or a hospice benefit during the measurement year. The numerator consisted of the number of members in the denominator whose most recent blood pressure (both systolic and diastolic) was adequately controlled. This most recent blood pressure reading must have occurred after the diagnosis of hypertension. If multiple blood pressure measurements occurred on the same date, or were noted in the chart on the same date, then the lowest systolic and lowest diastolic blood pressure readings were used. If no blood pressure was recorded during the measurement year, then the enrollee was assumed to have uncontrolled blood pressure. The baseline period consisted of calendar year 2016 (January 1, 2016 through December 31, 2016). Because no one was enrolled in WPC during the baseline period, Pilots defined the baseline population as the cohort that was enrolled in WPC from January 1, 2017 through June 30, 2018, per DHCS specifications. Pilots then gathered Medi-Cal data retrospectively for the baseline year for this enrollee population. This metric was reported annually. Whole Person Care Final Evaluation Report | Appendix B: Data and Analyses Methods for Pilot-Reported Metrics yoday uojenjeag Jeulj a4eD UOsJdd BJOYM | SII) Payoday-jojig 10) sSpoyray sasAjeuy pue ejeq :q xIpuaddy [Mejsts o9sIsUes4 UN x UN x UN x UN x UN x uN Xx ues UN x UN x UN x UN x UN x uN x o8aiq ues x x x x x x oulpseusag ues UN x uN x UN x uN x UN x uN xX oquaweses x x x x Vv x x ApIsIaAlY UN x YUN x UN x UN x UN x UN x 492e|d UN x UN x UN x UN x UN x UN x asuelo UN x uN x UN x UN x uN x uN xX eden x x x x x x AasauoW x x x x x x oulopual, UN x UN x UN x uN x UN x uN x ue UN x YUN x UN x UN x YUN x UN x sajasuy so] UN x UN x UN x uN x UN x uN xX s8ury x x x x x x usay x x x x x x EIS09) UN uN UN UN uN UN enuo9 x x x Vv x x x epowely m m m m m m rs s x x x x a a 2 a a a is m = c m at < m a © m = < m = © m = eee eS eet elle hele = ello 8s Dp a a av a a se ees See eS ase es oS |e eg es a a a a a a ° ° ° ° ° ° SS a as oe ao an (g sea, quauyjoiug | (7 sea, quaUUyjosug | (¢ sea, qUaUjOsUg | (Z4eaA JUaLUJJOIU | (T 4eaA JUaWI|OIUg (auyaseg “L@0Z) 9Ad ‘0Z0Z) SAd “6L0Z) tAd “810Z) €Ad ‘LU0Z) ZAd ‘9T0Z) TAd 6S-8T asy ‘aINSsalg POO|g [O1}UOD :IJa\ JUBA JOJ Bulioday :96T Wqlyxy WweJZolq yoJeasay Uo!enjenq pue s21WoU0r] YIeaH yoseasay Adljog yrjeaH JJ Ja}UaD VION raavya@ecye era) nas (powad siy3 404 Yoda 0} Y8nou—a A\Jea ulZaq jou pip sa: (pouiad siya 404 ajqejiene you sem eyep nq SOLA) payoday-jo|Id 10) spoyuyayy sasAjeuy pue ejeq :g xipuaddy | yoday uonenjeng jeuly a1eD UOSJad BJOUM (spoisad asay Jo |Je ye D439W S14} UO Yoda JOU pip 37 B43) BuJOday ION :YN Woe WesBosd Jo JuaU|}osUa Ing jaw S1y} UO Suodad sem 37 ayy) JUaW|JOIUJ :3 1uJ9W S14} UO BuJOdad sem 37 ay) Art|Iq :suosead UOISN|9x3 ; x x x x Vv x x eanquan UN x UN x UN x UN x UN x UN x eBwouos a x uN x UN x uN x UN x yN Xx ouejos UN x UN x UN x UN x YN x UN x eIseys ad x UN x UN x UN x UN x UN x IDdMIS x x x x x x ZndJp ejueS UN x UN x UN x UN x UN x UN x eye|D Blues UN x UN x UN x UN x YN x UN x oajyeW ues uN x uN x uN x uN x UN x uN Xx uinbeor ues on on m on mn mn g B g £ g 2 a 2 2 2 S c m = £ m = 2 m = c m = e m = © m = id gg 4 81/8 & BIS & B/S & B/S & BIS 4 B = & & | 5 Sao aes Sse Sse Sse eS Da a nz nz a a 3 a 3&8 | 8 Seis a8 8 | 8 a8 a] 8 a ae8/8 8 8 y0l1g a a 8 a a B ! So ° ° ° ° So as a = = 7 = (g sea,Q quauyjosug | (p4ea, quawyjosug | (¢ 4eaA JUaUjOUUA | (Z4eaA JUaUJOIUA | (T 4eaA UaW;OIUR (auyjaseg “L@0Z) 9Ad “070Z) SAd “60Z) tAd “8102) Ad ‘LU02) ZAd ‘9102) TAd ae teeet WeiZ01g YDJeasay UO!EN|eAW pue sd|WwoUdy Yyeaq Yoseasay Aotjod YiJeaH JOJ JajUaD VION yoday uojenjeag Jeulj a4eD UOsJdg BJOYM | SII) Payoday-jojig 10) spoyray sasAjeuy pue ejeq :q xIpuaddy [Asta o9sIsUes4 UN x UN x UN x UN x YN x uN xX ues UN x UN x UN x UN x UN x uN x o8aiq ues x x x x x x oulpseusag ues UN x uN x UN x uN x UN x uN xX oquaweses x x x x Vv x x ApIsIaAlY UN x YUN x UN x UN x UN x UN x 492e|d UN x UN x UN x UN x UN x UN x asuelo UN x uN x UN x UN x uN x uN xX eden x x x x x x AasauoW x x x x x x oulopual, UN x UN x UN x uN x UN x uN x ue UN x YUN x UN x UN x YUN x UN x sajasuy so] UN x UN x UN x uN x UN x uN xX s8ury x x x x x Vv x ay x x x x x x EIS09) UN uN UN UN uN UN enuo9 x x x Vv x x x epowely m m m m m m rs s x x x x a a 2 a a a is m = c m at < m a © m = < m = © m = ee ee see tells kee = ello 8s Dp a a av a a Bs eee Ss kes eS es se | eg es a a 2 a a a ° ° ° ° ° ° SS a a oe ao an (g sea, quauyjoiug | (7 sea, quaUUyjosug | (¢ sea, qUaUjOsUg | (Z4eaA JUaLUJJOIU | (T 4eaA JUaWI|OIUg (auyaseg “L@0Z) 9Ad ‘0Z0Z) SAd “6L0Z) tAd “810Z) €Ad ‘LU0Z) ZAd ‘9T0Z) TAd salaqeig YUM ‘Gg-09 asy ‘auNssaid POO|g [O1]UOD :IJapy JUBA JOj Buloday :/6T Uqiyxy weiZolq YoJeasay uolenjenq pue s21wWoU0sg YIeaH yoseasay Adljog yrjeaH JJ Ja}UaD VION raavya@ecye era) (S45 (powad siy3 404 Yoda 0} Y8nou—a A\Jea ulZaq jou pip sa: (pouiad siya 404 ajqejiene you sem eyep ing SOLA) payoday-jo|Id 10) spoyuyayy sasAjeuy pue ejeq :g xipuaddy | yoday uonenjeng jeuly a1eD UOSJad BJOUM (sposad asay Jo} |Je ye D439W S14} UO Yoda JOU pip 37 B43) BuOday ION :N Woe WesBosd Jo JuaU|}osUa Ing jaw S1y} UO Suodad sem 37 ayy) JUaW|JOIUJ :3 1uJ9W S14} UO BuJOdad sem 37 ay) Art|Iq :suosead UOISN|9x3 ; x x x x Vv x x eanquan UN x UN x UN x UN x UN x UN x eBwouos a x uN x UN x uN x UN x yN Xx ouejos UN x UN x UN x UN x YN x UN x eIseys ad x UN x UN x UN x UN x UN x IDdMIS x x x x x x ZndJp ejueS UN x UN x UN x UN x UN x UN x eye|D Blues UN x UN x UN x UN x YN x UN x oajyeW ues uN x uN x uN x uN x UN x uN Xx uinbeor ues on on m on mn mn g B g £ g 2 a 2 2 2 S c m = £ m = 2 m = c m = e m = © m = id gg 4 81/8 & BIS & B/S & B/S & BIS 4 B = & & | 5 Sais Sse Sse Sse eS Da a Dp nz a a 3 a 3&8 | 8 £228 a8 8 | 8 a8 a] 8 a ae8/8 8 8 y0l1g a a 3 a a B ! So ° So ° ° So as a a = 7 = (g sea,Q quauyjosug | (p4ea, quawyjosug | (¢ 4eaA JUaUjOuUZ | (Z4eaA JUaUJOIUA | (T 4eaA JUaW;OIUR (auyjaseg “L@0Z) 9Ad “070Z) SAd “610Z) tAd “8102) Ad ‘LU02) ZAd ‘9102) TAd ae teeet WesZ01g YDJeasay UO!eEN|eAW pue sd!wWwoUdy Yyeaq Yoseasay Aotjod YiJeaH JOJ JajUaD VION yoday uojenjeag Jeulj a4eD UOsJdd BJOYM | SIJa\| Payoday-jojig 40) spoyray SasAjeuy pue ejeq :q xIpuaddy [RAss o9sIsUes4 UN x UN x UN x UN x YN x uN xX ues UN x UN x UN x UN x UN x uN x o8aiq ues x x x x x x oulpseusag ues UN x uN x UN x uN x UN x uN xX oquaweses x x x x Vv x x ApIsIaAlY UN x YUN x UN x UN x UN x UN x 492e|d UN x UN x UN x UN x UN x UN x asuelo UN x uN x UN x UN x uN x uN xX eden x x x x x x AasauoW x x x x x x oulopual, UN x UN x UN x uN x UN x uN x ue UN x YUN x UN x UN x YUN x UN x sajasuy so] UN x UN x UN x uN x UN x uN xX s8ury x x x x x x usay x x x x x x EIS09) UN UN UN UN uN UN enuo9 x x x Vv x x x epowely m m m m m m rs s x x x x a a 2 a a a is m = c m at < m a © m = < m = © m = ee ee Sele t elle hele = ello se Dp a a av a a e 8 @/8 & 28/8 & 8/8 & BF F B/E BF Bl ay a a 2 a a a L ° ° ° ° ° ° SS a a oe ao an (g sea, quauyjoiug | (7 sea, quaUUyjosug | (¢ sea, qUaUjOsUg | (Z4eaA JUaLUJJOIU | (T 4eaA JUaWI|OIUg (auyaseg “L@0Z) 9Ad ‘0Z0Z) SAd “6L0Z) tAd “810Z) €Ad ‘LU0Z) ZAd ‘9T0Z) TAd sajaqgeig INoyUM ‘Gg-09 as8y ‘aiNssaig POO|g [O1]UOD :dUJaIA] JUBLIEA :8ET UGIYXA weiZolq YoJeasay uolenjenq pue s21wWoU0sg YIeaH yoseasay Adljog yrjeaH JJ Ja}UaD VION raavya@ecye era) TSS (powad siy3 404 Yoda 0} Y8nou—a A\Jea ulZaq jou pip sa: (pouiad siya 404 ajqejiene you sem eyep ing SOLA) payoday-jo|Id 10) spoyuyayy sasAjeuy pue ejeq :g xipuaddy | yoday uonenjeng jeuly a1eD UOSJad BJOUM (sposad asay 404 |Je ye D439W S14) UO Yoda OU pip 37 B43) BuJOday ION :YN Woe WesBosd Jo yuaU|}osUa Ing jaw S1y} UO Suodad sem 37 ayy) JUaW|JOIUJ :3 1uJ9W S14} UO BuNJOdad sem 37 ay) Art|! :suosead UOISN|9x3 ; x x x x Vv x x eanquan UN x UN x UN x UN x UN x UN x eBwouos a x uN x UN x uN x UN x yN Xx ouejos UN x UN x UN x UN x YN x UN x eIseys ad x UN x UN x UN x UN x UN x IDdMIS x x x x x x ZndJp ejueS UN x UN x UN x UN x UN x UN x eye|D Blues UN x UN x UN x UN x YN x UN x oajyeW ues uN x uN x uN x uN x UN x uN Xx uinbeor ues on on m on mn mn g B g £ g 2 a 2 2 2 S c m = £ m = 2 m = c m = e m = © m = id gg 4 81/8 & BIS & B/S & B/S & BIS 4 B = & & | 5 Sais Sse Sse Sse eS Da a Dp nz a a 3 a 3&8 | 8 £228 a8 8 | 8 a8 a] 8 a ae8/8 8 8 y0l1g 3 a 3 a a B ! So ° So ° ° So as a a = 7 = (g sea,Q quauyjosug | (p4ea, quawyjosug | (¢ 4eaA JUaUjOUUA | (Z4eaA JUaUJOIUA | (T 4eaA UaW;OIUR (auyjaseg “L@0Z) 9Ad “070Z) SAd “60Z) tAd “8102) Ad ‘LU02) ZAd ‘9102) TAd ae teeet WesZ01g YDJeasay UO!EN|eAW pue sdiwWwoUdy Yyyea}q Yyoseasay Aoljod YiJeaH JOJ JaqUaD VION RY ne olaye UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Variant Metric: Incarcerations per 1,000 Member Months Pilots reported the number of incarcerations per 1,000 member months. Two sub-metrics were reported: (1) the number of incarcerations per 1,000 member months for those age 14 or older as of June 30 of the measurement year, mainly reported in mid-year reports, and (2) the number of incarcerations per 1,000 member months for those age 14 or older as of December 31 of the measurement year, mainly reported in annual reports. Because this analysis focused on annual data, only the second sub-metric was included in this report. Pilots calculated the incarceration rate by dividing a numerator by a denominator, and multiplying the result by 1,000. The denominator consisted of a count of member months for all individuals enrolled in WPC at any time during the measurement year. Member months were based on WPC enrollment rather than Medi-Cal enrollment. Enrollees were excluded from the denominator if they used hospice services or a hospice benefit during the measurement year. The numerator consisted of the total number of incarcerations experienced by those in the denominator population; one enrollee could have multiple incarcerations during the reporting period. The baseline period consisted of calendar year 2016 (January 1, 2016 through December 31, 2016). Because no one was enrolled in WPC during the baseline period, Pilots defined the baseline population as the cohort that was enrolled in WPC from January 1, 2017 through June 30, 2018, per DHCS specifications. Pilots then gathered Medi-Cal data retrospectively for the baseline year for this enrollee population. This metric was reported twice per year, once for the sub-metric that included those age 14 or older as of June 30 of the measurement year, and again for the sub-metric that included those age 14 or older as of December 31 of the measurement year. Eka Appendix B: Data and Analyses Methods for Pilot-Reported Metrics | Whole Person Care Final Evaluation Report SoH Vata Paoday-Jo|!d 10) spoujyay| sasAjeuy pue ejeq :g xipuaddy| woday uolenjeng jeuly aseD UOSIad BIOUM x ae x x 4 x x o8aiq ues oulpseuiag YN x UN x UN x YN x UN x UN x ues UN x UN x YUN x YUN x UN x YUN x ojusWwelseS Vv x v z x x x x apIssaAly UN x UN x UN x UN x uN x UN x 492e|d UN x UN x YN x UN x UN x UN x asueio UN x UN x UN x YN x YN x UN x eden UN x uN x YN x UN x YN x YN x Aasaquoyy UN x UN x UN x UN x UN x YN x oulsopuay| UN x UN x UN x YN x UN x UN x uel a * x x x x sajasuy so] x x x ® x x sBuly UN x YN x UN x YN x UN x UN x way x x x x x x E1509. UN UN UN UN YN UN eu]U0} UN x UN x UN x YN x UN x UN x epowely m m m m m m 3 rd x P+ xs P+ S s S Ss Ss a c rr _ © mn = c in = £ m = ce mn _ © m = ee elt = i232 =¢ i/2 2 iS = Gl2 & = D a a a a a Ce es es eo ee ec eS Se Sees ee eS a o a a 2 a 3 3 3 5 3 3 > = =. os SH oA (g aea, quauyjosug | (pea, quawjjoiug | (¢ see, quawjosug | (Zz 4ea, quaWjosug | (T 4ea, qUaW|jo1U (auyaseg “T@0Z) 9Ad ‘0Z02) SAd “610Z) vad “10Z) Edd ‘LU02) ZAd ‘9T0Z) TAd SUJUOW| Jaquial] OOO'T Jad suonesaoseou| ‘Jay JUeeA JOj Buloday :66T Wqlux4 WweJZ01g youeasay UOljeNjeAq pue s2!WOU0DJ Yea GAME ERPS) Yyoseasay Aotjod YiJeaH JO JajUaD YIN yoday uojenjeaq Jeulj a4eD UOsJdg BJOYM | SII) Payoday-jojid 40) spoyyay sasAjeuy pue ejeq :q xIpuaddy Piss (spowiad asayp Jo} |Je 3 21439 $14) UO Oda, JOU pip 37 B43) BulOday ON 7YN (poisad siy} 40} yoda 0} y8noua Ajjea ulZaq Jou pip saijiAij9e WesZOJd Jo JUaW|jOJUA yNq ‘911}9W S14} UO BuljJOdaJ Sem 3) B43) JUaW|}OIUJ :3 (pouiad siuy 404 ajqejiene you sem ejep yng ‘21JaW sI4} UO Buloday sem 37 ayy) AI[IGENEAY "Vv :Suosead UOISN|9X9 ; UN x UN x UN x UN x UN x UN x eIN{UaA, x x x x a x a x PWOUOS qa x UN x UN x UN x UN x UN x oue|os UN x UN x UN x UN x UN x UN x eseys a x UN x UN x UN x UN x UN x D9DdMIS UN x UN x UN x UN x UN x UN x zniQ ejues uN x uN x YN x UN x UN x uN xX eue[D eqUes YN x UN x UN x UN x UN x UN x oajzey ues x x x x x x uinbeor ues oosioues4 x x x x x x ues mn mn m on mn mn g B g £ g 2 a 2 2 2 S c m = £ m = 2 m = c m = e m = © m = id g$ 4 81/8 & BIS & B/S & B/S & BIS 4 B = & & | 5 Sais Sse Sse Sse eS Da a Dp nz a a 3 a 3&8 | 8 £228 a8 8 | 8 a8 a] 8 a ae8/8 8 8 Aol 3 a 3 a a B id So ° So ° ° So as a a = 7 = (g sea,Q quauyjosug | (p4ea, quawyjosug | (¢ 1eaA JUaUjOuUA | (Z4eaA JUaUJOIUA | (T 4eaA UaW|}OIUR (auyjaseg “L@0Z) 9Ad “070Z) SAd “60Z) tAd “8102) Ad ‘LU02) ZAd ‘9102) TAd WweJZolg Youeasay UolJenjenq pue s21WOU0Dg YIeaH yoseasay Adljog yijeaH JJ Ja}UaD YION raavyaeeyeeTa] UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Variant Metric: Overall Beneficiary Health Pilots reported the percent of enrollees that provided a self-reported rating of their health as “Excellent” or “Very Good.” Two sub-metrics were reported: (1) the percent of enrollees reporting “Excellent” or “Very Good” overall health, and (2) the percent of enrollees reporting “Excellent” or “Very Good” emotional health. This metric was constructed from the Consumer Assessment of Healthcare Providers and Systems (CAHPS) survey. For each of the two sub-metrics, Pilots calculated the percent of enrollees who rated their health as “Excellent” or “Very Good” by dividing a numerator (number that reported those levels of health) by a denominator (number that answered the survey questions). The denominator consisted of a subset of all individuals enrolled in WPC at any time during the measurement year, who were enrolled a total of six months in WPC during the measurement year with multiple allowable gaps. Enrollees were excluded from the denominator if they used hospice services or a hospice benefit during the measurement year. The numerator consisted of the number of responses with answers of “Excellent” or “Very Good,” and was calculated separately for overall health and for mental or emotional health. Unlike other WPC metrics, the baseline reporting period for this metric was calendar year 2017 rather than 2016. This is because data on this metric could not be gathered before WPC enrollment began. This metric was reported annually. Whole Person Care Final Evaluation Report | Appendix B: Data and Analyses Methods for Pilot-Reported Metrics yoday uonenjeng [euly 42D Uossad ajoym | SIA) PaOday-jojlq 10} spoyray) SasAjeuy pue ejeg :g xipuaddy [Zs x x x x x o9sIsUes4 UN uN UN UN uN iis UN x uN x UN x UN x uN x o8aiq ues % ” % x x oulpseusag ues x x x x x oquawiees x x x x x ApIsIaAlY UN x YUN x UN x YN x YN x da2e|q UN x UN x UN x UN x UN x asuelo x X x x Vv x eden UN x UN x UN x UN x UN % Aasaquo;: UN x YN x UN x YN x UN x oulsopuayy x x ¥ x 9 2 uuey YN x YN x UN x YN x YN x sajasuy so] UN x UN x UN x uN x UN x sBuly Vv x x x Vv x Vv * ay x x x x x E1509) e4yU0D, UN x uN x UN x UN x uN x epawely m m m m m m rs s x x x x a S a 5 S S is m = c m at < m a © m = c m = =| m = ee ee Ss eet elle kel ee ello 8 = Dp a a av a a #2 8/F &£€ 8/F £F 2/F EF RIF F RIF F FI] soy a o a a 2 a L ° ° ° ° ° ° SS eA, =. oe a es (g sea, quauyjoiug | (7 sea, quaUUyjosug | (¢ sea, qUaUjOsUg | (Z4eaA JUaLUJJOIU | (T 4eaA JUaWI|OIUg (auyaseg “TZ0Z) 9Ad ‘0Z0Z) SAd “6L0Z) tAd ‘10Z) Edd ‘LU0Z) ZAd ‘9T0Z) TAd YYEaH [[e49AO - YYeaH Ayelijauag ||eI9AQ ‘dja JUeWe/ JOJ Bulioday :00Z Uq!YyXy WweJZolg YoJeasay uolenjenq pue s21wWoU0sg YIeaH yoseasay Adijod Y}22H JO} 4822 VION aural eS Ta) aa (poisad siug 40} y10da1 0} yBnoua Aluea ulaq you pip sa (poiiad siup 40} ajgejiene you sem eyep yng SoWII payoday-jojid 10) spouyay] sasAjeuy pue ejeq :g Xipuaddy| yoday uoljenjeng jeuly aseD UOSIdd BJOUM (spotsad asay Jo jJe ye 2439W S14} UO Yoda JOU pip 37 B43) Buoday ION :N Wy9e WeJBOId Jo yuaW|joJUa yng au siyy Uo BuNsoday sem 3] ay3) Aditi jaw S14} UO Buodad sem 37 ayy) JUaW|JOIUJ :3 :suosead UOISN|Ix3 y UN x uN x UN x uN x uN x eanquan YN x UN x UN x UN x UN x ewouos a x uN x YN x uN x uN x oUuR|os YN x YN x YN x uN x YN x eyseys a x UN x UN x UN x UN x IDdMIS uN x UN x YN x YN x UN x zndp eques YUN x UN x UN x YN x UN x eue[D ejUeS UN x UN x UN x YN x UN x oajze ues YN x YN x UN x YN x UN x uinbeor ues mn mn om mn m m g g g 2 g g 2 2 a 2 5 2 c m = £ m = a m —_ c m = © m = © m = See ne ee Eales cece ele Dp a Dp a a a 8 a a 3 a g | 2 & & 3 a a eS a a |S 3g & yolld 2 a 8 a B 2 E °o ° °° ° ° So 3. ao ae oe = a (g sea,Q quauyjosug | (p sea, quawjosug | (¢ 4eaA JUaUjOIUA | (Z4eaA JUaUJOIUA | (T 4eaA JUaW}OIUR (auyaseg “1@0Z) 9Ad “00Z) SAd “610Z) tAd “810Z) €Ad ‘LU0@) ZAd ‘9L02) TAd ae teeet WesZ01g YDJeasay UO!eEN|eAW pue sdiwoUcdy Yyyeaq Yyoseasay Aotjod YiJeaH JOJ JajUaD YIN yoday uojenjeag Jeulj a4eD UOsJdg BJOYM | SII) Payoday-jojid 10} spoyyay| SasAjeuy pue ejeq :q xIpuaddy [raza x x x x x o9sIsUes4 UN uN UN UN uN iis UN x uN x UN x UN x uN x o8aiq ues % ” % x x oulpseusag ues x x x x x oquawiees x x x Vv x x ApIsIaAlY UN x YUN x UN x YN x YN x da2e|q UN x UN x UN x UN x UN x asuelo x X x x Vv x eden UN x UN x UN x UN x UN % Aasaquo;: UN x YN x UN x YN x UN x oulsopuayy x x ¥ x 9 2 uuey YN x YN x UN x YN x YN x sajasuy so] UN x UN x UN x uN x UN x sBuly Vv x x x Vv x Vv x ay x x x x x E1509) e4yU0D, UN x uN x UN x UN x uN x epawely m m m m m m rs s x x x x a S a 5 S S is m = c m at < m a © m = c m = =| m = ee ee Ss eet elle kel ee ell 8 Dp a a av a a #2 8/F &£€ 8/F FE 2/F EF RIF F RIF F FI] soy a o a a 2 a L ° ° ° ° ° ° SS eA, =. oe a es (g sea, quauyjoiug | (7 sea, JuaUyjosug | (¢ sea, qUaUjOsUg | (Z 4eaA JUaLUJJOIU | (T 4eaA JUaWI|OIUg (auyaseg “TZ0Z) 9Ad ‘0Z0Z) SAd “6L0Z) tAd ‘10Z) Edd ‘LU0Z) ZAd ‘9T0Z) TAd yyeay [euoiowy - yyeay Aveloijauag |JeJaAQ :dJ}a JUBA JO} Buloday :TOZ UWq!YyXy weiZolq YoJeasay uolenjenq pue s21WoU0s] YIeaH yoseasay Adijod Y}22H 40} 4822) VION aural eS Ta) sole (poisad siug 40} y10da1 0} yBnoua Aluea ulaq you pip sa (poiiad siup 40} ajgejiene you sem eyep yng SoWII payoday-jojid 10) spouyay| SasAjeuy pue ejeq :g Xipuaddy| yoday uoljenjeng jeuly aseD UOSdad BJOUM (spotsad asay 40} |Je ye 2439W s1Yy} UO Yoda JOU pip 37 B43) Buoday ION “YN Wy9e WeJBOId Jo yuaW|joJUa yng aw siyy Uo BuNJodad sem 3] ay2) Adil jaw S14} UO Buodad sem 37 ayy) JUaW|JOIUJ :3 :suosead UOISN|Ix3 y UN x uN x UN x uN x uN x eanquan YN x UN x UN x UN x UN x ewouos qa x uN x YN x uN x uN x oUuR|os YN x YN x YN x uN x YN x eyseys a x UN x UN x UN x UN x IDdMIS uN x UN x YN x YN x UN x zndp eques YUN x UN x UN x YN x UN x eue[D ejUeS UN x UN x UN x YN x UN x oajzeW ues YN x YN x UN x YN x UN x uinbeor ues mn mn om mn m m g g g 2 g g 2 2 a 2 5 2 c m = £ m = a m —_ c m = © m = © m = See ne En ceee eec ee ule ele ne Dp a Dp a a a 8 a a 3 a a | 2 & & 3 a a eS a a |S 3g & yolld 3 a 8 a B 2 E °o ° °° ° ° So 3. ao ae oe = a (g sea,Q quauyjosug | (p4ea, quawyjosug | (¢ 4eaA JuaUjOUUA | (Z4eaA JUaUJOIUA | (T 4ea,A JUaW}OIUR (auyaseg “1@0Z) 9Ad “070Z) SAd “610Z) tAd “810Z) Ad ‘LU0@) ZAd ‘9L02) TAd ae teeet WesZ01g YDJeasay UO!}eEN|eAW pue sd!WwoUdy YyeaH Yyoseasay Aotjod YiJeaH JOJ JajUaD YIN RY ne olaye UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Variant Metric: Comprehensive Diabetes Care Pilots reported the percent of enrollees age 18 to 75 who had either Type 1 or Type 2 diabetes, who had controlled Hemoglobin A1c (HbA1c), with a value of less than 8.0%. Both types of diabetes were combined into this single metric. This metric closely followed the HEDIS measure for Comprehensive Diabetes Care, CDC-H8. According to DHCS specifications, WPC Pilots were expected to use both claim/encounter and pharmacy data to identify enrollees with diabetes for this metric, although an enrollee only had to be identified as having diabetes through one of the two methods to be included. Pilots calculated the percent of enrollees with controlled HbA1c by dividing a numerator (number with controlled HbA1c) by a denominator (number with diabetes). The denominator consisted of a subset of all individuals enrolled in WPC at any time during the measurement year who were age 18 to 75 as of December 31 of the measurement year, and had a diagnosis of Type 1 or Type 2 diabetes during the measurement year or the year prior to the measurement year. Enrollees were excluded from the denominator if they used hospice services or a hospice benefit during the measurement year. The numerator consisted of the number of members in the denominator whose most recent HbA1c test during the measurement year showed a level less than 8.0%. If no HbA1c test was conducted during the measurement year, then the enrollee was assumed to have uncontrolled HbA1c. The baseline period consisted of calendar year 2016 (January 1, 2016 through December 31, 2016). Because no one was enrolled in WPC during the baseline period, Pilots defined the baseline population as the cohort that was enrolled in WPC from January 1, 2017 through June 30, 2018, per DHCS specifications. Pilots then gathered Medi-Cal data retrospectively for the baseline year for this enrollee population. This metric was reported annually. ELUM) Appendix B: Data and Analyses Methods for Pilot-Reported Metrics | Whole Person Care Final Evaluation Report SoH [Sia Paoday-Jo|!d 10) spoujay sasAjeuy pue ejeq :g xipuaddy| wWoday uoenjeng jeuly aseD UOSJdag BIOUM UN x UN x UN x YN x UN x UN x o8aiq ues x x x x x x oulpseuleg ues uN x uN x UN x uN x uN x uN xX oquaweines x x x x Vv x x APISIOALY UN x UN x UN x UN x uN x UN x 492e|d x x x Vv x x v x aBuelQ UN x UN x UN x UN x YN x UN x eden x x * x x x AasaquoW, x 3 x x x x oulopuay UN x UN x UN x YN x UN x UN x uel YN x YN x UN x YN x UN x YN x sajasuy so] x x x ® x x sBuly x x x x x x way x x x x x x eS UN UN UN UN YN UN eu]U0} UN x UN x UN x YN x UN x UN x epowely m m o m m m x x x x x x aa a = = = a = mm = c im = = im = < m = € im = © im = ee eee elles alee aie 2 ele 2 3 2 2 2 z 2 2 © © © © © o 8 & a 8 & a 8 & a e 2 a is 2 2 o a & io}!d a a a a a a L 9 9 3 o 9 9 2. 2. 2 2. a >. (g aea, quauyjosug | (pea, quawyjosug | (¢ sea, quaujosug | (z4ea, yuaWjosug | (T 4ea,A JUaW|OIUR (auyaseg ‘TZ0Z) 9Ad ‘0Z0Z) SAd “610Z) vAd “8T0Z) €Ad ‘LU02) ZAd ‘9T0Z) TAd adie) Sajaqeig aAisuayasGWo ‘dja JUEILA JO} Suloday :70Z Uqlyxy RES ERE WeiZo1g yoseasay UOHenjenq pue sa!woUD3 Yea Yyoseasay Aotjod YiJeaH JO JajUaD YIN oday uojenjeaz Jeulj a4eD UOsJdg BJOYM | SIA) Payoday-jojiq 40) spoyrayy sasAjeuy pue ejeq :q xIpuaddy [Re 7a {spowiad asayp 40} |Je 3 21439 $14) UO JOda, JOU pip 37 B43) BulOday ION “YN (powiad siy} 40} yoda 0} y8noua Ajjea ul3aq Jou pip saijiAij9e WesZOJd Jo JUaWI|jOJUA yNq ‘911}9W S14} UO BuljOdaJ Sem 3) AY) JUaW}}OIUJ :3 (pouiad siup 40) ajgejtene you sem eyep ing :suosead UOISN|Dx3 ; x x x x x x eanquan UN x UN x UN x UN x UN x UN x ewouos qa x UN x UN x UN x UN x UN x oue|os x x x £ x x eseys a x UN x UN x UN x UN x UN x D9DdMIS x x x x x x zndp eques uN x uN x uN x UN x UN x uN xX eue[D eqUes x x x x x x o0a}e\W ues x x x x x x uinbeor ues oosioues4 UN x UN x UN x UN x UN x UN x ues mn mn m on mn mn g B g £ g 2 a 2 2 2 S c m = £ m = 2 m = c m = e m = © m = id gg 4 81/8 & BIS & B/S & B/S & BIS 4 B Se oss sn cis Sls Sto u|ecue Sues e| sess Da a Dp nz a a 3 a 3&8 | 8 £228 a8 8 | 8 a8 a] 8 a ae8/8 8 8 Aol 3 a 3 a a B id So ° So ° ° So as a a = 7 = (g sea,Q quauyjosug | (p4ea, quawyjosug | (¢ 4eaA JUaUjOUUA | (Z4eaA JUaUJOIUA | (T 4eaA JUaW|;OIUR (auyjaseg “L@0Z) 9Ad “070Z) SAd “610Z) tAd “8102) Ad ‘LU02) ZAd ‘910Z) TAd weJZ0ldg Yoeasay UO!}eNjeAy pue sd1WOU0DI YyedH ara terete yoseasay Adljog yijeaH JOJ Ja]UaD YION UCLA Center for Health Policy Research ele Health Economics and Evaluation Research Program Variant Metric: Depression Remission at 12 Months Pilots reported the percent of enrollees age 18 or older with major depression or dysthymia who reached remission measured at 12 months, plus or minus 30 days, after an index visit. One single metric was reported. This metric closely followed the Minnesota Community Measurement metric for depression care. Pilots calculated the percent of enrollees with depression remission at 12 months by dividing a numerator (number who reached remission) by a denominator (number age 18 or older with a diagnosis of depression). The denominator consisted of a subset of all individuals enrolled in WPC at any time during the measurement year who were of the appropriate age, and who had an index visit that met all of the following criteria: face-to-face visit or contact with a relevant provider, PHQ-9 result greater than 9, an active diagnosis of major depression or dysthymia, and no prior index visit during the measurement year. Enrollees were excluded from the denominator if they had an active diagnosis of bipolar disorder or personality disorder, if they were a permanent nursing home resident during the measurement year, if they used hospice services or a hospice benefit during the measurement year, or if they died prior to the end of the measurement year. The numerator consisted of the number of members in the denominator who had a PHQ-9 result of less than five, 12 months (plus or minus 30 days) after an index visit, assessed from December 2 prior to the measurement year through January 30 of the year after the measurement year. The baseline period consisted of calendar year 2016 (January 1, 2016 through December 31, 2016). Because no one was enrolled in WPC during the baseline period, Pilots defined the baseline population as the cohort that was enrolled in WPC from January 1, 2017 through June 30, 2018, per DHCS specifications. Pilots then gathered Medi-Cal data retrospectively for the baseline year for this enrollee population. This metric was reported annually. Whole Person Care Final Evaluation Report | Appendix B: Data and Analyses Methods for Pilot-Reported Metrics yoday uojenjeag Jeulj a4eD UOsJdg BJOYM | SIA) Payoday-jojig 10) sSpoyray SasAjeuy pue ejeq :q xIpuaddy PR Zs o9sIsUes4 UN x UN x UN x UN x YN x uN xX ues x x x YN x UN x UN x o3aiq ues x x x x x x oulpseusag ues UN x uN x UN x uN x UN x uN xX oquaweses x x x x x x ApIsIaAlY x x x x Vv x Vv x da2e|q x x UN x UN x UN x UN x asuelo UN x uN x UN x UN x uN x uN xX eden x x x x x Vv x Aasaquo;: UN x uN x UN x UN x uN x uN xX ouppopua, x x ¥ x Vv 2 Vv 2 uuey UN x uN x UN x uN x uN x uN xX sajaBuy soy UN x UN x UN x uN x UN x uN xX s8ury x x x x x x usay x x x x x x EIS09) e4yU0D, x x x Vv x x x epowely m m m m m m rs s x x x x a a 2 a a a is m = c m at < m a © m = c m = © m = ee ee Ss eet elle kel ee ell 8 Dp a a av a a e & @/8 & 2/8 & 8/8 & BF F B/F BF Bl ay a a 2 a a a L ° ° ° ° ° ° SS a a oe ao an (g sea, quauyjoiug | (sea, JuaUyjosug | (¢ seaA qUaUjOsUy | (Z 4eaA JUaLUJJOIU | (T 4eaA JUaWI|OIUg (auyaseg “L@0Z) 9Ad ‘0Z0Z) SAd “6L0Z) tAd “g10Z) €Ad “LU0@) ZAd ‘9T0Z) TAd SUIUOW ZT 3€ UOISSIWaY UOIssasdaq :I}ay JUeWeA JOj Bulioday :€Oz Uqiyxy WweiZolq YoJeasay uolenjenq pue s21WoU0s] YIeaH yoseasay Adijod Y}22H JO} 4922 VION aural eS Ta) 6vE (poisad siug 40) y10dai 0} YBnoua AlJea ulaq you pip sa SoHII payoday-jojid 10) spouyay] sasjeuy pue eyeq :g Xipuaddy| yoday uoljenjeng jeuly aseD UOSdad BJOUM (spotsad asay 40} |Je ye I439W s1Yy) UO Yoda JOU pip 37 B43) BuOday ION :YN W998 WeJ8oJd Jo yuaW|}o1Ua Inq (pouiad siyy 404 ajqejiene you sem eyep ing aw S1y} UO Suodad sem 37 ayy) JUaW||OIUJ :3 j@W Siy3 UO BuNJodad sem 37 aya) Ar|Iq :suosead UOISN|Ix3 y x x x x x einjus, YN x UN x UN x UN x UN x UN x ewouos a x uN x uN x uN x uN x YN X ouejos x x x x x x eiseys a x x x UN x UN x UN x IDdMIS * x x x v x x znd ejues x x x x x x ese|> eques UN x UN x YN x YN x UN S: YN x oajzeW ues uN x uN x uN x uN x UN x uN xX uinbeor ues m mn on m m m z z g g g z 2 a 2 2 2 2 c m = £ m = a m —_ c m = © m = © m = See ne ee nce eee ule eles Dp a Dp a a a 8 a 3&8 | 8 a 218 & & | 8 a 3a] 8 a 8/8 8 8 AO} 3 a 8 a 8 2 Hid °o ° °° ° ° So a a oe a ma a (g sea,Q quauyjosug | (p sea, quawyjosug | (¢ 4eaA JuaUjOUUA | (Z4ea,A JUaUJOIUA | (T 4eaA JUaW}OIUR (auyaseg “1@0Z) 9Ad “070Z) SAd “610Z) tAd “810Z) Ad ‘LU0@) ZAd ‘9L0Z) TAd WweJZ01g youeasay UOljeNjeAq pue s21WOUIJ Yea GATE ERPS) Yoseasay Aotjod YiJeaH JOJ JajUaD VION RY ne olaye UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Variant Metric: Major Depressive Disorder - Suicide Risk Assessment Pilots reported the percent of enrollees age 18 or older with a diagnosis of major depressive disorder (MDD) who had a suicide risk assessment completed during the visit in which a new diagnosis or recurrent episode was identified. One single metric was reported. This metric closely followed the suicide risk assessment measure endorsed by the American Medical Association (AMA)-convened Physician Consortium for Performance Improvement, also adopted by the Federal Electronic Clinical Quality Improvement (eCQI) Resource Center. Pilots calculated the percent of enrollees who received a suicide risk assessment by dividing a numerator (number that received an assessment) by a denominator (number with major depression). The denominator consisted of a subset of all individuals enrolled in WPC at any time during the measurement year who were of appropriate age and had a diagnosis of major depressive disorder (MDD). The numerator consisted of the number of members in the denominator who had a suicide risk assessment completed during the visit in which a new diagnosis or recurrent episode was identified. The baseline period consisted of calendar year 2016 (January 1, 2016 through December 31, 2016). Because no one was enrolled in WPC during the baseline period, Pilots defined the baseline population as the cohort that was enrolled in WPC from January 1, 2017 through June 30, 2018, per DHCS specifications. Pilots then gathered Medi-Cal data retrospectively for the baseline year for this enrollee population. This metric was reported annually. i109) Appendix B: Data and Analyses Methods for Pilot-Reported Metrics | Whole Person Care Final Evaluation Report SOLO basa) PalOday-JoO|!q 10) spoujay| sasAjeuy pue ejeq :g Xipuaddy| woday uolenjeng jeuly aseD UOSag BIOUM x x x UN x UN x YN x o8aiq ues x x x x x x oulpseusag ues UN x uN x yN x uN x uN x YN xX oquauieioes x x x g x x APISJOALY x x x x Vv x v x 4a2e|d x x YN x UN x UN x UN x asueio uN x UN x YN x uN x uN x YN Xx eden x x * x x v x Aasaquo; uN x uN x UN x uN x uN x yN xX ournopuay, x x x x v x Vv x uueyw UN x UN x YN x UN x UN ¥ YN x sajesuy so} UN x UN x UN x UN x UN x UN x s8uly x x x x x x usay x x x x x x E109 e4}U0D, x x x Vv x x x epowely on nm on m mn om * 4 x 7 x x 2 & 2 2 2. S < m = e m = c m = < m = & m = © m a Es & & = & & = S 5 z 5 & e & slp 8 & a zp av az a a @ @ &@/$ & @/$8 & 8/8 &£ 8/F F BF F Bl ou a a 3 a a 8 E ° ° °° ° ° ° o = =. os SH oA (g sea, quauyjoiug | (7 sea, juaUUjjosug | (¢ seaA qUaUjOsUy | (Z 4ea,A JUaLUJJOIUY | (T 4eaA JUaUI|OIUg (auyaseg “TZ07Z) 9Ad “0Z0Z) SAd “6L0Z) tAd “810Z) €Ad “LU0@) ZAd “9L0Z) TAd juauussassy ysIYy Aploins - Japsosig anissasdaq sole) ‘Ja JUeeA JOJ Buljoday :~Oz Uq!yxy WweJZ01g youeasay UOleNjeAq pue s2!WOUODJ Yea GAME ERPS) Yoseasay Aotjod YiJeaH JOJ JaqUaD YIN oday uoenjeag Jeulj a4eD UOsJdd BJOYM | SIJa|\| Payoday-jojlq 10) spoyyay sasAjeuy pue ejeq :q xIpuaddy [Acts (sporsad asayy Joy jJe Je I14}9WU S1y UO Woda JOU pip 37 843) BuOday ION :YN (powad siyj 404 Yoda 0} YSnou—a Ajsea ulZaq Jou pip saiziAq2e WeJZOId JO JUaLU||OJUA JNq ‘DJA S14} UO BuOdaJ sem 37 ayy) JUaLW||OIUA :3 (pouiad siup 40} ajgejtene you sem eyep ing :suosead UOISN|Dx3 ; x x x x x eanquan UN x UN x UN x UN x UN x UN x ewouos qa x UN x UN x UN x UN x UN x oue|os x x x £ x x eyseys a x x x UN x UN x UN x D9DdMIS x x x x Vv x x znip eueS x x x x x x ese) ejues YN x UN x UN x UN x UN x UN x oajzey ues uN x UN x UN x UN x UN x UN x uinbeor ues oosioues4 UN x UN x UN x UN x UN x UN x ues mn on m on mn mn g B g £ 2 a 2 2 2 S c m = £ m = 2 m = c m = e m = © m = &. & gg 4 81/8 & BIS & B/S & B/S & BIS 4 B Se oss sn cis Sls Sto u|ecue Sues e| sess Da a Dp nz a a 3 & a 3 a a | 2 & & 3 a a 3 a a/8 8 8&8 Aol 2 a 2 a a B id So ° So ° ° So as a a = 7 = (g sea,Q quauyjosug | (p4ea, quawyjosug | (¢ 4eaA JUaUjOUUA | (Z4eaA JUaUJOIUA | (T 4eaA JUaW|;OIUR (auyjaseg “L@0Z) 9Ad “070Z) SAd “60Z) tAd “8102) Ad ‘LU02) ZAd ‘9102) TAd weJZ0ldg Yoeasay UO!}eNjeAJ pue sd1WOU0dg YedH ara terete yoseasay Adljog yrjeaH JO} Ja]UaD YION UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Variant Metric: Permanent Housing Pilots reported the percent of enrollees who were initially homeless, and then were permanently housed for longer than six consecutive months. One single metric was reported. This metric was created by DHCS. Pilots calculated the percent of enrollees who were permanently housed for longer than six months by dividing a numerator (homeless enrollees who reached a seven-month time point in housing) by a denominator (homeless enrollees who reached a six-month time point in housing). The denominator consisted of a subset of all individuals enrolled in WPC at any time during the measurement year who were initially homeless, and who reached a six-month time point in permanent housing between December 1 of the prior year and November 30 of the measurement year. Enrollees were excluded from the denominator if they used hospice services or a hospice benefit during the measurement year. The numerator consisted of the number of members in the denominator who reached the seven-month time point in permanent housing between January 1 and December 31 of the measurement year. Unlike other WPC metrics, the baseline reporting period for this metric was calendar year 2017 rather than 2016. This is because data on this metric could not be gathered before WPC enrollment began. This metric was reported annually. Whole Person Care Final Evaluation Report | Appendix B: Data and Analyses Methods for Pilot-Reported Metrics yoday uojenjeag Jeulj a4eD UOssdd BJOYM | SIJa\| Payoday-jojig 10) sSpoyray sasAjeuy pue ejeq :q xIpuaddy [Ass o9sIsUes4 x x x x x ues x x x 4 se ogaiq ues x % x ¥ x oulpseusag UN UN UN uN UN ‘lee x x x x 3 x oquawiees x x x x 3 x ApIsIaAlY UN x YUN x UN x YN x YN x da2e|q x x x uN x YN x aguelo x X x x a x eden x x x x x AasauoW x x x UN x YN x oulsopuayy UN x UN x UN x uN x UN x ue x x x x x sajasuy so] x x x uN x YN x s8uly YUN x YN x UN x YN x UN x ay x x x x x EAS09 UN UN UN UN uN enugy x x x x a x epowely m m m m m m rs s x x x x a a 2 a a a is m = c m at < m a © m = c m = =| m = > £ &/2 & &/2 & &/2 £ &/2 &£ El/2 & & Dp a a av a a oO o cy oO o oO $ & a 8 a a 8 & a 3 2 a 8 2 a gS & a qolld a a 2 a 2 a L ° ° ° ° ° ° 3 a a oe ao es (g sea, quauyjoiug | (7 sea, quaUUyjosug | (¢ sea, qUaUjOsUg | (Z4eaA JUaLUJJOIU | (T 4eaA JUaWI|OIUg (auyaseg “TZ0Z) 9Ad ‘0Z0Z) SAd “6L0Z) tAd ‘10Z) Edd “LU0@) ZAd ‘9T0Z) TAd SUISNOH JUBULLUJad ‘IJa) JUeeA JOJ Buljioday :soz Wqlyxy WweJZold YoJeasay uolenjenq pue s21WoU0s] YIeaH yoseasay Adijod y}22H JO} 4822) VION aural eS Ta) schol (poised siuy 40} q0dau 03 y8noua AjJea ulBaq jou pip sal SoH payoday-jojid 10) spouyay] sasAjeuy pue eyeq :g Xipuaddy| yoday uoljenjeng jeuly aseD UOSIdag BJOUM (spoyiad asayy 404 |Je e D1439W S143 UO T4odad JOU pip 37 aya) BuJoday ION :YN oe WesBosd Jo quawi||osua yng (poiiad siup 40} ajgejiene you sem eyep yng ja S14} UO BUljJodad sem 37 ayy) JUaW||OIUJ :3 j@W Siy3 UO BuNJodad sem 37 aya) Ar|Iq :suosead UOISN|Ix3 y UN x uN x UN x UN x uN x eanquan YN x UN x UN x UN x UN x ewouos a x v x Vv x v x x ouejos x x x x 3 x eyseys a x UN x UN x UN x UN x IDdMIS uN x UN x YN x YN x UN x zndp eques YUN x YN x UN x YN x UN x eue[D ejUeS UN x UN x UN x YN x UN x oaj}e ues YN x UN x UN x YN x UN x uinbeor ues mn mn om mn m m g g g 2 g g 2 2 a 2 5 2 c m = £ m = a m —_ c m = © m = © m = See ne ee nce eee ule eles Dp a Dp a a a 8 a a 3 a a | 2 & a 3 & 8 3 a a/8 8 8&8 ill 3 a 8 a B 2 E °o ° °° ° ° So 3. ao ae oe = a (g sea,Q quauyjosug | (p sea, quawjosug | (¢ 4eaA JUaUjOIUA | (Z4eaA JUaUJOIUA | (T 4eaA JUaW}OIUR (auyaseg “1@0Z) 9Ad “00Z) SAd “610Z) tAd “810Z) €Ad ‘LU0@) ZAd ‘9L0Z) TAd ae teeet WesZ01g YDJeasay UO!}eEN|eAW pue sd!WwoUdy YyIeaH Yoseasay Aotjod YiJeaH JOJ JajUaD VION RY ne olaye UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Variant Metric: Housing Services Pilots reported the percent of enrollees who were homeless, and who received housing services after being referred to housing services. One single metric was reported. This metric was created by DHCS. Pilots calculated the percent of enrollees who received housing services after being referred by dividing a numerator (number who received services) by a denominator (number referred to services). The denominator consisted of a subset of all individuals enrolled in WPC at any time during the measurement year who were referred for housing services between January 1 and December 31 of the measurement year; these services were limited to those received after the enrollee’s first WPC enrollment date within the measurement year. Enrollees were excluded from the denominator if they used hospice services or a hospice benefit during the measurement year. The numerator consisted of the number of members in the denominator who received housing services after being referred. Unlike other WPC metrics, the baseline reporting period for this metric was calendar year 2017 rather than 2016. This is because data on this metric could not be gathered before WPC enrollment began. This metric was reported annually. Eis) Appendix B: Data and Analyses Methods for Pilot-Reported Metrics | Whole Person Care Final Evaluation Report soMneW, Vass Paoday-Jo|!d 10) spoujyay| sasAjeuy pue ejeq :g Xipuaddy| woday uolenjeng jeuly aseD UOSIag BJOUM UN x uN x UN x UN x YN a o8aiqg ues x x ¢ , x oulpseuiag uN uN yN uN uN ‘lee x x x x x ojusWwelseS x x x x x apIssaAly x x x x x da2e|q x x x UN x YN x asuelo uN x uN x UN x uN x uN x eden x x x x x AasaquoW, uN x uN x UN x uN x uN x oulsopua, x x x x x uueW UN x UN x YN x YN x UN it sajesuy so} x x x YN x YN % sBuly x x x x x way x x x x x 1809 uN uN YN uN uN ens) uN x uN x YN x UN x uN x epawely m mn m mn m m 3 rd x P+ xs P+ a a a a a a c rr _ © mn = c in = £ m = ce m _ © m = & & 5 & & 5 s 6&6 a|/s 6 aB]|s 8 B/S & BlS SB BIS & B > & £le & &/> & Ele & E|2 & Ble & & Dp ap a a yn Dp oO o oO oO o oO 8 & a 8 & a 8 & a e 2 a is 2 2 o a & io}!d a a 2 a a a L ° ° ° ° ° ° > = =. os SH oA (g sea, quauyjoiug | (7 sea, juaUjjosug | (¢ seaA qUaUNjOsUy | (Zz 4ea,A JUaLUJJOIUg | (T 4eaA JUaWI|OIUg (auyaseg “TZ0Z) 9Ad ‘0Z02) SAd “610Z) vad “10Z) Edd “LU0@) ZAd ‘9T0Z) TAd SOIIAIIS BUISNOH :DJ}aJ JUeeA JO} BuljOday :90Z Wq!Y4XA aaa eect) Weigold youeasay UO!}enjeAJ pue sd1wWwoUody Yyye3ay yoseasay Aotjod YiJeaH JO JajUaD YIN yoday uojenjeaq Jeulj a4eD UOsJdg BJOYM | SIJa|\) Payoday-jojig 40) spoyyay sasAjeuy pue ejeq :q xIpuaddy Piss DdM 40 3No paddoug :a (sporsad asay? Joy jJe Je 14}9W s1Yy UO Woda JOU pip 37 B42) BuOday ION :YN (powad siyj 404 Yoda 0} YSnou—a Ajsea ulZaq Jou pip saizAq2e WeJZOJd JO JUaLU||OJUA JNq ‘DJA S14} UO BuOdas sem 37 ayy) JUaLU||OIUY :3 (pouiad siup 40} ajgejtene you sem eyep ing aw sly uo Sunsodas sem 37 aya) Ariiiq :suosead UOISN|Dx3 ; x x x x x eanquan x x x ¥ =f x ewouos a x x x uN x uN x ouejos UN x YN x UN x YN x UN x eyseys a x x x x x D9DdMIS x x x x x znd9 eques UN x YN x UN x UN x UN x e4e|D e]UeS x x x x x oajzey ues x x % x x uinbeor ues oosioues4 x x x x x ues mn mn m on mn mn g g g £ g 2 a 2 2 2 S r= m = Sc m = 2 m = a m = c m = © m = = 3 o, 3 S 3 Wes s a a &. | 3 4 28] 98 & B!o6 & a|so 5 B!s 6 Blo 6&6 8 = 5 5 = & 5 a & & z & & > & Suse Da a Dp nz a a oO oO oO cy o oO $ & a 8 2 a S| & a g & a ci 2 a gs g a qld 2 a 2 a a B I! So ° So ° ° So as a a = 7 = (g sea,Q quauyjosug | (p4ea, quawjosug | (¢ 4eaA JUaUjOUUA | (Z4eaA JUaUJOIUA | (T 4eaA JUaW;OIUR (auyjaseg “L@0Z) 9Ad “070Z) SAd “610Z) tAd “8102) Ad ‘LU02) ZAd ‘9102) TAd weJ3ol, JJeasay UO!JEN|eA: Ue SI1LWUOUOD: eo} id Y YY UOHeNjenz pi 4 4yeeH Facer ec ye] yoseasay Adljog yijeaH JOJ Ja]UaD YION UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Variant Metric: Supportive Housing Pilots reported the percent of enrollees who were homeless, and who received supportive housing after being referred to supportive housing. One single metric was reported. This metric was created by DHCS. Pilots calculated the percent of enrollees who received supportive housing after being referred by dividing a numerator (homeless enrollees who received supportive housing) by a denominator (homeless enrollees referred to supportive housing). The denominator consisted of a subset of all individuals enrolled in WPC at any time during the measurement year who were referred for supportive housing between December 1 of the prior year and November 30 of the measurement year; these services were limited to those received after the enrollee’s first WPC enrollment date within the measurement year. Enrollees were excluded from the denominator if they used hospice services or a hospice benefit during the measurement year. The numerator consisted of the number of members in the denominator who received supportive housing after being referred. Unlike other WPC metrics, the baseline reporting period for this metric was calendar year 2017 rather than 2016. This is because data on this metric could not be gathered before WPC enrollment began. This metric was reported annually. Whole Person Care Final Evaluation Report | Appendix B: Data and Analyses Methods for Pilot-Reported Metrics yoday uonenjeng [euly 42> Uossad ajoym | SII) PaOday-jojlq 10} spoyjay) SasAjeuy pue ejeg :q xipuaddy [Meets o9sIsUes4 x x x x x ues UN x UN x UN x UN x uN x o8aiq ues x 4 x ¥ x oulpseusag UN UN UN uN UN ‘lee YN x YN x UN x YN x UN x Ojuawes2eS x x x x x ApIsIaAlY UN x YUN x UN x YN x YN x da2e|q UN x x x UN x UN x asuelo UN x uN x UN x UN x uN x eden UN x UN x UN x UN x UN % Aasaquo;: UN x YN x UN x YN x UN x oulsopuayy UN x UN x UN x uN x uN x ue YN x YN x UN x YN x YN x sajasuy so] x x UN x uN x YN x s8uly x x x x x usay x x x x x EAS09 UN uN UN UN uN enugy x x x x x epowely m m m m m m rs s x x x x a a 2 a a a is m = c m at < m a © m = c m = =| m = ee ee Ss eee elle kel ee ello 8 Dp a a av a a # & @/¢ &@ 2@\/% &@ 2/8 @ Gi|e EF ale E 2] aon a a 2 a a a L ° o ° ° ° ° SS eA, =. oe a es (g sea, quauyjosug | (7 sea, JuaUyjosug | (¢ sea qUaUjOsUA | (Z 4eaA JUaLUJJOIU | (T 4eaA JUaW]|OIUg (auyaseg “L@0Z) 9Ad ‘0Z0Z) SAd “6L0Z) tAd “g10Z) €Ad ‘LU02) ZAd ‘9T0Z) TAd SUISNOH aAioddns :dja|Aj JUeWeA JO} BuljOday :/0Z7 Wq!YyXA weJZolq YoJeasay Uolenjenq pue s21WoU0s] YIeaq yoseasay Adijod y}22H JO} 4822 VION aural eS Ta) acl (poisad siug 40) y10dai 0} YBnoua AlJea ulaq you pip sa (poiiad siup 40} ajgejiene you sem eyep yng SoH payoday-jojid 10) spouyay] sasAjeuy pue eyeq :g Xipuaddy| yoday uoljenjeng jeuly aseD UOSIad BJOUM, (spoisad asay Jo} jj ye 2439W S14} UO Yoda JOU pip 37 B43) BuOday ION “YN Wy9e WeJBOId Jo yuaW|joJUa yng jaw S14} UO Buodad sem 37 ayy) JUaW|JOIUJ :3 jaw Siy3 UO BuNJodad sem 37 ay3) Ari|Iq :suosead UOISN|Ix3 y UN x uN x UN x uN x uN x eanquan YN x UN x UN x UN x UN x ewouos a x * x x x ouejos UN x UN x UN x UN x UN x eISeYyS a x UN x UN x UN x UN x IDdMIS UN x UN x YN x UN x UN x znd ejues x x x x x ese|> eques YN x YN x UN x UN x oayeW ues UN x uN x uN x UN x uinbeor ues mn mn om mn m m g g g 2 g g 2 a 2 2 2 2 c m = £ m = a m —_ c m = © m = © m = See ne En ceee eec ee ule ele ne Dp a Dp a a a 8 a a 3 a a | 2 & & 3 a 2 3 a a/8 8 8&8 Al 3 a 8 a 8 2 Hid °o ° °° ° ° So a a oe a ma a (g sea,Q quauyjosug | (p4ea, quawyjosug | (¢ 4eaA JUaUjOUUA | (Z4ea,A JUaUJOIUA | (T 4eaA JUaW;OIUR (auyaseg “1@0Z) 9Ad “070Z) SAd “610Z) tAd “810Z) Ad ‘LU0@) ZAd ‘9L0Z) TAd ae teeet WesZ01g YDJeasay UO!JeEN|eAW pue sd!WoUdy YyIeaH Yoseasay Aotjod YiJeaH JOJ JajuaD YIN RY ne olaye UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Universal Metric: Comprehensive Care Plan Pilots reported the percent of enrollees who received a comprehensive care plan, accessible by their entire care team, within 30 days of enrollment and within 30 days of the enrollee’s anniversary of enrollment in WPC. Two sub-metrics were reported: (1) the percent of enrollees who received a comprehensive care plan, accessible by the entire care team, within 30 days of enrollment, and (2) the percent of enrollees who received a comprehensive care plan, accessible by the entire care team, within 30 days of the enrollee’s twelve-month anniversary date of enrollment in WPC. This metric was created by DHCS. For each of the two sub-metrics, Pilots calculated the percent of enrollees with a comprehensive care plan by dividing a numerator (number with a plan within 30 days of enrollment or anniversary) by a denominator (number of enrollees that were new or had an anniversary). The denominator consisted of the number of enrollees who were either new to WPC, or who had a twelve-month anniversary as an enrollee in WPC, depending on the sub- metric. The numerator consisted of the number of members in the denominator population who had a comprehensive care plan within 30 days of enrollment, or their twelve-month anniversary of enrollment, depending on the sub-metric. Unlike other WPC metrics, the baseline reporting period for this metric was calendar year 2017 rather than 2016. This is because data on this metric could not be gathered before WPC enrollment began. This metric was reported annually. ai Appendix B: Data and Analyses Methods for Pilot-Reported Metrics | Whole Person Care Final Evaluation Report SoH Fla) paloday-Joj!d 10) spoujay sasAjeuy pue ejeq :g xipuaddy| woday uolenjeng jeuly aseD UOSIdad BIOUM x ae x x 4 x o8aiq ues oul x x x x x [paeuseg ues x x x x x ojuswelseS a x x x x apIssaAly x x x x x dBI2\d x x * Vv x Vv x asueio x 2 x x 3 x eden x x * x x AasaquoW, x x x x x oulsopual, x oo x x x uueyw x * x x x sajasuy so] x v x Vv x x x sBuly x x x x x usay x x x x x E1509. euquoD x x x x x epowely m m m m m m 3 rd x P+ xs P+ S s S Ss Ss a c rr _ © mn = c in = £ m = ce mn _ © m = & & 5 & & 5 s 6&6 a|/s 6 aB]|s 8 B/S & BlS SB BIS & B Se Eee Ble eels = fle 2 ele £ os D a a a a a oO o oO oO o oO 8 & a 8 & a 8 & a e 2 a is 2 2 o a & iO}!d a o a a 2 a L 3 3 3 5 3 3 > = =. os SH oA (g aea, quauyjosug | (pea, quawjjoiug | (¢ see, quaujosug | (Zz 4ea, quaWjosug | (T sea, qUaW|jo1U (auyaseg “TZ0Z) 9Ad ‘0Z02) SAd “610Z) vad “10Z) Edd ‘LU02) ZAd ‘9T0Z) TAd quawjoiug jo sAeq O€ UIYUM - UB|d Je) aAIsuayasdWOD :dIJJa\ [ESJAAIUN) OJ Buloday :g0z Uq!uXy WweJZ01g youeasay UOIeNjeAq pue s2!WOU0DJ Yea Yyoseasay Aotjod YiJeaH 4OJ JajUaD YIN adele) yoday uojenjeaq Jeulj a4eD UOsJdg BJOYM | S19) Payoday-jojlg 40) spoyray sasAjeuy pue ejeq :q xIpuaddy [Reka (sporsad asay? Joy jJe ye 14}9WU s1y UO Woda OU pip 37 B43) BuOday ION “YN (powad siyj 404 Yodas 0} YSnou—a Ajsea ulZaq Jou pip saijiAq2e WeJZOJd JO JUaLU||OJUA JNq ‘DJA S14} UO BuOdaJ sem 37 ayy) JUaU||OIUJ :3 :suosead UOISN|Dx3 ; x x x x x eanquan x x x x a x PWOUOS a x x x x x oue|os x x x £ x eyseys a x x x x a x D9DdMIS x x x x x zndp eques x x x x x ese) ejues x x x x x o0a}e\W ues x x x x x uinbeor ues oosioues4 x x x x x ues mn mn m on mn mn g g g £ g 2 a 2 2 2 S c m = £ m = 2 m = c m = e m = © m = id & & & & 5 5 6&6 B/S &5 BIlS &6 Blo &6 BI]S & BIlS & B = — & | 5 & 68 | 5 a |= e ¢@/| 3 Ss 5 8° 5 Da a Dp nz a a oO oO oO cy o oO $ & a 8 2 a S| & a g & a ci 2 a gs g a qolld 3 a 3 a a B I! So ° So ° ° So as a a = 7 = (g sea,q quauyjosug | (p4ea, quawyjosug | (¢ 4eaA JUaUjOUUA | (Z4eaA JUaUJOIUA | (T 4eaA JUaW;OIUR (auyjaseg “L@0Z) 9Ad ‘0Z0Z) SAd “610Z) tAd “8102) Ad ‘LU02) ZAd ‘9102) TAd weJZ0ldg Yoeasay UO!}eNjeAJ pue sd1WOU0Ig YIedH ara terete yoseasay Adljog yijeaH JO} Ja}UaD YION SoH [Scloa) paloday-Joj!d 10) spoujay| sasAjeuy pue ejeq :g xipuaddy| woday uolenjeng jeuly aseD UOSIag BIOUM x ae x q x o8aiqg ues oul x x x x Ipseulag ues x x x x oyuaweses x x x x APISIOALY x x x x 4222 |d v x x x v x asueio x 2 x x eden x x * x AasaquoW, x 3 x x oulopuay x x x x wey x x x x sajasuy so] x x Vv ¥ ® sBuly Vv x x x v x way x x x x EOD e4jUu0D x x x x epawely m m o m m m x x x x x x S s S Ss Ss a = mm = c im = = im = < m = € im = © im = &. & &. &. & &. ¢ & B/|s 8&8 8/5 & Bile & BIS KF Bl & B ee ee) eee es alee aie 2 ele 2 3 2 2 2 z 2 2 © © © © © o 8 & a 8 & a 8 & a e 2 a is 2 2 o a & io}!d a a a a a a L 9 9 3 o 9 9 2. 2. 2 2. a >. (g 4ea, quaujjosug | (p4ea, quawyjosug | (¢ sea, quaujosug | (zea, yuaWjosug | (T 4ea,A JUaW|OIUR (auyaseg ‘0Z0Z) SAd “610Z) vAd “8T0Z) €Ad ‘LU02) ZAd ‘9T0Z) TAd “TZ07Z) 9Ad quaw|joiuz jo Aiesuaniuuy YUOI-anjam| JO SAB OE UlYUM - UL} Bed aAIsUaYyaIGWOD ‘IU}aI\J |eSIBAIUN) JO} BUIOday :60Z Uq!YXy WweJZ01g youeasay UO!eNjeAq pue s2!WOUDJ Yea Yyoseasay Aotjod YiJeaH JO JajUaD YIN adele) yoday uojenjeaq Jeulj a4eD UOsJdg BJOYM | SJ) Payoday-jojig 10) spoyrayy sasAjeuy pue ejeq :q xIpuaddy [Rkes (spousad asay} Jo} jJe ye I1439WU s1y UO Woda OU pip 37 842) BuOday ION :YN (poisad siy} 40} yoda 0} y8noua Ajjea ulZaq Jou pip saijiAij9e WesZOJd Jo JUaWI|jOJUA yNq ‘D11}9W S14} UO BulyJOdaJ Sem 3) B43) JUaW|}OIUJ :3 (pouiad siup 40) ajgejtene you sem eyep ing aw sly uo Sunsodas sem 37 aya) Arliiq :suosead UOISN|Dx3 x x x x eanquan x x x a x PWOUOS a x x x x oue|os x x x £ eyseys a x x x x D9DdMIS x x x 3 x znip eueS x x x x esejp Blues x x x x o0a}e\W ues x x x x uinbeor ues z x % i oosioues4 v v v d ues mn mn m on mn mn g g g £ g 2 a 2 2 2 S c m = £ m = 2 m = c m = e m = © m = id & & & & 5 5 6&6 B/S &5 BIlS &6 Blo &6 BI]S & BIlS & B = — & | 5 & 68 | 5 & s | 5 e ¢@/| 3 Ss 5 8° 5 Da a Dp nz a a oO oO oO cy o oO $ & a 8 2 a S| & a g & a ci 2 a gs g a qolld 3 a 3 a a B I! So ° So ° ° So as a a = 7 = (g sea,Q quauyjosug | (p4ea, quawyjosug | (¢ 4eaA JUaUjOUUA | (Z4eaA JUaUJOIUA | (T 4eaA JUaW|;OIUR (auyjaseg “L@0Z) 9Ad “070Z) SAd “610Z) tAd “8102) Ad ‘LU02) ZAd ‘9102) TAd weJZ0ldg Yoeasay UO!}eNjeAy pue sd1WOU0DI YIJedHq ara terete yoseasay Adljog yrjeaH JJ Ja}UaD YION UCLA Center for Health Policy Research errs Health Economics and Evaluation Research Program Appendix C: Data and Analyses Methods for Narrative Reports Overview of Data and Analysis Methods for Narrative Reports Data Source The UCLA evaluation team used data from ten rounds of narrative reports (PY 2 — PY 6 mid-year and annual) submitted by WPC Pilots to the California Department of Health Care Services. Data in these reports covered January 2017 through December 2021. In these reports, WPC Pilots were asked to report on program achievement, success, and progress as well as on program challenges, barriers, and lessons learned in three major domains: care coordination, data and information sharing, and data reporting. WPC Pilots were also asked to report on outcomes and sustainability of WPC. A complete overview of reporting requirements for these narrative reports can be found in Attachment GG Special Terms and Conditions. Methods All narrative reports were reviewed for completeness and imported into the qualitative analysis software NVIVO. To facilitate analysis, all reports were organized by WPC Pilot. Both inductive and deductive coding methods were applied for analysis. After developing an initial codebook based on sections outlined in the narrative reports (deductive coding), the codebook was subsequently refined to reflect emergent themes in the data (inductive coding) and to eliminate redundancies and repetitions across sections of the report. All narrative reports were coded and reviewed by at least two members of the team, and five primary themes from the initial coding process were identified: (1) care coordination; (2) data and information sharing; (3) identifying, engaging, and enrolling eligible beneficiaries; (4) biggest barriers to WPC success; and (5) WPC outcomes and sustainability. An additional round of coding was conducted to identify and quantify specific subthemes within the data. Only the most prevalent subthemes were included in the final evaluation report. Limitations The qualitative analysis of narrative reports relied on self-reported data from participating WPC Pilots. While efforts were made to validate responses and perspectives within and across the data sources when possible, there is potential for responses to have been subject to response or social desirability bias. Due to the concurrence of WPC with other programs focused on Whole Person Care Final Evaluation Report | Appendix C: Data and Analyses Methods for Narrative JREiyA Reports rls} UCLA Center for Health Policy Research ' Health Economics and Evaluation Research Program redesign of care processes and payment, the effects of WPC cannot fully be separated from other programs. Appendix C: Data and Analyses Methods for Narrative Reports | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program SISSIES AE: Appendix D: Data and Analysis Methods for Lead Entity Surveys Data and Analysis Methods for Lead Entity Surveys Data Sources PY 3 LE Survey To gain insight into WPC implementation in the early stages of the program, UCLA administered a PY 3 survey from July-September 2018 to key program staff from Lead Entities (n=27) participating in WPC Pilots. The survey included 74 closed and open-ended questions on various domains: * Questions about the local context of the Pilot and motivation for participation; ¢ Questions about WPC infrastructure, resources and implementation; * Questions about intra- and inter-agency communication, decision-making and collaborative processes and participation in learning collaboratives; * Questions about processes developed regarding potential and current WPC enrollees; and * Questions about program monitoring activities, performance trends and perceived impact of WPC. The PY 3 survey assessed health information technology infrastructure, specific activities related to project implementation, ratings of level of effort, staffing and workforce development, participation in quality improvement activities, and challenges and solutions. COVID-19 Impact Survey To gain insight into WPC Pilots’ response to the pandemic, UCLA administered a COVID-19 impact survey in April 2020 to WPC LEs (n=25). Napa and Plumas (of the Small County WPC Collaborative) did not complete a survey; Plumas was no longer participating in the WPC Pilot at the time. Whole Person Care Final Evaluation Report | Appendix D: Data and Analysis Methods for Lead Entity Surveys eer ret) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program The brief, rapid response survey assessed (1) how WPC infrastructure and integrated care delivery approach may have helped with local response to COVID-19, and (2) the impact of the COVID-19 pandemic on WPC enrollment, staffing, and services. PY 5 LE Survey To gain insight into WPC implementation in the later stages of the program, UCLA administered a PY 5 survey from July-September 2018 to key program staff from Lead Entities (n=25) participating in WPC Pilots. Napa and Plumas (of the Small County WPC Collaborative) did not complete a survey; Plumas was no longer participating in the WPC Pilot at the time. The survey included 55 closed and open-ended questions on various domains: Additional detail on data sharing infrastructure and resources; * Care coordination processes and supports; e Specific housing related services; ¢ Integration of health and social services; e Perceived impact of WPC; and « Sustainability and the transition to CalAIM. PY 6 LE Survey In PY 6, UCLA fielded an additional survey to LE leadership in all WPC Pilots during the waiver extension year (n=26). LEs that did not participate in PY 6 were asked to complete with perspective through PY 5 (Solano, as well as Mariposa and San Benito of the Small County WPC Collaborative). Surveys provided additional information on WPC implementation, changes to WPC since the PY 5 survey, and updates on sustainability planning and progress on transition to CalAIM. All Surveys For all four surveys, questions constituted a variety of structures including yes/no, multiple choice, ranking, Likert scale, and matrix. Surveys were pilot-tested among stakeholders at a selection of Pilots. Following pilot testing, UCLA revised the structure and content of the survey to address stakeholder feedback before deploying the final version of the survey to all Lead Entities. Surveys were administered via SurveyMonkey. WPC Pilot contacts at each Lead Entity were emailed a link to complete the survey and were instructed to involve additional team members who were most knowledgeable about implementation of specific WPC domains. Surveys were Appendix D: Data and Analysis Methods for Lead Entity Surveys | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research errs Health Economics and Evaluation Research Program filled out predominantly by leaders (directors, administrators, and program managers) in each Lead Entity. The survey instruments are available in Appendices O and P. Methods Data were analyzed using Excel and Stata. Descriptive analyses were conducted to assess Lead Entity characteristics on the different survey domains. Members of the UCLA team recoded responses to open-ended questions or responses to Likert Scale and matrix questions as needed to appropriate categories. Throughout the final evaluation report, UCLA presents the most recent survey results — where appropriate, UCLA presents multiple data points over time. Limitations The analysis of the surveys relied on self-reported data from participating WPC Pilots. While efforts were made to validate responses and perspectives within and across the data sources when possible, there is potential for responses to have been subject to response or social desirability bias. Due to the concurrence of WPC with other programs focused on redesign of care processes and payment, the effects of WPC cannot fully be separated from other programs. Whole Person Care Final Evaluation Report | Appendix D: Data and Analysis Methods for Lead Entity Surveys eer ret) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Appendix E: Data and Analyses Methods for Follow-up Interviews with Lead Entity and Frontline Staff Overview of Data and Analysis Methods for Follow-up Interviews Data Source To gain in-depth understanding of WPC implementation, UCLA conducted semi-structured interviews with key informants from all participating WPC Pilots (n=26). Interviews were conducted from June to September 2021 and lasted roughly 90 to 120 minutes. UCLA conducted interim interviews (n=27) from September 2018 to March 2019. WPC Pilot contacts were asked to include individuals with expertise on the county’s WPC implementation and care coordination processes. Each WPC Pilot participated in at least two interviews: one with frontline staff (i.e., care coordinators, Public Health Nurses, frontline supervisors, social workers), and one with key leadership and management (i-e., WPC Directors, project managers). Interviews were conducted with WPC Pilots via Zoom video conferencing and recorded with software or handheld audio recorders. Interviews were led by a member of the UCLA evaluation team, with input from additional members, as appropriate. A total of 58 interviews were conducted with 167 individual key informants. Interviews focused on greater understanding of concepts such as care coordination workflows, data sharing infrastructure, communication and decision-making processes, impact of COVID- 19, and inter-agency collaboration with partner organizations. Additional topics included: the general impact of WPC, synergy with other projects, leadership and staff buy-in, recommendations for ongoing implementation of the program, and plans for sustainability of key WPC components and transition to CalAIM. See Appendix X for the interview protocol used for both frontline staff and Lead Entity interviews. Methods Interviews were transcribed verbatim using Rev.com transcription services and de-identified prior to analysis. A codebook was developed based on key evaluation questions and interview content, using both inductive (i.e., based on emergent themes from coding of initial interviews) and deductive coding (i.e., based on a priori themes and components of the interview protocol). After establishing a codebook, the transcribed interviews were distributed among five members of the study team for coding analysis. During the coding process, study team members met regularly to discuss emerging themes and refine the codebook as needed. See E¥P Appendix E: Data and Analyses Methods for Follow-up Interviews with Lead Entity and Frontline Staff | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program SISSIES ARR: Exhibit 210 for the qualitative codebook used for the qualitative analysis. Analyses was completed using NVivo software. Limitations Follow-up interviews relied on self-reported data from participating WPC frontline staff and key leadership and management. While efforts were made to validate responses and perspectives within and across the data sources when possible, there is potential for responses to have been subject to response or social desirability bias. Due to the concurrence of WPC with other programs focused on redesign of care processes and payment, the effects of WPC cannot fully be separated from other programs. Whole Person Care Final Evaluation Report |Appendix E: Data and Analyses Methods for Follow-up [IEA Interviews with Lead Entity and Frontline Staff erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Descriptive analyses were conducted to assess partner organization characteristics on the survey domains. Limitations PY 5 partner surveys relied on self-reported data from participating partner organizations from WPC Pilots. While efforts were made to validate responses and perspectives within and across the data sources when possible, there is potential for responses to have been subject to response or social desirability bias. Due to the concurrence of WPC with other programs focused on redesign of care processes and payment, the effects of WPC cannot fully be separated from other programs. Appendix F: Data and Analyses Methods for Partner Surveys | Whole Person Care Final Evaluation Report eer ret) UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 210: Codebook Used for Preliminary Coding of Follow-up Interviews, PY 6 NODES Respondent Role Who are respondents, how involved in WPC County and Organizational Context Description of other programs that may overlap with WPC (Health Homes, PRIME, etc.), LE motivation for participating in WPC, rural/urban, etc. WPC Program Summary of Pilot and core elements of the Pilot; includes changes over time, & how pilot funded Pandemic impact Impact of pandemic on Pilot, Pilot response, and any specific services provided to COVID-19 impacted individuals. May double-code with other domains. Pilot Leadership and Governance Governance structure (e.g., admin committees), frequency of meetings, how decisions made re: Pilot program design, operations, etc. Partners Any references to established relationships with other organizations or to departments/divisions within same umbrella organization (e.g., partnership changes, quality of communication, factors affecting engagement, etc. This does not include one-time interactions with frontline staff at other organizations/departments) Data sharing/ IT Infrastructure Any references to data sharing, HIE or other data repository, case management software or other infrastructure for tracking referrals, services, & care coordination or to facilitate reporting/outcome tracking Enrollee outreach and engagement Any references to strategies used to outreach to or identify individuals eligible for WPC, engage them in care, or when to disenroll / graduate from care Care Coordination Definition of care coordination, how care coordination works (e.g., needs assessment, care plan, referral tracking), who is on the care coordination team, Accountability, how WPC staff communicate with one another or with other providers in the community Other Services References to other services provided as part of WPC, including housing support, recuperative care, BH care, sobering center stays, etc. EY/M) Appendix E: Data and Analyses Methods for Follow-up Interviews with Lead Entity and Frontline Staff | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program SIRS ISS ae: Staffing Any references to recruitment or retention, turnover, caseload, type of staff used, supervisor & staff orientation, supervisor/staff skills & training, staff concordance with target populations, references to burnout, compassion fatigue, etc. Community engagement Any references to inclusion of client/enrollee or staff perspectives in WPC planning, implementation, or al Contracting and Contract Incentives Any references to contracting with the state or with WPC partners, factors affecting time intensity or specialized knowledge for contracting, effectiveness of contract incentives, and perceived utility for CalAIM. [Also include include references to RFP/RFA, MOU, data sharing agreements that were signed, etc.] Diversity, equity, or inclusion Any references to Pilot efforts to address disparities, or consider DEI in program planning, implementation, or evaluation activities. Lessons Learned, Facilitators, or Barriers Lessons learned, Facilitators, or Barriers (anticipate double-coding with other content) WPC Outcomes Perceived Impact, including benefits and unanticipated consequences, including client successes. WPC Sustainability and transition to CalAIM Factors affecting sustainability of WPC, plans during transition to CalAIM, perceptions of CalAIM, etc. Pilot-Internal Evaluation & Ql Activities Internal evaluation activities & QI Technical Assistance and Desired Support for State Perceptions of provided TA or of QI activities, what they wish the state had done Illustrative and Interesting quotes Social Determinants of Health (new) Explicit references to social determinants of health, social needs, social factors Other Any important content that doesn’t fit elsewhere Whole Person Care Final Evaluation Report | Appendix E: Data and Analyses Methods for Follow-up [REY/ Interviews with Lead Entity and Frontline Staff rls} UCLA Center for Health Policy Research ' Health Economics and Evaluation Research Program Collections / Sets: *® = County/LE e Legacy, Expansion, New e Program Size (Target Pop): Small (<=1,000), medium, Large (10,000+) * Program Structure: Centralized vs. De-centralized « Program Structure: Some contracted vs. All Contracted vs. Not Contracted e Cost: Large, medium, small e Target population: High Utilizers, SMI/SUD, Chronic Physical Conditions, Homelessness and/or At Risk of Homelessness, Justice Involved ® Interview Type: Leadership and Strategy, Frontline Supervisor; Frontline Staff E¥{cq) Appendix E: Data and Analyses Methods for Follow-up Interviews with Lead Entity and Frontline Staff | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Appendix F: Data and Analyses Methods for Partner Surveys Overview of Data and Analysis Methods for Partner Surveys Data Source To gain a comprehensive understanding into WPC implementation, UCLA developed a survey for participating partners from WPC Pilots. The interim partner survey was conducted from July to October 2018, and included a total of 227 partners from 25 Lead Entities. A total of 227 partners from 25 Lead Entities participated in the survey. Partner surveys from two counties were excluded: Plumas withdrew from participation, another delayed implementation due to fires (Sonoma). The final partner survey was conducted from June to August 2020, with various types of partner agencies, including community clinics, hospitals, private human and social service providers, county mental health and housing agencies, probation/law enforcement agencies, private mental health and substance abuse agencies as well as other types of county and private agencies. A total of 166 partners from 25 Lead Entities participated in the survey. Partner surveys from two counties were excluded: Plumas withdrew from participation, and Napa did not participate. The majority of questions in the final partner survey were identical to questions from the PY 5 LE survey; the PY 5 partner survey was more limited in scope than the PY 3 partner survey. Questions explored specific activities related to project implementation, ratings of level of effort, staffing and workforce development, changes in collaboration as a result of WPC, and challenges and solutions to project implementation. Questions constituted a variety of structures including yes/no, multiple choice, ranking, Likert scale, and matrix. Final partner surveys were conducted via Qualtics. WPC Pilots provided an email link to their partner agencies to complete the survey. Partners were advised to involve additional team members as needed to ensure questions were answered by the person most knowledgeable about specific WPC domains. Surveys were mainly completed by leaders (directors, administrators, and program managers) of the partner agencies. Methods Data were analyzed using Excel and Stata 12. Whole Person Care Final Evaluation Report | Appendix F: Data and Analyses Methods for Partner Surveys UCLA Center for Health Policy Research Health Economics and Evaluation Research Program SIASAISSI AP: Appendix G: Data and Analyses Methods for PDSA Reports Overview of Data and Analysis Methods for PDSA Reports Data Source WPC Pilots were required to submit Plan Do Study Act (PDSA) reports for Universal and Variant metrics semi-annually and annually in order to report on quality and performance improvements. WPC Pilots were also required to submit a PDSA Pilot summary worksheet. Pilots organized PDSAs into category types that included: (1) ambulatory care, (2) care coordination, (3) comprehensive care plan, (4) data, (5) inpatient utilization, and (6) other. DHCS provided Pilots with a template for PDSA reporting. WPC Pilots were asked to report the following for each PDSA project: (1) WPC Lead Entity, (2) project lead (name/phone number/email), (3) reporting period, (4) PDSA project, (5) target population, (6) PDSA size, (7) status, (8) PDSA type, (9) start date, (10) recent revision date, (11) report date, (12) project description, (13) revision, (14) results, and (15) next steps. Methods PDSAs reports were sent to UCLA by DHCS and reviewed for completeness. UCLA received PDSAs for the following reporting years: PY 2 mid-year through PY 6 annual. PDSA reports were compiled into Excel and categorized by both Pilot and reporting year. Counts were developed for PDSA type and length of days per PDSA project by PDSA type, Pilot, and reporting year. Counts of PDSA reports were also calculated based on continuity through all reporting periods. Whole Person Care Final Evaluation Report | Appendix G: Data and Analyses Methods for PDSA Reports erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Appendix H: WPC Services Offered through PMPM Bundles and FFS Methodology In order to categorize the services reported by WPC pilots into eleven common service groups, UCLA used (1) WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6; (2) PYS (2020) LE survey (n=25); and (3) WPC Annual Invoices from PY 2 to PY 6. Pilots had the flexibility to provide services that would best fit the needs of their target populations and could be delivered with existing or newly developed infrastructure and resources. While no single service was specifically required by the program, all Pilots were expected to provide care coordination and housing support services as needed to address the needs of beneficiaries. Additionally, services delivered by Pilots could only be identified through an examination of bundled (PMPM or per-member per-month) or specific services (FFS or fee- for-service) that Pilots used to report to DHCS and receive payment. Bundled services varied in what combinations of services were included and associated costs, as they were tailored by each Pilot to fit the needs of the population they expected to serve. As part of the LE survey in 2020, UCLA asked Pilots to identify which of 20 services were offered through each PMPM and FFS category. For this analysis, two Pilots in the Small Counties WPC Pilot (San Benito and Mariposa) were analyzed separately as each used different bundles of services and had different rates. Napa and Plumas counties were excluded from this service analysis because Napa did not respond to the LE Survey and Plumas dropped out of WPC in PY 3. Categories that were added in 2021 after the 2020 LE survey were excluded from this analysis when information on which services were provided through these categories was not available. These were primarily COVID-19-related services. From the 20 specific services included in the survey, UCLA aggregated the findings into 11 categories of services: (1) Outreach; (2) Care Coordination; (3) Housing Support; (4) Benefit Assistance; (5) Employment Assistance; (6) Sobering Centers; (7) Medical Respite; (8) Transportation; (9) Health Education; (10) Legal Services; and (11) Re-Entry Services. In Exhibit 211 services offered through each PMPM and FFS category are shown along with the rate of each category for each program year that were pulled from the WPC Annual Invoices. The rate was used to calculate the total service cost per enrollee e}:{0) Appendix H: WPC Services Offered through PMPM Bundles and FFS | Whole Person Care Final Evaluation Report SJJj pue bi:tam) S8|PUNG WidINid Y8NOJY} pasayO SBdIAJaS IdM :H XIpuaddy| yoday uonenyjenq jeuly aie} UOsJadg BjOUM OsTs OsTs OsTs v/N v/N x] xX YyaeasynOC 389435 SsajaWOH 82 epawely ‘{sua}unoduq JUawasesuy pue Aso8a}eD S44 yoeasjno :uolsuedx] T'WZ# [eG TET$ TETS TETs TETs TETs ¥ g]esdajas Wodsued} quale oz epawely > SOWOH |E91PAIN HA “207# 8G AsoBaye5 S44 vSTS STs STs vSTs vSTs x sesajod 07 epowely ~$10}EUIP400D 212} 9SINN Aso8aje5 S44 - SOWOH [EI1Pe| HE “GOZ#H 12d zors zors zors zOTs zors x SHO 6 epawely 3e squaljed JOJ uoljeulpsooD Aso8aye5 S44 a1eD Hal Para|dwo “6T# 12d SSTS SSTS SSTS SSTS SSTS x x auljdjay er epawely — Juawyjeal| Japsosiq asn AsoBajeD S44 aoUeISqNS 0} S|e10d OTH 120 SSTS SsT$ SsT$ SST$ SST$ x ageyury] TL epawely - Juawi}eal| Japsosiq as) AloSaye) S44 aoueysqns 0} sjeWOd 9TH ‘12d 6727S 6@7$ 6277S 627S 6727S x sanoy ‘39e]U09 OT epowely Jaseueyy aeD /M 8uljsa} Aso8ajeD S44 Bnig - UOIsIaAIG GNS “ST# 1G 6¢¢s 6¢7$ 627s 6z2$ 6227S x sanoy ‘siaqunooua ysIA | 6 AlodajeD S44 epawely NOD - UOISIAIG GNS “STH 12 6¢@zs 67zzs 6727S 67z$ 6zz$ x sinoy juaussassy | g AloBajeD S44 epowely 7 UOISIBAIG GNS “ST# IFC 6€¢S 6EzS 6EzS 6EzS 6€z$ x x skep | £ Aio8aye5 S44 epawely pag - JajUap BuLiaqos “pT# [9d SSZ'TS | SSZ‘TS | SSZ‘TS | SSZ‘TS | SSZ‘TS x x aouejsisse jeda| | p AioBa}e S44 epowely jenpiaipul--aoueysissy [2897 3g uo}eoNpy BulsnoH “g# [2G arey aqey yey a7ey arey [2 & ae z ge Sf lf ° [° awen Alo3a3e) ‘Ai0893e5 30lld tzoz | ozoz | etoz | stoz | cztoz |f |® |S |2 & 3 a seg 3 | lm lo © |5 ls |= la |S 18 g |e | fe iF le la |B I le | 3 |8 |g e 8 [8 |@ ja & iF > = |5 |F |e B jo |% 8 3 ° |e 3 [+ |g TZ0Z 1 LTOZ ‘sarey jenuuy payelossy pue ‘SadiAsas payelzossy ‘salio3a}e) WidiNd Pue S44 :TTZ UwqIYXg WeJZ01g Youeasay Uolenjenq puke so1woUu0s3 YyeaH ae ES Yyoseasay Adljod YiJeaH 4OJ JaqUaD YIN yoday uonenjeag Jeulj 2423 uossad ajoym | S44 Pue Sa|PUNG WdINd Y8NOIY} paseyO SE21INAS IdM :H XIpusddy [acts bles bles bles pls vrs Xx 2 4811 ajpung Z AsoBa3e) epawely sadias Juawageueyy a1eD WdWd Tes Tees Tees Tees Tees X T 481, a|pung T Asose3e3 epawely sadialas Juawaseuey| ase) WdNd ors | O0r$ v/N v/N v/N Xx (sjexou) ey epawely uoljejUaWind0g pue SujuaalS AsoSazep S44 quawssassy YyeaH ‘O/# 12d o0zs oozs V/N v/N v/N x (sjay0y) sjuauissassy Ty epawely Adquq payeulpsoo9 *69# [2d AsoSajeD S44 oors o0rs v/N W/N v/N x {asegewoH) BE epowely uonejuawinsog pue suluaeJ9¢ AsoSaj}e5 S44 qUaUssassy YI/28H “Pgo# led 00z$ o0zs v/N v/N v/N xX (asegawioy) sjuaussassy LE epawely Adquq pazeulps005 “9g9# [2G AsoSaye) S44 06z$ 06z$ 06z$ v/N v/N x SUO}}290] a/qissa20y SE epawely ‘saaiasas A2e20apy pue Asosaye5 S44 juawi|}o1u3 syyaUag “G6r# 120 oszs oszs oszs v/N v/N x x WiesBO1d ardsay “Br# [PC ee epawely AsoBaje9 S44 SST$ SST$ SST$ v/N VIN x SANOY — $}9e]UOD UONESIAEN ze epawely 18) aul|djaH “p9T#Ied | AloBeze9 S44 oos‘vs | o0s‘rs | o0s‘rs | v/N v/N qual|9 Jad sasuadxa Te epawely 3/Q/3||a-punj aouejsissy Aso8ajeD S44 Bulsnoy wa 1-10Us “20TH 12d oss oss oss v/N v/N x| xX ypeey-ul 0g epawely {ssaqunodug Juawasesug pue AsoBajeD S44 yoeadjng :uolsuedx] ¢'y/# [ed oots oots ooTs v/N v/N xX| XxX awoH/AyI9e4 67 epawely ‘sdajunoduq Juawasesuy pue Aso8ajeD S44 yoeasng :uolsuedx3 7'V/# 180 31ey 37ey ayey yey yey 2 2 z z s 3 3 & s 2 2 awen Aioga3e) ‘AioZa3e> 3Olld T1207 0z0z 610Z 8t0z dtoz |S | |= |B = 2 ils 1% |& ie o & io |F ls |f |= |2 |= [> |9 |B <= |§ |e J§ |S im 3 I> is |S a le |= lo |> (S |2 [2 Ie [8 gis |6 |& 18 jm |3 2 js |& 2/2 |8 |F S |5 |= [4 |s a? lf (5 |e |e 8 jo |& 2 g @ |S 3 ia |e a 5 a 3 8 wiesZolg yoseasay Uo!eNjeng pue so!woU0dg Yea Hae yoseasay Adljog YaJeaH 404 JajUua YIN Sdd pue filam) S8|PUNG WidINid Y8NOJY} pasayO SBdIAaS IdM :H XIpuaddy| yoday uonenjenq jeuly aie} UOSJadg BjOUM “vis LUIS Lots ZvT$ V/N x] X Texiajay | p Mogae S44 way pue uauissassy Suluaal9s €€TS eels €€T$ 6EzS v/N x ‘Ades0npy suyauag | ¢ Alo8aje) S44 usay vEL'ZS v/N v/N v/N v/N x] xT x xl xix] xix (9 4a!) Aeys way Bu07 € AlosayeD e4S0D WdiNd esjuo> VTS V/N V/N v/N v/N xX| xX] Xx xX} xX] xX!) xX] xX g Jal] Juawaseuew 7 Miosaje), ByS0D ase9 anisuayasdwio WdiWd e4U0D g9zes v/N V/N v/N v/N X| X| xX X|X |X| xX] XxX V Joly Juawaseuew T Mogaie) e4soD ase9 anlsuayaiduioD WdWd e4jU0D v/N oos‘rs | 00s‘ | o0s‘r$ v/N xX S44 | T Ato8a1e5 $44 83505 SadIAlas UO!}Isues] BUISnOoH e4]U0D SZs$ sZss v/N v/N v/N x x} xX] Xx (asegawoy) ajpung sasivlas OT Aiosa3e epawely payes8e3u] pue SuisnoH YyyeeH WdiNd SZs$ SLS$ SLS$ SLS$ v/N x x| x] x € a1 6 Aiogaie> epauely alpung sao|Asas pazesZa}u) WdiWd pue SuisnoyH ‘yyeeH o0rs o0rs o0rs o0rs v/N x x| x] x Z29IL g AiogayeD epawely a|pung saaiaias pazes89qu) WdilNd pue Suisnoy ‘yyeaH ooEs ooEes o0Ees o0Es V/N x xX| xX) xX T Jal, ZL Mosaje) epawely ajpung saziaias payes3aqu} WdWWd pue Suisnoy ‘yye8H ese'Ts | ese‘TS | |Ese'TS | ESEe'TS v/N x] x weal a17e1ydAsg g Aiodaje) epawely J@a41¢ Ja]UaD YYeaH IsMIL WdWd v/N v/N v/N TL@s$ TLZs$ ajpung ad1Ales g Arodayep epawely Bululeysnsg Adueual »g BulsnoH WdWWd v/N v/N v/N v7es v7es ajpung adiAjas y MosayeD epawely uolsued) Suisnoy pasueyug WdWd v/N v/N v/N STEs Stes suolsued, € Asosaje> epawely Ayes BulssnN pal|yys Wdlld a7ey a7ey arey aqey arey |7 ie = z = So ey @ |Z |e je wen Alozaje) ‘A0833e5 3Olld tzoz | ozoz | etoz | stoz | ctoz |p @ |S 2 |& |e S |e |b |e |Z 5 |¢ |= le |& lo ig IF le 8 8 w |e |= |e 2 2 |P |v |§ |= & je |& |8 | 12 |S 12 I§ |& = |8 /& jo |B 3 |* |2 Is ls 8 s |> |é \8 3 (8 = a 5 a 3 8 AARC WERE wesZo1g yoseasay uoNenjenq pue so!woU0d3 YI]eaH Yyoseasay Adljod YiJeaH 4OJ JaqUaD YIN Jeuly 242 Uosiad a}oYM | S44 PUE Se|PUNG did Y8NOAY? Pa1ayO SEdINIBS IdM 7H yoday uonenjeag uaddy (igts Szzs v/N v/N v/N v/N x] xX l x] X quawasesug *g yoees3no | z AiodajeD S44] sajaduy soz 6L¢$ V/N V/N 7 V/N xX x} x Tx daqua) Buuagos | T Aio8aje) S44 | sajasuy so7 v/N zst'ts | zst‘t$ | zst‘t$ | zsT'TS x| x/ x Xx x x| x oney € Alosaye) sBUIy MO7/UO!}EUIPsOOD ased ‘dwu0> WdlWd v/N LSTS LST$ LSTS LSTS x x x] x] x z Asosaje} sBuly WdWd v/N 9zss 97s$ 9z7ss 97s$ x| x] x x x! xl x uO!}eUIPJOOD a1e> T Atogaye) ssuly WdiNd v/N S7z'es | Szz‘zs | s7zz‘z$ | szz‘zs x x x x| xX Adeoonpy Iss | 7 Aio8a}e) S44 SsuD v/N 99TS 99T$ 99T$ 99T$ xlx X x| x quawasesug | € Alosaie) S44 sduiy v/N sozs sozs sozs sozs x x x uonessaqu| Ayunuo) | Z Arosa}e) S44 sBuly v/N osts osTs OsTS OsTS xX x x xX| XxX yun | T Aio8aie) S44 SBUIY ase) aatyadnoey wa) OYS TLTS TLTS TLTS TLTS v/N x yoddns Buisnoy ajesapow g Alosaye5 ulay Wdild 008‘TS | 008‘TS | O08‘TS | OO8'TS | OO8'TS | X| X| X| Xx xX x uolqeulpsoo) y Aiosaye) usa) uonesa2se2U]-3s0q AeG-06 WdiWd oss oss osts oss oss X| X| xX x x UO!JEUIPIOOD B42 JdM ¢ Aiogaie) usa) WdiNd oozs oozs o0zs oozs oozs x x sadiAlas juawAo| dwg @ Mosaie) usa) WdiAd oss o8srs o8srs O8vs osrs x) xX eN SUISNOH T Aiosaje), ulay WdiAd £L§ 2Ls LL v/N v/N quawjeall | 6 Alo8aye5 S44 wley uoljes8aqu| AylunWiWwoZ ooEes ooEes oes v/N v/N spod ase) | g Aloda}e5 S44 wey v/N S8s S8s v/N v/N x x aue9 ayidsay | 9 Aio8aqeD S44 way v/N 06s 06s 06S v/N x [euajay pue uoleuojuy | ¢ Alo8aqeD S44 usay arey arey arey arey seu le |e lo z 3 (2 |F |@ jF |f |g wien AioZa3e5 ‘Aio333e) Olid tzoz | ozoz | etoz | stoz | ctoz |f @ |S 2 |& |e S |e |b |e |Z 3/8 |m (8 (2 eS |e fe (8 [8 a le |= le |2 [S |2 IP le |§ |= gis |6 |& 18 jm |3 2 js |& 2/8 |8 |S |S j2 |= [2 |S |3 = 5 5 iF |e Bio |& a 5 a a |t io ® 5 FI wesZolg Yoseasay Uo!eNjeng pue so!woU0dg Yea Pree eauesea yoseasay Adljog YaJeaH 404 JajUua YIN Sdd pue (Slam) S8|PUNG WidINid Y8NOJY} pasayO SBdIAIaS dM :H XIpuaddy| yoday uonenjenq jeuly aie} UOSJadg BjOUM LLS$ v/N v/N v/N v/N x] x] x x} x[x] x (SN3-ans) ¥oddns vt Aiofaje) | sajaBuy so] pue ‘uolesiaey “Uawuasesuy WdWd dapsosiq asp adueysqns T6z'€$ V/N V/N V/N v/N xX xlx! xl x uol}eUIP1O0} asey padue4yUR €T AloZaje9 | sajaBuy so7 save) Sulgpiig pue jenuapisay WdiNd vetzs | v/N v/N v/N v/N x led] zt AsodareD | sajeduy $07 ae) Sulspiig pue jenuapisay WdWd 1949 SUIZplig pue jenuapisay €0T'TS v/N v/N v/N v/N x] xX (usi) TT Aiodaye9 | saja8uy soy quaidisay ad1Asas aAlsua}u] WdWd 629'ts | W/N v/N v/N wIN | xX] Xx] Xx] Xx x] x] x] x uo}eUIp4005 a1} OT AioBaie5 | sajauy S07 pasueuyug - Aujua-ay aoijsne WdWWd 60S v/N W/N v/N win |X| X X x] x] x] xTx 31eD WNPY g AioBaqe9 | sajaBuy so7 papuayxg - Aqua-ay aonsnr Wdld Tess v/N v/N v/N v/N x] x X xX} xX/| xX] xX] xX Jessajay Ayunuw0) £ Mo8aye9 | sajaBuy so7 unpy - Anua-ay aaysnr WdlNd 60S v/N W/N v/N win |X| X X x} x] x] x 1e119J94 9 AioBaqe | saja8uy so7 wer ynpy - Ayqua-ay aorsne WdiWd ovs’6s | V/N v/N v/N v/N X x| x x} x] x] x SOOIAI9S g Mogae) | sajesuy 507 ade aaljesadnoay a}e1YydAsq WdWd vST'9S W/N V/N V/N VIN x x! xX x] x) x] x SadIAas aJeD aAesadnday y Mogaje) | sajaduy so7 WdWd v7rs v/N v/N v/N v/N x x x] x) x] x (ss) € Alodajep | saja8uy so7 sa2iAiasg yoddns Adueual WdlNd o8es W/N W/N v/N v/N X X x} x] x] x SIMS z Mogae) | sajasuy so7 uoddns aiez ssaussajawoy WdWd SEBS V/N V/N V/N v/N xX x|x| x saa|Asas Ased0apy syyeuag T Alodaje) | sajasuy so7 Wdid BTSs V/N V/N v/N v/N xix!) x] x (swea, yaas3s) | € Aio8a}e} S44 | sajasuy so7 juawasesug »g y2ea1]NO a7eu a7ey a7ey arey arey |P 2 = 7 = & if (8 j= |e 2 wen Aiogaze) ‘Aioga3e> ECE] tzoz | ozoz | etoz | stoz | ctoz |p |S |S [2 |& |e |S |e |e 8 |3 5 |€ |n [8 |& 2 1 |? fe (8 8 ge lf |g 7 lp ie |B |g ie | @ jo |& 12 |S Im [3 12 IS | 2/8 |S jo jB /3 |* [2 Is ls 3 so |7 js |8 3 /§ |e w 3 wo 8 i. maneeRuRaR WeIZ01g Yseasay UO!}eN|eAW pue sdiwoUcdy Yyyeay youeasay Aaijod YyiJ2aH 40J JaUaD VION yoday uonenjeag Jeuly a2) uosiad ajoym | S44 pue Sa|pUNg WdWid Y8NosY} paiayo Sed1MEaS DdM :H xIpuaddy BEEfs osz$ osz$ oszs osz$ osz$ l l [x] x quawasesuq g ydeeI3NO | T A10B|3e9 S44 esodie 00z$ 00z$ 00z$ V/N 7 x 4032907 SuIsnoH yp Miodajze> une, WdlWd z9v$ 790s 790s 79ovs v/N x| x] x x| x] x] x saBuay|eyd |2190s-o4sASq ¢ AioBaye) une x@|dwo> pue suolipuoD Wdid YaeaH jeqUa] ajesapo; 0} PII YM sjenpialpul Jo} Juawaseueyj ased ovss ovss ovss oss ovss x] x] x x] x{| x] x quawaseue 7 ArozayeD uel ase) paseg-3ulsnoy, WdWNd oLzs oLzs oLz@s oz7$ ozs x| x| x x] x] x] x jquawazeue T Aiogae9 uve ase) aaisuayaidwo) WdiNd SvTS Svs Svs v/N v/N x ans queuneas | 2 AioSeie) S44 une NWUSpISay JO aposidy + Pay 8. GINS [eUapisey Aep +06 09S 09$ 09$ v/N v/N x yuauissassy 1VddS-IA | 9 AloSaie) Sid une c6Es z6ES @6ES v/N v/N x S|ENpIAIpul ssajawoH | ¢ AioBa}e> S45 une, jo Juawasesuq paseg-pjal4 oos‘v$ | o0s‘rs | o0s‘vs | ToZ‘zs v/N x a4 U-BAOW Jual|D | p A1oBa1eD S44 une, Sezs Ses Ses Lvs Lots x Ue[g 1D pasa}Uad-UOSiag | ¢ A1oBa}eD S44 une, LvTs Lvts Lvts Lvts Lots x x [euajay | z AlodayeD S44 une, pue ‘\uauissassy ‘Suluaasas 06S 06S 06s 06s 06s xX [euajay pue uolewojuy | T Alo8ajeD S44 ue, 99Z$ v/N v/N v/N v/N x| xX] x x] x) x] xX Pooysoqusien s,vIVIN ZT Aso8aye9 | sajasuy so7 WdWWd TLU'TS v/N v/N v/N v/N x| x x] x) x] x dead OL gt Aso8aje9 | saja3uy so7 WdWd zs v/N v/N v/N v/N x| xX] Xx x] x] x] x aye) JO suoisued, ST Aso8aje) | sajaguy so7 - xa}dwo9 Ajjes1pa Wdild 37eu 37eu a7ey a7ey a7eu a ie z z = 3 ey s = 2 2 wen Aiogaze) ‘Au0883e5 30lld Tz0z oz0z 610z stoz ctoz |f |® |& j2 |& |e [S |e je |o |e Zein le & es |F la S |g a le |= le |2 [S |2 IP le |§ |= e |e |5 |2 |B |@ |S j4% 5 jf 2/2 |8 |F Ss |5 |= |e |s Bi? | |5 |e |s ge & = 3 3 [+ |9 WweJZ01q youeasay uolenjeaq pue s2|WoU0dy YyIJeaH Pearce yoseasay Adljog YaJeaH 404 JajUua YIN Sdd pue VA:tom S8|PUNG WidINid Y8NOsY} pasayO SBdIAIaS dM :H XIpuaddy| yoday uonenjenq jeuly aie} uoSJag BjOUM 00s‘v$ | 00S‘v$ | 00S‘rS v/N V/N ajpung ul-aaow | z AsoBajeD S44 @BUeIO T8T$ T81$ T8T$ T8T$ T8T$ xX|xX| xX] xX) xX xX| xX x aue9 aatjesadnday | T Aiodaje9 S44 asuelO v/N v/N v/N 80€$ 686$ x| xix x x] xX sadiasas juswaseueyy z Alosaiep Aasa}uoW asea paseg Aylunuwo> WdiWd 902$ 902$ 902$ 686$ 80ES x x] x!) x] x SadIAJaS JUBWASeUeY T Alosaye) Aalaquo/W ase) paseg Ayjunwiwo) WdWd 80€$ 80s 80€$ B0E$ v/N x| x] x WD Sia4JoM uersIDUeL4 OT Aasaquow AsoSaj}e5 S44 vls'zs | vests | pzs‘zs | vzs‘z7$ v/N x] x] x Buisnoyay pidey | 6 AiosaxeD S44 AasaquoW szs‘zs | szsz$ | szs’zs | ses'z$ | W/N x1 x] x qoddng | g Aio3a1e9 say | Aaloquow Adueual »g UO!eSIAeN Suisnoy LT@s LT@s L1@$ L412 v/N x yeasqsuns Jajuaa BSuyagos | 9 AioBaje) S44 Aasaquow 887$ 887s 8387S 887$ L1@s x x] xX daqua Bulsagos | y Alodaje5 S34 Aasajuow LLs LLs LLs dus 887s xX| xX) xX] xX yoddns | ¢ Aio8aj}e5 S45 Aasaquo/ pue quawace|d Sulsnoy v9ss v9ss v9ss v9Ss v9ss X| X| xX XxX) xX) xX] xX) xX] xX ajpung 7 Atogaye) | oulsopuay| PUIPJOOD aj} Wa] YOYS WdWNd 9T8s 9T8s 9T8s 9T8S 9T8s X| X| xX xX) xX) xX} xX) xX] xX a|pung T Asogaje) | oulsopuay) yoneulpsoo5 Aysuaqul Y3IH WdiNd OsTs OsTs OSTS osts osts X x| x] x} x] x yioddns | z Aio8a3e9 si4 | oupopuaw jeuolsues) YyYeaH jejUs\y v/N vsts vsTs vSTs vsTs xX) X| xX X| xX) xX sadiasas aydsay |eoipaw | T Alo8aie) S44 oulsopua|\| 68E'TS | G8E'TS | G8ETS | E8E'TS | EBE'TS x} xix syoddns suisnoy z Alosaje5 esodueyy Wdid TeL'ts | t2z'ts | tzz'ts | tzz'ts | TzZ‘ts x| x] x] x x} xT x] x woreuipsoo0 T MoBaje) esodueyy ae) aAlsuayasdwo> WdiNd ooss ooss ooss ooss ooss x aue9 ayidsay | 7 Aso8ajeD S44 esoduey aqey aqey aqey ajey arey |7 ie = z = ge 5 S = ° S awen Aloga3e) Aso#a3e 20jld tzoz | ozoz | etoz | stoz | ctoz f @ |S |2 2 |e S Je |b aaa Zein le & SS |F la S |B a le |= le |2 [S |2 IP le |S |= gis |6 [2 18 jm |3 2 ls |& 2/8 |8 |S |S j2 |= [2 |S |3 a 5 5 iF |e @ io e & = 3 3 [+ |9 aaa cect weisold youeasay UOljenjeAZ puke sd!wWwOUoddy Yyyje3aqH Yyoseasay Adljod YiJeaH 4OJ JaqUaD YIN Jeuly 2423 Uossad ajoYyM | S44 Pue Sa|PUNG Wd Y8N014} pasa}HO SE21N2S DdM :H yoday uonenjeag uaddy Pits W/N v/N | 8Zt‘T$ | W/N v/N | l 0 € MioBaye5 S44 | OjUaWweseS Szzs S77 S77 s7zs S7zzs x| x] x x| x x| x Sdd [euayay pue yoeas3ng | Zz AsoSajeD S44 | oyuaweides LS2S LS7$ LS@$ v/N v/N xX| xX] x) x x skeq peg +d3I | T Asodaied S44 | ojuaweses 690S 690$ 690$ 690$ 690S xX x x] x] x] x quawaseueyy z Arosaje) Oplssanry ase) Yoddns Suisnoy WdlNd oses oses | ose$ oses oses x| x] x x) x|x{| x] x] x quawaseuew Se) NY T Aogaie) ApISIOATY WdWd 6EZS 6EZ$ 6€7$ 6EzS v/N x| x) x] x Adevonpy syyeuag | 7 Ao8aie S44 aplssanry €97$ €97$ 6€7S 6EzS 6EzS x x] x) x] xX yoeasyng/suluaeis | T ArosaieD S44 Splssonly esz'zs | ese’es | ozt'es | ett'zs | etr'zs xX] x] x x| Xx x| x quawiedesu3 vy Aiogaie3 Jaded WdWWd gests | ses‘Ts | Zsz‘tS | €09'T$ | €09'TS x | x x| x] x SEd|Alag BUISNOH € Mogaie) Je0e|d WdWd 9 9 €1Z’6s | 9788s | 9788s x| x] x X wieido1g 212) aridsay [es1paW z@ Mogae) 490d g9‘ots | 99‘0TS WdiWd zz'ts | zvz'Ts | T9E'TS | Tzs'TS | Tzs'tS x| x] x x | x X woNeU!psoo5 T Mogaie> 489d aue) xa|dwo9 anlsuayasdwio5 WdWd 096$ 096s | 096$ v/N W/N x] XxX] x x] x] x] x] x saoias Ayigeureisns 9 AioBaieD aBuesO 33 uolyeBIAeN SUISNOH Wdlid vests | vests | vests | V/N v/N X SOOIAI9S y MoBaIe> ‘3BURIO aseajay pue yseay-ul [ler WdiWd 80z$ 80¢$ 807s 80z$ 807$ X X x | x uonesiAeN ¢ Mogae) aBuesO B YyoeaIINOC DyWIIEd¢ [INS WdWWd 9Tz$ gtzs | 9tz$ 91z$ 91z$ x| x] x x|x xx SdOPIAOAd J83USD z@ Mogaiey 38ueJO uj-doig Aq papiaoid saoinsas WdWWd ageyul] pue aaiwoddns T2t$ T21s T21s T21s T21s x| xX] Xx x| x x] xX SOIIAIaS UOITESIAEN, T Alodaje) agueio SS9|WOH IUI/D 8 [eUdSOH WdWWd a1ey 37ey ayey aqey yey a ie = z = 3 ey i Ee 2 9 awen Aloga3e) ‘AioZa3e> 3Olld 1z0z 0z0z 610z 8t0z coz |# |® |= la |& [2 |S la |e |& [2 5 |€ ln [8 (© 2 1 |? fe (8 8 w |e |= |e 2 2 |P |v |§ |= & je |& |8 | 12 |S 12 I§ |& 2 |8 /8 |S lS 2 |* 18 |S [2 8 s |> |é \8 3 (8 = wa 3 wo $ i. WweJZ01q youeasay uoljenjenq pue s2|WoUody YyyeaH Pearce yoseasay Adljog YaJeaH 404 JajUua YIN S44 pue (Elam S8|PUNG WdINid Y8NOsY} pasayO SBdIAIaS dM :H XIpuaddy| ywoday uoNenjenq jeuly aie} UOSJag BjOUM VETS v/N v/N v/N v/N SadiAsas ayidsay jeaipay) | T Aio8a}e S44 ‘o9sjaues4 ues 7S6'ES v/N v/N v/N v/N x x| xX x] x| x swea Aynoy YsiH g Alosajep o8aiq ues WdWd T89$ v/N v/N v/N wN | xX] x] x] x] x x| x] x] x € aseYd UO}eIB9qU| B21MaS 7 AsoBaye) o8a1g ues WdlNd Ts8$ v/N v/N v/N v/N Xx] xX| xX] xX x| xX] x) Xx * 7 8SeYd UO!EIBAU] BI1AIaS T Asogaje) oaiq ues WdWd vos v/N v/N v/N v/N X X + Jaqunoaug | T Aio8aye5 $44 o8aiq ues quawasesu g yoeas3nO €87$ V/N v/N v/N v/N x x] x x x T Aiodaje9 | oulpieuiag WdWWd ues 8t7s V/N v/N v/N v/N x SadIAlag BulsnoH +55 | 7 AJoBa}eD S44 oulpieuiag ues LT2s v/N v/N v/N v/N x] x) x] x xX} xX] xX] xX] Auanoy yoeesan9 paseg-pjaly | T Aro8aje9 S44 | oulpseuieg ues 9E6'TS | 9€6'TS | 9E6'TS | JEE'TS | OEETS x| x] x suoddns 7 Mo8a3e9 | ojuag ues WdlWd “s9'T§ | ZS9'TS | Zg9‘TS | “ZS9‘TS | zs9‘TS | X| X| xX] X| X] Xx] xX] x x uoleU!ps005 T Alosaye9 oquag ues a4e) aAlsuayasdwo7 Wdid 99€$ 99€$ 99€$ 99€$ 99€$ x] xX quawasesug *g yoeesnC | T A1oBaje> S44 oyuag ues 787s 787s 787s 787s 787s x | Xx X SOd1MES, € Aiogaye) | oyuawieses uoleZIAeN g JUBLAZeUeW, WdlNd ase) Aqisuaqu| J2M07 Less Less Less dess zess x| xX xX SOOIAI3S Z AloZaje9 | oyUaweses uol,eSIAeN *g JUBWAaSeUeY] Wdid asea Aysuazu| JeysIH sles sles ses SLES SLES x| x x| xX] x) x a|pung SuisnoH T Alodaje) | ojuaweies WdWWd aey aew aey a1ey arey a ie z z = 3 3 iF jt 2 9 wen Aloza}e) ‘Aioga3e> 30lld Tz0z oz0z 610z stoz coz |# |® |= ja |& |e | [a |S one 3% |m [8 |8 [2 |S [2 la [8 8 w je | la lo 3 |> lu |§ |= & je |& |8 | 12 |S 12 I§ |& 2/8 |S jo jB /3 |* [2 Is ls eB] 8 |" \® 18] la |S |e 8 a are tcret WeIZ01g Yseasay UO!}eN|eAW pue sdiwoUcdy Yyyeay Yyoseasay Adljod YiJeaH 4OJ JaqUaD YIN Jeuly 2423 Uossad ajoYyM | S44 Pue Sa|PUNG WdId Y8N0I4} pasa}HO SE21MAS DdM :H yoday uonenjeag uaddy (iets 9s$ v/N v/N v/N v/N X Xx x] x] x] Xx WoleuIpsoo9 a4eD T Aodaye) | uinbeor ues WdWd LETS v/N v/N v/N v/N x x x] x) x] x wea] uoljes3aqul SH | € Aoda}e) S44] uInbeor ues v/N 9ss 9ss 9ss 9s$ xX x] x] x] x uoleulpsoo) ase) | z Asodaje9 S44 | uinbeor ues sss V/N v/N v/N v/N x] xX x! x ade) aanjesadnday | T Aogayep S44 | ulnbeor ues 090'TS v/N v/N v/N v/N x x x wea] ae) g Aio8a}e) oosiouesy SINNH Ajisuayul Y3IH SWdINd WdiNd ues tvs v/N v/N v/N 7 xX SOIIAIOS UOIEZIIGeIS y Modaze> ‘oosoues4 Adueual pue Suisnoy Wdid ues BvEes v/N V/N v/N v/N x| xX SOdIAI3S € Aiogaje) oosiauely uoljsued| Sulsnoy pasueysuy Wdid ues STES v/N v/N V/N V/N x x] xX SOIIAJBS UO!JLUIPIOOD ase}, 7 Alozaye) oosiouely Wdld ues 9s v/N v/N v/N v/N SOdIAIaS, T Aiogaye9 ‘oosioues4 quawasesuq pue yoees3no WdWd ues v/N OTs gts gts V/N x x] x saaiaias | / A1oBa1e) S44 oosiauesy quawasesuq pue yoees3nO ues ess v/N V/N 7 V/N xX x| xX sadlAlas uol}esIAeN pue | 9 Alo8ajeD S44 oosiguel4 uol}eulpsoo> payasse] pidey ues sszs V/N v/N v/N v/N x| xix sadialas | ¢ Alo3aye5 S44 ‘oosioues4 uolsuedx] Aujug payeuipsoo ues €8$ v/N v/N v/N v/N x x) XT RTS SAdIAlag Ja]UaD adinosay | AlodajeD S44 oosinuely ues v/N vers vETs vets vets x| x] x x] xX suoipuos | ¢ Alo8aje> $44 oosiouely joAsd pue jesIpaw ues Joy ayidsay jerIpayy ul sAep v/N Ovts Ovts Ovts Ovts ans | z AsoBaie9 S44 ooSINUE4 doy BSuijas 1wW43 GNs ul sAep ues a7ey a7ey arey aqey arey |7 ie = z = So ey @ |Z |e je wen Alozaje) ‘A0833e5 Old tzoz | ozoz | etoz | stoz | ctoz |p @ |S |2 |& |e S |e |b |S |Z 3S 8 |m (8 (2 eS |e fe (8 [8 w |e |= |e 2 2 |P |v § |= & je |& |8 | 12 |S 12 I§ |& 2 |8 /8 /& lS 2 |* 18 |S [2 8 s |> |é \8 3 (8 = wa 3 wo o ° 8 a WweJZ01q Youeasay Uolenjenq pue s2|WoU0sy YyJeaH praerteer yoseasay Adljog YaJeaH 404 JajUua YIN Sdd pue Nica) SO|PUNG WdINid Y8N0JY} pasayO SBdIAIaS IdM :H XIpuaddy| yoday uonenjenq jeulj aie) UOSJag BjOUM SLTS V/N v/N v/N v/N l x sjeuajay pue yoseaiyno | ¢ AioBa}e5 S44 znd eUeS 000‘e$ v/N v/N v/N v/N yoddns Asueuay | 7 Aio8a}e5 S44 znd9 eyUeS v/N 00s‘v$ | O0S‘vS | O0S‘vS | O0S‘”S yoddns Buisnoy | T AioBajeD S44 znig eyues LL0°%S v/N v/N v/N v/N x x x] xX] x SUBJ] @WOH BUISINN g Aiodaje) | ede|D ejues WdWd €88S Vv/N v/N v/N V/N xX x| xX x! x! x] x | Juawaseuew aiep Wal 3u07 MosajeD eye[9 eques WdWWd v9E'TS v/N v/N v/N v/N x| x] xX x x] xX quawaseueyy aseD Wa] PI € Asogaje) | eve|) eques WdWWd €8zts | w/N v/N v/N v/N x| x] x X X| x | qeweseuew e1e9 We] WoYs z Mogaye | ere [5 ees WdWd LETS v/N v/N v/N v/N x| xX xX| xX] x) xX] x yuoddns| T Alosajep | euejD ejues Jaad pue uoneniqeyay WdWWd 000'TS v/N v/N v/N v/N x|xX| xX) xX] xX) xX] xX] x] x] x quawissassy | ¢ Aio3aje) S44] ere] eqUeS Bulsnoy — JP4488y 8 SSeD0y v/N oOTs OOTs v/N v/N X X x | Xx ypeaiano quaned | Aiodaie) S44] e4e[9 eques 9vzS v/N v/N v/N v/N x x| x] xX 4 x] xX uole3s Bulsagos | € Asodajey S44 | e4e/D eJUeS 9LES v/N v/N v/N v/N x| X x] x] x aydsay jesipayy | z Asodajzep S44] eye|D eqUeS ooTs v/N v/N v/N v/N x] x] x! x x x quawiasesuq pue yoeaqno | T Aiogdaje5 S44 | e4e/D elUeS 6728S V/N v/N v/N v/N x| x] x x x x] xX quawysnipy Z Asogaje) | oaje|\ ues WOL YM saa|asas Aranod9y WdiWd pue yyeaH |eJolAeyag 9€9S V/N V/N v/N v/N xX x x} x!) x] Xx quewysnipy T Mio8ajep, oayeyy ues WO. YUM ssaul|aM 0} se3plig WdWd oEers v/N W/N W/N v/N X xX| xX] xX] x a|pung ase3 Aysuazu) Mo] € Arodaje) | uinbeor ues WdWWd Toss v/N v/N W/N v/N X xX| xX] xX] x ajpung ase) Ayjsuaqu Y3lH z Arogaye) | uinbeos ues WdlWd aey aew aey a1ey arey a ie z z = 3 ey s = 2 2 wen Aloga}eD ‘AoZa3eD Old Tz0z oz0z 610z stoz ctoz |f |® |& j2 |& |e [S |e je |o |e 3 |s |m fe [2 fe lS |F ie [8 [8 a le |= le |2 [S |2 IP le |S |= e |e |5 (2 |8 |@ |S j4% 5 if 2/8 |8 |& js [3 |= [2 BS [2 a\’ je 2 ele | |g 8 2 & = 3 3 [+ |9 are tcret WweJZ01g Youeasay UO!JenjeAq pue so|WoUosy YyIjeaH Yyoseasay Adljod YiJeaH 4OJ JaqUaD YIN Jeul4 2429 Uosiad ajoym | S44 PUe Se|PUNG did Y8NoAU? PalayJO SEd!ABS IdM 7H yoday uonenjeag uaddy [irdcts v/N 69TS 69T$ 69TS 69T$ xX l x l xT xX l x SOdIAIaS YIeAIINO aiqow | z Asosa1eD S44 eunquay, v/N 671s 621s 62TS 6zTS x xX) xX wesBo1g ae) aanjeiadnday | T A1oBa3eD S44 eanjuej, v/N Q9E'TS | 99E'TS | 99E'TS | 99E'TS X| X| xX Xx) X|X| xX) xX] xX ajpung T Aio8aje) ewouos juawaseue;, ase) aAlsuajul WdWd v/N O€TS O€TS v/N v/N sasiaias | z Alo8a}e5 $44 euiouoS ade) aAljesadnsay Wal WOYs v/N 6rS 6rS 6v$ 6vs x x x x] xX saaiaias | T AioBaje> S44 eBWOUOS quawasesuq pue yoeas3no v/N vses sys vsrs vsrs X| X| X X|X |X| xX] XxX ajpung WdWid T Mogaie) ouejos WdiAld 9T8$ 9T8$ 9T8$ 9T8$ 9T8s x| Xx xX} xX] xX) xX] Xx quawaseue ase SuIsnoH 7 Aosa}e), eyseys Wdild sess sess sess S6s$ s6ss x| xX x] xX SODIAJAS [EDIPI|\) T Alosaye5 B]SeYS WdiAd oos‘v$ | 00s’rs | O0s‘rs v/N v/N x yoddns Adueual | ¢ Aio8a}e5 $45 eySeYS vETS vers vers v/N v/N x 4aqUaD SIS aiqowy | z A10daze5 S44 eqseys oszs oszs oszs v/N v/N xX daquay Sulagos | T Aio8aje5 S44 eyseys TLS V/N V/N v/N V/N x xX WdWWd syoddns y Atodaje) znd eyues Bulsnoy 9}e|pawsa}u] WdWWd 8IZs V/N v/N v/N v/N xX xX WdlWd € Aio8aje9, znd eyues syoddns Sulsnoy aalsuayu| WdlWd Toss V/N v/N v/N v/N x| x xX) x!) x aipung WdiNd YeeH 21uI/D Z Asosaje) znaQ eyues Wdid zoss V/N V/N v/N v/N ajpung T AiosaieD. znd eyues WdlWid WiJeeH JeJolneyag WdWd 00s v/N v/N v/N v/N x (90u) | § AroSa3e5 S44 zndQ eUueS Jaqyuaz aie) aanesadn2ay oes v/N v/N v/N v/N x Ayqisya | y AsoBaxeD S44 znd eyues pue juauussassy ‘Suluaasas a7eu a7eu areu aed ae /F |e le [3 | (8 [5 |e le [2 |S awen Aio%a3e) ‘Aosa3e) 30lld tzoz | ozoz | etoz | stoz | ctoz | |S |S |Z |& |e |S |e |b |e |F 3 lg |e 8 es IF ie (8 [8 a le |= le |2 [S |2 IP le |S |= gis |6 [2 18 jm |3 2 ls |& 2 |8 (8 |§ 3 j2 = |2 2 a eS aig | ff 8 2 a 5 a 3 [+ |9 WweJZ01q Youeasay Uo!enjenq pue s2|WoU0dy YyIJeaH ANDER EREE yoseasay Arlo YaJeaH 404 JajUa YIN SdJj pue (fam) S8|PUNG WidINid Y8NOJY} pasayO SBdIAIIS JdM :H XIpuaddy| uoday uoNenjenq jeuly aie} UOSJag BjOUM ‘yeah UAaAIS a4} U! pasajjo JOU sem AJOBa}e9 ay} SajeI!pU! Y/N *AJOBa}e9 UdAIZ ay) Aq paplAoid SeM ad!AJaS BY] SB}LI!PU! X :SAION “9 Ad 02 Z Ad Wy sajonuy jonuuy IdM pue ‘TLZ0Z 4equia2ag-ZT0z Alenuer ‘soday uolnzi/147) pun yuaLUjjosU AjJaUIOND asOJ UOSIad aJOYM ‘ Ad WO SABAINS 30|14 JdM :294NOS v/N ve7s ve7s ve7s ve7s X] xX] Xx xX} X/| xX] xX] xX ajpung € Arosaye) Banque, UOl}eUIPsOOD aJeD paseq-pjal4 WdWd v/N ozzs ozes ozs ozs x| x x] Xx uolJeulpsoo9 a4eD 7 AtoBaeD einjUua, Wdld v/N BTEs Bres stes Stes x| x a|pung quewasesuy T Aso8aye9, eanquan Wdld v/N v/N OsTs v/N v/N x x x] x uonesiaen | p Aio8ajeD S44 eunqua, uoneaiiddy |ass/iss v/N o0rs 000‘TS v/N v/N sadiasas Asejiouy | ¢ Alo8ajeD S44 eunquan, pue ydeaino payesiel 37ey 37ey 37ey ayey yey 2 2 zg z s 3 Ey & = 2 2 awen Ajogajze), ‘AloZaj3e> 3Olld Tz0z 0z0z 610z stoz ctoz |f |® |& [2 |& |e |S |e je |o |e 5/6 lm lo (© 5 |S |= la |S 18 wm le | la |w [@ |3 |B ln [8 |= ®& je |& |§ [2 12 |S 12 I§ |S 3 |B |S jz |g /8 |2 |e |S | 5° | | 2 |g ge 8 a 3 ele 3 [+ |9 pveeterr WeiZ01g Yseasay UO!}eN|eAq pue sdiwoucdy YyyeayH Yyoseasay Aaljod YIjeaH JOJ JaqUaD VION December 2022 Appendix |: Pilot Primary Target Populations and Reporting Overall WPC Program UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 212 provides an overview of the primary target populations by WPC Pilot. Each Pilot developed and defined their own target population(s). Primary target populations were defined as those groups that each Pilot aimed to directly influence and designed their services to address the specific needs of these groups. Exhibit 212: Primary Target Population by Pilot Chronic At-risk-of- Physical Homeless- | Justice- WPC Pilot High Utilizers Conditions | SMI/SUD Homeless ness Involved Alameda x x Contra Costa x Kern x x x x Kings x x Los Angeles x xX x x x x Marin x x x Mendocino x Monterey x Napa x x Orange x x Placer x x x x x x Riverside x Sacramento x x San Bernardino x San Diego x x x San Francisco x San Joaquin xX x x x San Mateo x Santa Clara xX Santa Cruz x x Shasta x Solano x x Sonoma x x Ventura x San Benito (SCWPCC) x x x Mariposa (SCWPCC) x x Appendix |: Pilot Primary Target Populations and Reporting | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Chronic At-risk-of- Physical Homeless- | Justice- WPC Pilot High Utilizers Conditions | SMI/SUD Homeless ness Involved Plumas (SCWPCC) x x Source: Initially provided in PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; verified in Pilot specific case studies in February-April 2022. Note: SCWPCC is the Small County Whole Person Care Collaborative. SMI/SUD is serious mental illness and substance use disorder. In Exhibit 213, the target populations of individual enrollees identified by each Pilot in their quarterly Enrollment and Utilization Reports are listed. Pilots varied in whether they reported only on individual-level inclusion in their primary target populations or expanded to report on additional target populations. The COVID-19 target population was added during PY 5 and was not included as a primary target population due to its delayed implementation. Exhibit 213: Enrollee Target Populations Reporting by WPC Pilot, PY 2 to PY 6 2 : 2 BE? = 5 eee | gs 2 a = = = ° WPC Pilot S = Alameda x x x Contra Costa Kern x x x x x Kings x x x Los Angeles x x x Marin x x x Mendocino x x x: x Monterey x x x x x Napa x x x Orange x x x x x Placer x x x x x Riverside x x x x x x Sacramento x x x x x San Bernardino x x San Diego x x x x x x San Francisco x % x San Joaquin x x x x x x San Mateo x x x Santa Clara x x x x x x x Santa Cruz x x x x x x x Shasta x x x x x ScWPCC x x x x x x x Solano x x x x x x x Whole Person Care Final Evaluation Report | Appendix |: Pilot Primary Target Populations and Reporting erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program a ; = a w en . a £Boa eS) 3 Say oo in saz 2 a sa8 33 a E23 = £ eee ae S csés 2 8 22 32 8 WPC Pilot ie Ss Sonoma x x x x x Ventura x x x x x Source: Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Includes 237,603 unique enrollees in WPC Pilots with a target population reported. When count for a target population was less than ten individuals, it was not reported. SMI/SUD is serious mental illness and/or substance use disorder. SCWPCC is the Small County Whole Person Care Collaborative. In the following section, we describe the original target population of each WPC Pilot as described in their application, updates to the target population after implementation as described by Pilot leadership in UCLA-led interviews and the target populations of individual enrollees identified in WPC Quarterly Enrollment and Utilization Reports. We also describe UCLA’s ultimate determination of each Pilot’s primary target population(s). Alameda’s Target Populations Description from Application In their application, the Alameda County Health Care Services Agency (HSCA) identified the target populations of their WPC Pilot as three primary groups: 1. Care Coordination Population — Individuals with complex conditions who may be receiving care management in one system, but actually need care coordination that crosses multiple systems. 2. High Users of Multiple Systems — Medi-Cal beneficiaries who have come in contact with at least two of the following systems: medical, mental health, substance abuse treatment or criminal justice. Individuals are identified using data from the managed care plan, Alameda Alliance for Health, and Alameda County Behavioral Health Care Services. 3. Homeless Persons — Medi-Cal beneficiaries who meet at least one of the Housing and Urban Development (HUD) category definitions of homelessness. Changes during WPC and Primary Target Population Determination Through UCLA conducted interviews, Alameda County HCSA indicated that their target populations included individuals that are on Medi-Cal and had a history of homelessness in the past two years, high utilizers of multiple systems, and Medi-Cal beneficiaries already in a care management program (full-service partnerships). UCLA determined that the primary target populations for Alameda were high utilizers and the homeless. Appendix |: Pilot Primary Target Populations and Reporting | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research er) Health Economics and Evaluation Research Program Pilot Reporting of Target Populations by Enrollee In WPC Enrollment and Utilization Reports, Alameda only reported individuals in four target populations (Exhibit 214). These target populations included the primary target populations of their Pilot as well as two additional target populations. Exhibit 214: Alameda WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD_| Homeless | Homelessness | Involved COVID-19 Individual-level | X x x x Target Populations Reporting Pilot’s Primary | X x Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Contra Costa’s Target Populations Description from Application In their application, Contra Costa Health Services indicated that their target population was “Medi-Cal recipients who are primarily and repeatedly accessing health care services in high- acuity settings due to the complexity of their unmet medical, behavioral health and social needs.” More specifically, the Pilot used data to identify individuals with the following in one year: skilled nursing facility stay, more than six ED visits, more than six inpatient days or more than two inpatient admissions. They aimed to use their data warehouse to develop a data- driven, real-time algorithm to identify individuals that meet the target population criteria. Changes during WPC and Primary Target Population Determination Through UCLA conducted interviews, Contra Costa indicated that they developed a sophisticated predictive risk model that included information from a variety of county sources. These data sources included information on a potential enrollee’s service utilization, chronic conditions, justice involvement and social determinants of health. Contra Costa’s primary target population was solely high utilizers to provide enrollment flexibility. Pilot Reporting of Target Populations by Enrollee Whole Person Care Final Evaluation Report | Appendix |: Pilot Primary Target Populations and Reporting erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program In Contra Costa’s enrollment and utilization reports, they reported WPC enrollees in one target population: high utilizers. Given that their predictive risk model aimed to identify individuals that were high utilizers or are at-risk of becoming a high utilizer, their individual reporting aligns with their primary target population (Exhibit 215). Exhibit 215: Contra Costa WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD | Homeless | Homelessness | Involved COVID-19 Individual-level | X Target Populations Reporting Pilot’s Primary | X Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Kern’s Target Populations Description from Application In their application, Kern Medical Center (KMC) identified their target population as high utilizers, defined as high utilizers of emergency and inpatient services, with a focus on individuals that are homeless, at-risk of homelessness or have been recently incarcerated. Additionally, all enrollees were required to be eligible for Medi-Cal. The local health plans were supposed to provide lists of individuals that met these criteria. Changes during WPC and Primary Target Population Determination Through UCLA conducted interviews, KMC indicated that changes to their target populations occurred due to changes in their program. The original intention was to identify high utilizers through lists provided by the two local health plans. However, KMC identified several limitations to this method, including: e Homeless individuals and those at-risk of homelessness were not identified or captured by the health plans. Soon-to-be-released or recently incarcerated individuals were not captured by the health plans. e The contact information provided by the health plans was typically not current or effective. Appendix |: Pilot Primary Target Populations and Reporting | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program As a result, KMC modified their outreach and recruitment process to include referrals from the Housing Authority, in addition to the placement of a physician within jail that identified soon- to-be-released inmates for inclusion in the program. KMC also created a website and email address that allowed for self-referral into the program. As a result, the target population no longer required individuals to be high utilizers - if need was identified through these other recruitment mechanisms, the individual was enrolled. As a result, UCLA identified the primary target population for Kern as high utilizers, homeless, at-risk-of-homelessness and justice- involved. Pilot Reporting of Target Populations by Enrollee Through access to several data sources, including behavioral health data and social determinant assessments, KMC was able to assess enrollees for all target populations identified by the State, apart from COVID-19. These reported target populations included those that were targeted by the Pilot (high utilizers, homeless, at-risk-of-homelessness and justice-involved) and target populations not directly targeted by the Pilot (chronic physical conditions and SMI/SUD; Exhibit 216). Exhibit 216: Kern WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD | Homeless | Homelessness | Involved COVID-19 Individual-level | X x x x x x Target Populations Reporting Pilot’s Primary | X x x x Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Whole Person Care Final Evaluation Report | Appendix |: Pilot Primary Target Populations and Reporting iembereee UCLA Center for Health Policy Research ia Health Economics and Evaluation Research Program Kings’ Target Populations Description from Application Kings Area Resource Enhanced Linkages (KARELink) aimed to reduce the number of adults with mental illnesses and co-occurring substance use disorders in their jails and to build a collaborative bridge to wellness for people with behavioral health issues who are homeless or at-risk of homelessness. The target population had to have a substance use disorder, mental health issue or chronic health condition of diabetes or high blood pressure. In their application, Kings County Human Services Agency (KINGS HSA) indicated that their primary target population was the high cost, high utilizers of services who accessed care primarily on a crisis basis via an emergency room or did not access care on an ongoing basis and were often incarcerated. Individuals had to have at least one of the following: 1. Substance use disorder 2. Mental health issue 3. Chronic health conditions (diabetes or hypertension) Changes during WPC and Primary Target Population Determination Through UCLA structured interviews, KARELink leadership indicated that their target population was primarily SMI/SUD with chronic physical conditions. High utilizers and justice-involved were a subset of this population, but were not required for enrollment. As a result, UCLA determined their primary target populations to include SMI/SUD and chronic physical conditions. Pilot Reporting of Target Populations by Enrollee Initially, KARELink reported on four target populations: high utilizers, chronic physical conditions, SMI/SUD and justice-involved (Exhibit 217). After some changes to their reporting process, they were no longer reporting on high utilizers and justice-involved. The data used to determine an enrollee’s target population came from the screening and assessment of the client by care coordinators. Exhibit 217: Kings WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD | Homeless | Homelessness | Involved COVID-19 Individual-level x x x x x x Target 400 | Appendix |: Pilot Primary Target Populations and Reporting | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Populations Reporting Pilot’s Primary x x Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Los Angeles’ Target Populations Description from Application In their application, Los Angeles County Department of Health Services identified six target populations for their WPC Pilot: 1) individuals experiencing homelessness, 2) justice-involved individuals or individuals who are high utilizers of acute care services due to 3) serious mental illness (SMI), 4) substance use disorder (SUD), 5) complex medical issues, and 6) high-risk pregnant women. There was an overlap between the populations and where they did not overlap they still shared similar traits, including difficulty engaging into programs and common challenges to manage debilitating social inequities. Therefore, individuals could enter through any target population. The homeless target population included all homeless or at-risk of homelessness individuals that were chronically homeless, had a physical or mental disability, had two or more chronic medical or behavioral health (e.g., mental health or substance use disorder) conditions, or were recent and/or recurrent care utilizers (e.g., multiple emergency department (ED) visits or hospitalizations for medical or psychiatric issues). The justice-involved target population included justice system-involved individuals who were at the highest risk of medical, psychiatric, and/or substance use decompensation with one or more of the following: 1) recent or recurrent acute care utilization, 2) multiple and/or complex chronic medical conditions, 3) serious mental illness, 4) substance use disorders, or 5) pregnancy. The mental health target population criteria varied depending on the program through which the enrollee were identified. For the Intensive Service Recipient (ISR) program, individuals must have had a severe mental health diagnosis and a minimum of six psychiatric hospital admissions in the previous year. For the Residential and Bridging Care (RBC) program, individuals must have had a serious mental illness and/or co-occurring substance use disorders in psychiatric inpatient units, or exited Institutions of Mental Disease (IMDs) and have been treated in Whole Person Care Final Evaluation Report | Appendix |: Pilot Primary Target Populations and Reporting erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program enriched residential settings. For the Kin to Peer (KTP) program, individuals must have lacked family or healthy social support systems and have been eligible for the ISR or RBS programs. The substance use disorder target population had to have a substance use disorder and at least one of the following: 1) three or more ED visits related to SUD within in the past year, 2) two or more inpatient admissions for physical and/or mental health conditions, 3) three or more sobering center visits within the past year, 4) homeless (meeting HUD criteria), 5) part of foster system, 6) more than two residential SUD treatment admission within the past year, 7) history of two or more incarcerations with drug use, 8) drug court referral (to either Sentence Defender Court or Women’s Re-Entry Court, and/or 9) history of overdose in the past two years. The medically complex target population consisted of individuals with the Transitions of Care (TOC) program who were admitted to a Lanterman-Petris-Short (LPS) Act general acute care hospital who were on the LANES (Los Angeles Network for Enhanced Services) HIE with three or more admissions (medical or psychiatric) within the last six months and at least one of the following: 1) one or more avoidable hospital admissions related to a chronic medical problem, 2) homelessness, 3) SUD, 4) mental health disorder, and/or 5) incarceration within the last month. The expectant mothers target population included pregnant women with one or more of the following: 1) homeless or at-risk of homelessness, 2) physical or mental disability, 3) chronic medical or behavioral health condition, 4) soon to be or recently released from incarceration. Changes during WPC and Primary Target Population Determination Through UCLA conducted interviews, Los Angeles indicated that target populations remained as described in the application. As a result, UCLA determined Los Angeles’ primary target populations included all six standardized target population groups. Pilot Reporting of Target Populations by Enrollee Los Angeles’ WPC Pilot reported on all six target populations identified by DHCS (Exhibit 218). In order to determine who was reported in each target population, they used data collected on target populations and homeless status from different programs in the pilot. If target populations information was unavailable, they determined enrollee’s status based on program enrollment. For example, all individuals in the sobering centers were included in the SMI/SUD target population and all individuals in the re-entry programs were included in the justice- involved target population. Appendix |: Pilot Primary Target Populations and Reporting | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Exhibit 218: Los Angeles WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD | Homeless | Homelessness | Involved COVID-19 Individual-level | X x x x x x Target Populations Reporting Pilot’s Primary | X x x x x x Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Marin’s Target Populations Description from Application In their application, County of Marin’s Department of Health and Human Services (Marin HHS) focused on two target populations: 1. Individuals who experienced homelessness or were at-risk of homelessness (including those released from institutions) and 2. Individuals who experienced complex medical conditions, behavioral health issues, and/or lacked social supports that interfered with standards of care, which resulted in high utilization and costs. More specifically, the latter population included the top 10% of Medi-Cal beneficiaries by spending who had a diagnosis of a mental disorder, substance use disorder, traumatic brain injury, dementia or opioid use, two or more chronic conditions, and/or repeated incidents of avoidable emergency use, hospital admissions or nursing facility placement. Changes during WPC and Primary Target Population Determination Through UCLA interviews with Pilot leadership, Marin HHS indicated that their target population had expanded to include three groups. These groups were linked to their per- member-per-month (PMPM) bundles that provided care coordination. The homeless target population received housing based case management. The high utilizers received comprehensive case management. Lastly, individuals with a mental illness, substance use disorder and/or other health conditions that were not eligible for specialty Medi-Cal mental health plans received case management for individuals with mental health conditions and Whole Person Care Final Evaluation Report | Appendix |: Pilot Primary Target Populations and Reporting erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program complex psychosocial challenges. As a result, UCLA identified their primary target populations as high utilizers, homeless and at-risk-of-homelessness. Pilot Reporting of Target Populations by Enrollee In enrollment and utilization reports, Marin HHS reported on three target populations: high utilizers, homeless and at-risk of homelessness (Exhibit 219). The high utilizer target population aligned with the complex Med-Cal beneficiary population. The homeless and at-risk of homelessness populations aligned with the homeless target population. The third target population that aimed to address individuals with mental health conditions and complex psycho-social challenges often did not meet the SMI/SUD criteria because those with SMI could be eligible for specialty Medi-Cal mental health plans. Exhibit 219: Marin WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD_| Homeless | Homelessness | Involved COVID-19 Individual-level | X x x Target Populations Reporting Pilot’s Primary | X x x Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Mariposa’s Target Populations Description from Application In their application, Mariposa County Human Services Department indicated that their target population would be individuals with a behavioral health condition (mental health, substance abuse or co-occurring diagnosis) and one or more of the following: e Repeated incidents of emergency department (ED) use, hospital admissions or nursing facility placement © Two or more chronic conditions © Homeless or at-risk of homelessness e Recently released from institutions (e.g., hospital, county jail, institutions for mental diseases, skilled nursing facility, etc.) or connection to the criminal justice system. EI Appendix |: Pilot Primary Target Populations and Reporting | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research . 7 December 2022 Health Economics and Evaluation Research Program Changes during WPC and Primary Target Population Determination During UCLA structured interviews, Mariposa indicated that their target population had evolved through implementation. Their focus shifted to high users of the ED due to the small size of the local ED (four beds). Their target population was then defined as high utilizers (three or more ED visits or one hospital admission per year) who had SMI/SUD and any of the following: homelessness, chronic conditions or justice-involved. As a result, UCLA identified their primary target populations as high utilizers and SMI/SUD. Pilot Reporting of Target Populations by Enrollee While Mariposa reported on all seven of the DHCS-designated target populations, the focus of their program was high utilizers and SMI/SUD (Exhibit 220). In order to determine a potential enrollee’s utilization and SMI/SUD status they used data from the managed care plan in addition to self-report and observation. Exhibit 220: Mariposa WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD | Homeless | Homelessness Involved COVID-19 Individual-level | X x x x x x x Target Populations Reporting Pilot’s Primary | X x Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Mendocino’s Target Populations Description from Application In their application, Mendocino County Health and Human Services Agency (HHSA) indicated that their target population would be individuals with a SMI. They would prioritize high utilizers of mental health and/or medical services and those who experienced homelessness or housing instability, co-occurring SUD and/or recent interactions with the criminal justice system. In addition, enrollees needed to be eligible for Medi-Cal. Changes during WPC and Primary Target Population Determination Whole Person Care Final Evaluation Report | Appendix |: Pilot Primary Target Populations and Reporting a erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Through structured interviews, UCLA determined that the target population for Mendocino County HHSA was still individuals with SMI, but in order to prioritize enrollees, they also required that enrollees fit into at least two other DHCS-defined target population groups: homeless, at-risk of homelessness, high utilization and justice involvement. UCLA determined their primary target population was SMI/SUD. Pilot Reporting of Target Populations by Enrollee In their enrollment and utilization reports, Mendocino County HHSA reported on all target populations (Exhibit 221). All of their enrollees were in the SMI/SUD target population. Because self-report was the data source for their target population, it is likely errors occurred in the target populations. Additionally, different agencies had different methodologies for reporting which resulted in inconsistencies among their population. Exhibit 221: Mendocino WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD | Homeless | Homelessness | Involved COVID-19 Individual-level | X x x x x x Target Populations Reporting Pilot’s Primary xX Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Monterey’s Target Populations Description from Application The Monterey County Health Department aimed to target homeless and chronically homeless Medi-Cal beneficiaries or Medi-Cal eligible individuals, which included those recently released from jail. Potential enrollees had to have two or more of the following: e Two or more mental health unit admissions in the prior year, * Two or more chronic health diagnoses ® Two or more ED visits within the past 12 months, @ One or more hospital admission within the prior 12 months or, Ea Appendix |: Pilot Primary Target Populations and Reporting | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program e Two or more prescribed medications (antidepressants, antipsychotics, mood stabilizers, diabetes medication, antihypertensives, cholesterol lowering medications, inhaled corticosteroids and bronchodilators, seizure medications and anticoagulants). More specifically, Monterey County intended to use the HUD McKinney-Vento Homeless Assistance Act definition of homeless and the 2016 HUD Hearth definition of chronically homeless. Changes during WPC and Primary Target Population Determination Through UCLA interviews with Pilot leadership, Monterey County Health Department indicated that after implementation, they continued to focus on homeless individuals. They did not provide services to individuals that were at-risk of homelessness, rather they needed to already be living on the streets to receive services. The majority of the enrollees were also high- utilizers. UCLA determined that the primary target population of Monterey was homeless. Pilot Reporting of Target Populations by Enrollee Monterey County WPC pilot reported on six of the seven DHCS-defined target populations: high utilizers, chronic physical conditions, SMI/SUD, homeless, at-risk of homelessness, and justice- involved (Exhibit 222). Although they reported on many of the target populations, the main target population of the program was homeless individuals. The other criteria were not a requirement to participate and were used mainly to prioritize those that were enrolled in the program. Exhibit 222: Monterey WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD_| Homeless | Homelessness | Involved COVID-19 Individual-level | X x x x x x Target Populations Reporting Pilot’s Primary x Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Napa’s Target Populations Description from Application Whole Person Care Final Evaluation Report | Appendix |: Pilot Primary Target Populations and Reporting erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program In their application, Napa County Health and Human Services Agency (HHSA) indicated that their target population would be individuals experiencing homelessness or at-risk of homelessness. They would prioritize these individuals for enrollment if they were high system users and have a physical disability, serious mental illness or substance use disorder, or co- occurring disorders. Changes during WPC and Primary Target Population Determination Through structured interviews with UCLA, Napa County HHSA indicated that they have mainly focused on chronically homeless individuals during the first phase of their Pilot. They used the HUD definition of homelessness and found that most of their chronically homeless enrollees have a SMI, SUD or other physical disability. However, they were no longer focusing on the criteria they outlined in their application for prioritizing enrollees. In addition, due to unexpected difficulties in gaining access to partner data, it was difficult to determine whether or not potential enrollees had the priority criteria prior to completion of a release of information consent form during the enrollment process. Ultimately, UCLA determined that their primary target populations were homeless or at-risk-of-homelessness. Pilot Reporting of Target Populations by Enrollee In their enrollment and utilization reports, Napa County HHSA reported on three target populations (Exhibit 223). They aimed to target homeless and individuals that are at-risk of homelessness, starting the program by only enrolling those that have been chronically homeless. Exhibit 223: Napa WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD_| Homeless | Homelessness | Involved COVID-19 Individual-level | X x x Target Populations Reporting Pilot’s Primary x xX Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Orange’s Target Populations Description from Application Ey Appendix |: Pilot Primary Target Populations and Reporting | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program In their application, County of Orange Health Care Agency (HCA) indicated that they would target two populations: 1) homeless and 2) SMI and SMI homeless. The first target population was individuals experiencing homelessness. To ensure that this target population would benefit from WPC services, they focused on those individuals that had visited the ER for care, particularly those that accessed the ED two or more times in a rolling three-month period. The second target population included individuals with serious mental illness (SMI) and SMI homeless. Given that these individuals were served through the County’s Behavioral Health Services and regulations prevented sharing of data from Behavioral Health, these individuals could not be properly identified through the initial homeless search. Changes during WPC and Primary Target Population Determination Through structured interviews, UCLA determined that the target population of Orange HCA’s WPC pilot had evolved slightly from what was originally proposed in their application. Specifically, the target population of the Pilot was defined as homeless individuals. Individuals experiencing homelessness with SMI was a subpopulation of their target population. In general, individuals were engaged and enrolled into the Pilot through contacts with participating emergency departments, clinics and shelters and through outreach programs known to individuals experiencing homelessness. The additional criteria listed in the application was thus not required, but would likely be met given the method of engagement. UCLA determined that their primary target population were homeless and SMI/SUD. Pilot Reporting of Target Populations by Enrollee In their enrollment and utilization reports, Orange HCA reported on six target populations (Exhibit 224). The at-risk-of-homelessness target population was only used when an enrolled individual had initially secured housing. Once in the at-risk-of-homelessness target population, individuals were disenrolled from the pilot if they remained housed for six months. Exhibit 224: Orange WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD | Homeless | Homelessness | Involved COVID-19 Individual-level | X x x x x x Target Populations Reporting Pilot’s Primary x x Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Whole Person Care Final Evaluation Report | Appendix |: Pilot Primary Target Populations and Reporting iembereee UCLA Center for Health Policy Research ia Health Economics and Evaluation Research Program Placer’s Target Populations Description from Application In their WPC application, Placer County Health and Human Services (HHS) indicated that they would focus on several target populations for their pilot to ensure serving enough individuals even though Placer is not a small county. They aimed to serve 450 adult individuals throughout the duration of the program who fit the following target populations: 1. History of repeated incidents of avoidable ED use and hospital readmissions (top 5% of their service population in terms of cost of services) 2. Two or more chronic health conditions (including heart disease, diabetes, COPD, unmanaged cholesterol, obesity, and high blood pressure) 3. Severe mental health diagnoses and/or substance use disorder 4. Currently homeless or at-risk of homelessness 5. Scheduled for release from jail and meet at least one WPC target population criteria Additionally, individuals needed to be eligible for Medi-Cal. Changes during WPC and Primary Target Population Determination Through structured interviews with UCLA, they indicated that they had purposefully kept their target population as broad as possible in order to allow for flexibility in their program. Not only would they be able to serve more individuals, but they would also be able to test strategies to help a variety of populations. Ultimately, UCLA determined that Placer’s primary target populations included all six DHCS-defined groups. Pilot Reporting of Target Populations by Enrollee At the individual-level, Placer reported enrollees in the six original target populations ( Exhibit 225). They did not report on inclusion in the COVID-19 target population after it was added to the program. Exhibit 225: Placer WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD_| Homeless | Homelessness | Involved | COVID-19 Individual-level | X x x x x x Target Populations Reporting Appendix I: Pilot Primary Target Populations and Reporting | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research . 7 1DYLer-1 Lay L0y ed Health Economics and Evaluation Research Program Pilot’s Primary | X x x x x x Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Riverside’s Target Populations Description from Application In their application, Riverside University Health System (RUHS) was targeting probationers with the following criteria: e New probationers ¢ On probation for at least one full year e At-risk of or experiencing homelessness e Have a behavioral health diagnosis e Have a physical health diagnosis Potential enrollees would be screened and enrolled at their first probation visit. Changes during WPC and Primary Target Population Determination During UCLA structured interviews, RUHS leadership indicated that their target population remains probationers. UCLA determined their primary target population was justice-involved. Pilot Reporting of Target Populations by Enrollee Initially, RUHS believed that enrollees needed to meet all six original target populations designated by DHCS for WPC. However, after the first year of enrollment, DHCS clarified that only screening and Medi-Cal eligibility was required. As a result, all enrollees are in the original six target populations in the first year, but are no longer in all the target populations starting in the second year (Exhibit 226). Exhibit 226: Riverside WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD | Homeless | Homelessness | Involved COVID-19 Individual-level | X x x x x x x Target Whole Person Care Final Evaluation Report | Appendix |: Pilot Primary Target Populations and Reporting erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Populations Reporting Pilot’s Primary x Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Sacramento’s Target Populations Description from Application In their application, the city of Sacramento indicated that their Pilot would target individuals with repeated incidents of avoidable ED use and/or hospital admissions, defined as two or more ED visits or inpatient hospitalizations or one ED visit and two or more comorbid conditions, and those who are homeless or at-risk-of-homelessness. Additionally, potential enrollees would need to be Medi-Cal enrolled or eligible and reside in Sacramento County. Changes during WPC and Primary Target Population Determination Through structured interviews, UCLA determined that the target population of Sacramento’s WPC Pilot remained high utilizers that are homeless. The data used to determine an enrollee’s eligibility has evolved over implementation. Sacramento initially tried to get a list of potential enrollees from the health plan but found it was too difficult to outreach and engage through this method. They then transitioned to a hot-spotting method, which sought out locations where their target populations tended to be and developed a referral system at the ERs and hospitals. Ultimately, the pilot’s primary target populations were homeless and high utilizers. Pilot Reporting of Target Populations by Enrollee In their enrollment and utilization reports, Sacramento initially reported on all target populations apart from justice-involved (Exhibit 227). Through clarification on reporting requirements with DHCS, they stopped reporting on all the target populations that were not in their target population criteria (chronic physical conditions and SMI/SUD). Sacramento had strict eligibility criteria and therefore, individuals that were not reported as high utilizers and homeless or at-risk of homelessness were likely misreported. Exhibit 227: Sacramento WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD | Homeless | Homelessness | Involved COVID-19 Appendix |: Pilot Primary Target Populations and Reporting | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Individual-level | X x x x x Target Populations Reporting Pilot’s Primary | X x Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. San Benito’s Target Populations Description from Application In their application, San Benito County Health and Human Services Agency indicated that their target population would be individuals who are homeless or at-risk of homelessness and have one or more of the following: ¢ Behavioral health condition (mental illness, substance abuse or co-occurring diagnosis) e Repeated incidents of ED use, hospital admissions or nursing facility placement * Two or more chronic conditions ¢ Recently released from institutions or connections to the criminal justice system. Additionally, enrollees needed to be between 18 and 64 years old and eligible for Medi-Cal. Changes during WPC and Primary Target Population Determination During UCLA structured interviews, San Benito indicated that through implementation the focus of the program had shifted to high-utilizing individuals that are homeless or at-risk of homelessness. This shift was mainly brought on by their first enrollees, whom typically were homeless or at-risk of homelessness and had a connection to the criminal justice system. Without evidence of high utilizations in the past, the goals of the Pilot to reduce the use of avoidable ED use and inpatient hospitalization were not going to be realized and these individuals were not benefiting from the services provided. Additionally, these first enrollees were often disenrolled quickly due to lack of engagement. UCLA determined the primary target populations to be high utilizers, homeless and at-risk-of-homelessness. Pilot Reporting of Target Populations by Enrollee Whole Person Care Final Evaluation Report | Appendix |: Pilot Primary Target Populations and Reporting Try UCLA Center for Health Policy Research Health Economics and Evaluation Research Program While San Benito reports on all seven of the DHCS-designated target populations, the focus of their program was high utilizers, homeless and at-risk-of-homelessness (Exhibit 228). In order to determine a potential enrollee’s utilization and homelessness status they used data from the hospital in addition to self-report and observation. Exhibit 228: San Benito WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD | Homeless | Homelessness | Involved COVID-19 Individual-level | X xX x x x x x Target Populations Reporting Pilot’s Primary | X x x Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. San Bernardino’s Target Populations Description from Application In their application, San Bernardino County’s Designated Public Hospital, Arrowhead Regional Medical Center (ARMC) indicated they aim to target the most vulnerable population at-risk for frequent, emergency medical and behavioral services. In order to determine the population, they collected data from ARMC, Public Health, and Behavioral Health and scored individuals based on emergency visits, inpatient hospital stays and urgent care visits. ARMC planned to update the list yearly and methodology for scoring as necessary. Initially, the scoring has been based on the following rubric: agora o KS Point Value Given Hospital medical inpatient 1 point per day ED encounter 3 points per encounter/admission/event Psychiatric/SUD inpatient admission 3 points per admission Psychiatric/SUD acute care 1 point per day Urgent/express/crisis care 1 point per event Public health utilization 0.5 point per encounter Flagged as Chronically Homeless (overrides either | 300 points below) Most recent prior residence homeless 200 points Most recent prior residence temporary (receiving | 150 points services, so at risk of homelessness) Appendix |: Pilot Primary Target Populations and Reporting | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Most recent prior residence permanent (receiving | 100 points services, so at risk of homelessness) This rubric was supposed to prioritize individuals that are both high utilizers and homeless or at-risk of homelessness. In addition, enrollees needed to be Medi-Cal eligible. Changes during WPC and Primary Target Population Determination ARMC continued to use a list of potential enrollees created using a scoring algorithm. However, there have been updates to the scoring algorithm. For example, the algorithm initially counting each inpatient day has been changed to counting each admission. Additionally, there were no longer elements about homelessness in the algorithm and instead chronic physical conditions have been included. ARMC used this system so that everyone in the county had the opportunity to be part of the Pilot. They were concerned that if they used referrals, there would be bias towards certain providers. The focus of the program was to address individuals with high utilization. Chronic physical conditions helped prioritize those individuals with potential for intervention. Ultimately, UCLA determined that high utilizers was the primary target population. Pilot Reporting of Target Populations by Enrollee In enrollment and utilization data, ARMC reported on two target populations that aligned with their target population scoring algorithm: high utilizers and chronic physical conditions (Exhibit 229). Exhibit 229: San Bernardino WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD | Homeless | Homelessness | Involved COVID-19 Individual-level | X x Target Populations Reporting Pilot’s Primary | X Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Whole Person Care Final Evaluation Report | Appendix |: Pilot Primary Target Populations and Reporting iembereee UCLA Center for Health Policy Research Health Economics and Evaluation Research Program San Diego’s Target Populations Description from Application In their application, the County of San Diego Health and Human Services Agency indicated that their target population would be high-cost, frequent users of ED and/or inpatient services identified by the Medi-Cal managed care plans who: e Are currently experiencing homelessness or are at-risk of homelessness and e Have a mental health condition, substance use disorder, or chronic physical health condition/s In addition, enrollees needed to be Medi-Cal eligible. San Diego defined high users as individuals having more than $40,000 in Medi-Cal paid claims and at least five ED visits or three inpatient hospitalizations. They aimed to exclude individuals with terminal illnesses. Changes during WPC and Primary Target Population Determination Due to the normal lag in Medi-Cal claims, which resulted in a delay identifying high-utilizers with health conditions or behavioral disorders, San Diego has focused less on lists of eligible enrollees from their managed care plans and relied more on community referrals. San Diego still defined their target population as individuals that are homeless or at-risk of homelessness and high utilizers. However, they have made a few exceptions to the high utilizer criteria if it was apparent that the individual had high need and was likely to end up a high utilizer without intervention. San Diego intended for the additional criteria included in the target population definition to assist in prioritizing enrollees and describe the enrolled population. UCLA determined the primary target populations to be high utilizers, homeless and at-risk-of- homelessness. Pilot Reporting of Target Populations by Enrollee San Diego reported on all six original target populations designated by DHCS (Exhibit 230). For first two quarters of 2018, they were building their relationship with the justice system and therefore were not able to systematically capture information on this target population. Additionally, as they developed the system used to capture all the information needed to determine an enrollee’s target populations, there was a potential lag in the time to collect the necessary information. As a result, the most complete target population information might not have been available in the first months of enrollment. Appendix |: Pilot Primary Target Populations and Reporting | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Exhibit 230: San Diego WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD | Homeless | Homelessness | Involved COVID-19 Individual-level | X x x x x x Target Populations Reporting Pilot’s Primary | X x x Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. San Francisco’s Target Populations Description from Application In their application, the San Francisco Department of Public Health (SFDPH) indicated that their target population was Medi-Cal enrolled homeless adults. In order to prioritize individuals for WPC services, SFDPH developed a risk-based stratification of the homeless population. Severe risk has been defined as the top 5% of urgent/emergency services and individuals homeless for more than 10 years (in SFDPH’s Coordinated Care Management System (CCMS)). High risk was defined as the top 5% of urgent/emergency services and individuals homeless for less than 10 years (in CCMS). Elevated risk included individuals who were not part of the top 5% of urgent/emergency services and were homeless for less than 10 years (in CCMS). Changes during WPC and Primary Target Population Determination Through UCLA conducted interviews, San Francisco indicated the target population remained individuals experiencing homelessness identified through CCMS. They continued to use historical data to stratify their target population into severe risk, high risk and elevated risk. UCLA determined the primary target population was homeless. Pilot Reporting of Target Populations by Enrollee In San Francisco’s enrollment and utilization reports, they reported WPC enrollees in three possible target populations: high utilizers, homeless, and COVID-19 (Exhibit 231). All enrollees were included in the homeless target population. Whole Person Care Final Evaluation Report | Appendix |: Pilot Primary Target Populations and Reporting Geant eee UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Exhibit 231: San Francisco WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD | Homeless | Homelessness | Involved COVID-19 Individual-level | X x x Target Populations Reporting Pilot’s Primary xX Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. San Joaquin’s Target Populations Description from Application In their application, the San Joaquin County Health Care Services Agency indicated that they would target three populations: 1. Adult Health Plan of San Joaquin (HPSHJ) that are assigned to the FQHC look-alike clinics and are over utilizers of the emergency department 2. Adults with a mental health and/or substance use disorder 3. Adults experiencing homelessness or at-risk of homelessness upon discharge from the hospital, medical center, psychiatric health facility, or county jail In addition, the enrollee needed to be a Medi-Cal beneficiary. Changes during WPC and Primary Target Population Determination Through UCLA conducted interviews, San Joaquin indicated that all enrollees had to fit into at least one target population, but often they fit into more than one. An enrollee might be referred for homelessness, but then later identified as a high utilizer as well. Data came from referral forms, EHS, HMIS, HIE, jails, anong many other sources. UCLA determined that high utilizers, SMI/SUD, homeless and at-risk-of-homelessness were the primary target populations. Pilot Reporting of Target Populations by Enrollee San Joaquin reported individuals in all DHCS-defined target populations except chronic physical conditions ( Appendix I: Pilot Primary Target Populations and Reporting | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Exhibit 232). San Joaquin did not use SMI/SUD in 2017 because partners were not providing the data as they were finalizing data sharing agreements. Many enrollees had mild to moderate mental illness rather than serious mental illness so were not identified as having mental illness. They added justice-involved later in 2018. Exhibit 232: San Joaquin WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD | Homeless | Homelessness | Involved COVID-19 Individual-level | X x x x x x Target Populations Reporting Pilot’s Primary | X x xX x Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. San Mateo’s Target Populations Description from Application In their application, San Mateo County Health System identified three target populations for their Pilot. These target populations included: e High utilizers with mental illness and/or medical conditions who present frequently to EDs, Psychiatric Emergency Services (PES), and/or have avoidable or extended stays in residential treatment e High utilizers with untreated SUD ¢ High utilizers with similar clinical profiles previously listed, but are also identified homeless or recently released from jail Changes during WPC and Primary Target Population Determination San Mateo has found in practice that these categories were often fluid. As initially designed, the target population was supposed to map to specific teams, but this has not been the case. Asa result, the PMPM bundle did not accurately tell which services the client was receiving. If enrollees got a Behavior Health and Recovery Services (BHRS) “touch”, they were in that bundle, but Bridges to Wellness served people in all three target populations and across all PMPMs. The initial list of enrollees was identified through referrals and lists of individuals with Whole Person Care Final Evaluation Report | Appendix |: Pilot Primary Target Populations and Reporting Try UCLA Center for Health Policy Research Health Economics and Evaluation Research Program more than four ED visits. Ultimately, UCLA determined that high utilizers was the primary target population. Pilot Reporting of Target Populations by Enrollee All enrollees were in the high utilizer target population (Exhibit 233). San Mateo determined if an enrollee was also included in the SMI/SUD target population depending on the services the enrollee received. Enrollees were included in the homeless target population based on registration information from their electronic health record. This information was not always up to date and it is likely that the number of enrollees experiencing homelessness has been under reported. Exhibit 233: San Mateo WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD | Homeless | Homelessness | Involved COVID-19 Individual-level | x x x Target Populations Reporting Pilot’s Primary | X Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Santa Clara’s Target Populations Description from Application In their application, Santa Clara Valley Health and Hospital System (SCVHHS) indicated that their target population was high utilizers of multiple systems (HUMS) who are Medi-Cal enrolled, engaged in two or more systems of care and in the top 5% of utilizers for SCVHHS encounters over the past year. While they acknowledged that many individuals within this population have co-occurring physical and behavioral health issues, experience homeless and/or be justice- involved, they believed the program could make the most impact with the top 5% HUMS. Changes during WPC and Primary Target Population Determination Through UCLA conducted interviews, Santa Clara indicated that the Center for Population Health Improvement (CPHI) aggregated data from SCVHHS departments (e.g., Santa Clara Valley Medical Center, Office of Supportive Housing, Custody, Behavioral Health) and Valley Health Appendix |: Pilot Primary Target Populations and Reporting | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Plan claims. Based on these data sources they developed a statistical point system which assigned different values depending on the patient’s type of clinical encounters in the past year (e.g., emergency and psychiatric encounters receive more points than an ambulatory care visit; inpatient stays are capped at 75th percentile). Santa Clara targeted the top 10% high-scoring individuals for enrollment in the program (~10,000 potential clients). Ultimately, this system aimed to identify high utilizers, which UCLA determined as the primary target population. Pilot Reporting of Target Populations by Enrollee In Santa Clara’s enrollment and utilization reports, they identified individuals in all possible target populations (Exhibit 234). Exhibit 234: Santa Clara WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD | Homeless | Homelessness | Involved COVID-19 Individual-level | X x x x x x x Target Populations Reporting Pilot’s Primary | X Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Santa Cruz’ Target Populations Description from Application In their application, the County of Santa Cruz Health Services Agency (HAS) identified the WPC Pilot target population as adult Medi-Cal beneficiaries with at least one of the following characteristics: Repeated incidents of avoidable emergency use, hospital admissions, or nursing facility placement * Two or more chronic conditions e Mental health and/or substance use disorders e Currently experiencing homelessness Whole Person Care Final Evaluation Report | Appendix |: Pilot Primary Target Populations and Reporting erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program e At-risk of homelessness and require intensive housing support to live in the community due to their mental illness, substance use disorder and co-occurring health condition ¢ Post incarceration; could include probation or parole status. Changes during WPC and Primary Target Population Determination Through UCLA conducted interviews, Santa Cruz indicated that they focused on those with co- occurring behavioral health (including SUD) and physical chronic conditions. In particular, they focus on high-cost chronic conditions, but they also took into account high-utilization or medication history when determining if an individual met their criteria. UCLA determined the primary target populations were chronic physical conditions and SMI/SUD. Pilot Reporting of Target Populations by Enrollee While the WPC Pilot reports on all seven target populations, the main focus of their pilot was individuals with co-occurring behavioral health and chronic physical conditions (Exhibit 235). This has been reflected by the fact that almost all enrollees were in the SMI/SUD target population, except for individuals with mild or moderate mental illness. Exhibit 235: Santa Cruz WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD | Homeless | Homelessness | Involved COVID-19 Individual-level | X x x x x x x Target Populations Reporting Pilot’s Primary x x Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Shasta’s Target Populations Description from Application In their application, the Shasta County Health and Human Services Agency (HHSA) indicated that their target population was adults ages 18 to 64 with two or more ED visits or hospitalizations in the last three months and are homeless or at-risk of homelessness. Potential enrollees also needed to fulfil one or more of the following criteria: Appendix I: Pilot Primary Target Populations and Reporting | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program HASTE Aa: ¢ SMI diagnosis e SUD diagnosis e Undiagnosed/undisclosed opioid addiction Changes during WPC and Primary Target Population Determination Through UCLA conducted interviews, Shasta County HHSA indicated that their target population was high utilizers with an emphasis on individuals with chronic illness, SUD and homelessness. UCLA determined that their primary target population was high utilizers. Pilot Reporting of Target Populations by Enrollee While Shasta reported on all target populations except for justice-involved and COVID-19, the pilot aimed to provide services for individuals that met the high utilizer criteria (Exhibit 236). Exhibit 236: Shasta WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD_| Homeless | Homelessness | Involved COVID-19 Individual-level | X xX x x x Target Populations Reporting Pilot’s Primary | X Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Whole Person Care Final Evaluation Report | Appendix |: Pilot Primary Target Populations and Reporting err UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Solano’s Target Populations Description from Application In their application, Solano identified their target populations as individuals with the highest medical utilization, repeated incidents of avoidable ED use, and two or more chronic and serious health conditions, with at least one being mental health and/or substance use disorders. Enrollees were identified using data from Partnership Health Plan. Changes during WPC and Primary Target Population Determination Through UCLA conducted interviews, Solano indicated that outreach and enrollment was originally intended to be based on a list compiled by the managed care organization which would identify high utilizers with chronic conditions. However, they found that individuals on the list were not always appropriate for the program and some individuals were not willing to participate in the program. Therefore, they expanded their approach to include referrals from community based organizations (CBOs), emergency departments and clinics. Individuals referred into the program still needed to meet the Pilot eligibility criteria (e.g., high utilizer with two or more chronic conditions, one of which must be SMI and/or SUD). Solano expanded its definition of high utilizers but individuals still needed to have repeated, avoidable ED use. The majority of enrollees were homeless or at-risk of homelessness. Ultimately, UCLA determined that high utilizers and SMI/SUD were the primary target populations. Pilot Reporting of Target Populations by Enrollee While Solano reported on all DHCS-designated target populations, the pilot target population of the pilot included only the high utilizer and SMI/SUD populations (Exhibit 237). Solano captured the additional target populations due to the information already being collected for reporting purposes. Appendix |: Pilot Primary Target Populations and Reporting | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research err) Health Economics and Evaluation Research Program Exhibit 237: Solano WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD | Homeless | Homelessness | Involved COVID-19 Individual-level | X x x x x x x Target Populations Reporting Pilot’s Primary | X x Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Sonoma’s Target Populations Description from Application In their application, the County of Sonoma Department of Health Services Behavioral Health Division indicated that their target population has been individuals who are homeless or at-risk- of-homelessness who also have a serious mental illness and at least one of the following: e ~=Co-occurring health conditions including substance use disorders « High users of emergency services e Served by multiple agencies In addition, the enrollee needed to be eligible for Medi-Cal. They also indicated that they would focus on elderly individuals who are difficult to place since they often experience the longest waits for appropriate placement. Changes during WPC and Primary Target Population Determination Through UCLA conducted interviews, Sonoma County indicated that their target population had changed from their initial application. In particular, individuals did not need to have a severe, persistent mental illness and Sonoma also worked with individuals with high/moderate mental health conditions. Additionally, included individuals could be high utilizers of mental health or medical emergency room services. UCLA determined the primary target populations as SMI/SUD, homeless and at-risk-of-homelessness. Whole Person Care Final Evaluation Report | Appendix |: Pilot Primary Target Populations and Reporting err UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Pilot Reporting of Target Populations by Enrollee While Sonoma County did report on all but two of the target populations designated by DHCS (no justice-involved or COVID-19 reported), the specifically targeted populations of the Pilot were the SMI/SUD, homeless and at-risk of homelessness populations (Exhibit 238). Exhibit 238: Sonoma WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD_| Homeless | Homelessness | Involved COVID-19 Individual-level | X x x x x Target Populations Reporting Pilot’s Primary x xX x Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Ventura’s Target Populations Description from Application In their application, Ventura County Health Care Agency identified their target population as adult (ages 18 or older) high utilizers with at least four ED visits and/or two inpatient visits. Furthermore, the Pilot prioritized individuals who are homeless or at-risk of homelessness and/or with SUD or mental illness. All enrollees needed to be Medi-Cal eligible. Changes during WPC and Primary Target Population Determination Through UCLA conducted interviews, Ventura indicated that they went with a general target population in order to have the most flexibility. As a result, Ventura would be able to serve any high-need population including individuals with multiple chronic conditions, SMI/SUD, or currently experiencing homelessness. High utilizer was their primary target population. Pilot Reporting of Target Populations by Enrollee While the pilot aimed to provide services for individuals that met their high utilizer criteria, they reported on five target populations (Exhibit 239). The pilot used a four-point question to determine if an enrollee is homeless. Appendix |: Pilot Primary Target Populations and Reporting | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program HASTE Aa: Exhibit 239: Ventura WPC Pilot Target Populations Chronic High Physical At-risk of Justice- Utilizers Conditions | SMI/SUD | Homeless | Homelessness | Involved COVID-19 Individual-level | X x x x x Target Populations Reporting Pilot’s Primary x Target Populations Source: Whole Person Care Pilot Applications (n=25) 2016; PY 3 Follow-up Interviews with Lead Entities (LE) and Frontline Staff (n=27), September 2018-March 2019; Pilot specific case studies review in February-April 2022; and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6. Whole Person Care Final Evaluation Report | Appendix |: Pilot Primary Target Populations and Reporting Moday uoenjenq jeul4 a4eD UOSIad BJOYM | SYSd JdM JO Sa|dwWex; aAije.jsnj|| paysaIas :f xIpuaddy Ea ysnouyy ‘xapay Jsajlue|] YZNosU) syodai payleyap PaAladaJ S1OJEUIPIOOD BJD IdM ‘SquedIoIWed WweJso1d JM JO S1OJLUIPIOO B1eD DdM 0} asn juawedap Asuasiawa jo UoleoyI}OU AW} |ead aplAoid pjnom yey} Wes3oud e UO paxJOM apIsJaAly 900'T SUSIA JuaWedag Anuasiaw rae) Asojeinquiy apissanry “asn GJ 0} saaeusaye puly pue ‘Sddd 41843 YIM saajjouua azisel!Wey ‘saunseaw aAljejUaAed aseaJoUl 0} SEM UO!}UA}U! AUL ‘SWaysAs adUeJSISSe |eEJOIAeYaq pue ‘|edI|pa “JeI90S ajeSIAeu Way} djay 0} ydeOd ssauljam & 0} paudisse sem aajjojua yoea aiayM WesZ01d e paudisap Aayl uead yoea %g Aq saa|joiua DdM Aq susia (Q3) JUaWyedagq AdUaSJaWy Jo asn ay} adnpai 0} sem [e083 s,oUulD0pUua|\| See'T aseg Ajoqeinquiy oulsopuay| “ueIDIUI[D YEAH |esolAeyag pasuaciy pue (WIJ) uela|uysa | jeoipal,] Asuasuawy ue YIM payers a4aM SHUN [1D ‘Wodsues} adue;nqwe guizij!jn {NOYIUM sadIAJas apPIAOsd 0} SEM | VD JO |eO3 DY ‘saluaZJawWa YYeay jesoIMeyaq/Yyeay jejuaW 4OJ sjeiuajas pue ‘uoljeyodsues} “uawaseueW ‘sjualussasse Yyjeay je}UaW pue |edIpaw :3ulpnjoul sadiAdas papiAoid yey} yun ajiqow e sem | 1WD *sdap|OYS}L}s J9YIO YIM UO!}eJOge}|O9 Ul AJUNOD epaweyy ul payeas9 Wes8oJd jojld e sem (11D) wea, wodsuel| pue juawssassy AjuNWWOD ay, c6L 30|!d wea} Wodsue pue juawssassy Ayunuiwo9 epawely ase) Asojejnquiy WSdd jo Asewuuins (skeq) yyauay 9aweNn WSdd 30|!d IdM adAy AloZaye) WSdd 9 Ad-t Ad ‘8d, Aso8aye9 Aq paqwugns sySdd IdM JO Sajdiuexg aaijesjsn||| PaIDa/aS :OvZ UG!Yxy SVSCd JdM $0 Sajdwexg aaizes}sny|| payaas :f xipuaddy WweJZo/d youeasay uo}enjenq pue s2}woUCs] yea yoseasay Arlo Yaa} 404 JaqUuaD YIN areola) Yau) SWSAd DdM JO Sajdwex] aniesysnj|| payajas :f xipuaddy| woday uolenjeng jeuly aseD UOSJd9q BJOUM, sulseus ejep 3ulojdxa Aq paysij|dwosse sem siyl “Ajjenuue %& Aq sasn Usiy ssajaUoy JO} SJAAIS B19 0} a]qe|lene suejd aied anisuayasdwiod 4S JO} suejd aseD aAlsuayasdwoD jo Jaquinu uejd a4e) JO Jaquinu ay} aseasou! 0} pawle ods!9ues4 UeS S8Z'T ay} aseasou| oosioues4 ues anisuayasduioD *suea} 31D ||e OF eyep jo Ayjiqissao0e paseasoul pue uoljeyUauNs0p ul Aduaysisuod pajeasd sly} ‘uolje]UaWINI0p JO} SjuawayNbas |je Suuaaod Aq uejd aed ayajdwod @ ayeaJ9 padjay |00} siy} Bulsp) “sauUewW jUAa}sIsUOD ssao0id @ Ul YIJeaY JO SJUeUILUJAJAp [2190s ay!| soyIoads quaw||olu3 ed} pue sauljapind ajeaJ9 0} Jayed JUaW|OJUa 0} |00} "pny Jo 4J84} Ul [00] JIPNe Ue payUawajdw! eve|D ejUeS L6T uonequawa)du] eue|9 ejues (WvA- 8VSQd) seseueW asep ewndgyje> eo} ae) “sdaQqUIBLU dM 40} WOddns | aaljyesadnsay JO UO!JeUIPJOOD BY} AAOJAUU! 0} Japso Ul sJadeUeW 0} pasajas ased eWI}dO|eD 0} SadIAJas ase aAl}esadnda4 salelyauag BUIAIIA1 SIBQUIAL gM Ul] 0} palule BBueIQ Oss‘T asuelo “Bulyodai pue sisAjeue eJep JO} WaysAs ay} asn Oo} Aq|ige ayy aseasoUl 0} SEM UO!UAJU! AY] ‘UO!eJ}SIU|LUpe WeIZOId pue suaSeuew ased jo spaau ay} JaaW 0} WJOy}e|d uoljeu!p4so09 aued asn-o}-Asea Ue Se GYVZIM d4u (auvZIM) uSIsapad 0} SEM We JIBY| ‘ASN 0} yjye}s 10y Asea jou wuose|d SEM ‘QUYZIM ‘WJ0j1e|d UOIJEUIPs1OOD aed s,ULLeIA| Sv | uolyeuIpsoo> aieg ue | uoneulpsoo) ase *Saa||O1Ua Jd M 104 paseasoul Xapayy sajIUe jo asn ‘suoijezwl| pue uojes1|dnp awios sem asay (skeq) adAy vSdd jo Asewuuns yysue7 owen WSGd 30[!d IdM Aioaye) WSdd 77Z0Z Jaqwiazeq WeJZ01g youeasay UOIeN|eAq pue s21WOUIJ Yea yoseasay Aotod YiJeaH 4OJ JajUaD VION Woday uolenjeng jeuly aseD UOSsiag BJOUM | SYSdd IdM JO Sajdwexy anljessn||| paloajas :f xIpuaddy Ea “aouejsisse Wess01d Bulsnoy 494} 0} JapIAQid Jed/U!|9 AY} 40 pamoyje ad10jsajes OU! payesouad suoday ‘aduejsisse Sulsnoy papseme aq 0} uosiad e JOJ S9x2} PY! Bw} Jo YBua| Sulpueyssapun ul JapiAosd quawdojanaq Jediulja ayy ple 0} JUe|aW seM SIy| “sawWO3}NO Sulyoes| Wesdold pue suolesijdde aouejsisse weisoid Sulsnoy aouejsissy yuawndop 0} spjalj eyep padojanap oyUawesdeS vl@ Sulsnoy oyuaweses “JO|!d ay} pue ‘SOHO “Ue9H JI|GNd Uaamjaq Suljeys e1ep Jo ssadoud ay} aU!|WeAI}s 0} SEM aAIWaIGo ay ‘auednsse Ayjenb ejep 104 sassazoid pue Suloday soja $]090}0Jd ysi|qejsa 0} pue sdiujawW BuljJodad SNHG SDH 404 e1eqg jo 40} pasn ezep jo Azljenb ay} aAcsdwi! 0} pawie eden 6£9 Ayyjeno aaoidwy] eden ejeg (7 a]9A9) *saa||O1Ua DgM || 0} painqijysip pue payeasa — (uolssaidaq) 319M SJepuUaled Sa!PAI}Ie AjyUO| ‘(UOISsasdap) $3J09S GOHd $9109S §-OHd AAOJdWI! 0} s8ulJaaW Yoddns Ajyaam 3aNpai 0} sjual|> Ysnosy} ayt| JO Aqjenb |jesaAo ,Saa|josua BAodsduu! OdM YUM sdnois 0} pafod e jUaWA|dUW! 0} pawuie ojlUag UeS 60S yoddns Ajpjaay Ajunod jews “wiea} aed ailjua ayy Aq ajqissaqze (spaeu sulsnoy pue ‘asn aduejsqns ‘spaau yyeay jeaisAyud pue jequaw BSulpnjoul “a‘!) uejd eased sued ase) dAISUdYaIGWOD e YUM sSaleidyauagq suljedioiwed (1Ma) sseuljaM Jo uolyodoud ay} aseasou! 0} pawle oayeyj ues 00s‘T 0} saspig oajyey) ues “uolejuauinsop dy} Bas UeD Wea} Bed as!]Ua ay} JeYr uolj}eulpsoos ased os uejd a1e9 paseys 34} 3403s 0} a4aYyM INO Buns JO |@Aa| YZIy pasu ay!| Sualeq pajuasaid sey Sad SIUL “Ss om sjenpiaipul uo}UaAJa}U! SU}JeU|P41009 a4ed pue sa13ojoUYyIE} ssajawoH (skeq) adAy vSdd jo Asewuuns yysue7 owen WSGd 30[!d IdM Aioaye) WSdd WweJZ0/d youeasay uoNen|enq pue so}WwoOU0s] yea} yaueasay Ad!|Od Yea} 4104 4a}UaD VION 7207 Jaquiaseq bial SVSCd IdM JO Sajdwiexg anizessny payajas :f xipuaddy| qoday uonenjeng jeuly asep UosJad BJOUM uoljejndod ayewul ay} 0} JdM asiaape 0} 291JJO Jyluays s,AjUNOD Uday YIIM 440M 0} Sem JUaIU! YL “QdM OJ! SjUAI|D payesjaoseoUl-ysod Bululejas pue Bul||o1ua 0} sJalseq ajqissod Ayluap! 0} pawie Way uoljuajad pue juaw|joiua TSS‘T | payesaoseoul-ysog usa 4810 “saa||olua IdM 404 sweay e4]Ua9 pue ‘pjaly WuaWwasesua dM Aq uoljeulpsood aed anisuaju! ‘Ayjenb ysiy ysnosyy Sem uolzezI|13N Jual}edul Sulsnpai Joy Ayjepouw Agewid ay ‘uolzezI]13N JUal}edul Ul UOIJONpa|al %S e Jo WnWiulW e aAalyse 0} pawie AjuNOZ esnjua/ 6L7'T einquap, *syafoid JuaWaAosduu! Ayjenb pue ‘Buljiodas wei3oud ‘sasodind Sulsoy1uoW snonul}Uu0d 404 (HID) YEAH JJ aduell|y e/UJOJI]eD [e4jUaD pue (3IHOS) @BueyoxZ Uo!IyeWOJU] Yay ZNuD eyUeS ay) Woy eyep Uol}ezI|IIN Gq pue jeydsoy avlao—ad 0} ssad0/d d1yeWaysAs e paysi|qejsa znD e]UeS ss9ZiN, YBIH 404 3uleYs ejeq uolzez1|!3N 7LE'T znd e]Uues “sjsanbau uoljeyodsuel} Aep awes 0} puodsad 0} (Ly) Wsued) JeuolZay eaiy SBuly YM SMOLJOM aAI}esJOge||Od ajeaJ9 0} SEM aAIalqo ay “aUeApe UI ||e9 0} paau 3y} }NOUYUM ‘sjusWUlOdde UO!jeWOdsUeI} 0} UO!}EUIPJOOD JaJIP YIM ssaussa|auoY Buldualadxa sjenpiAlpul AapIAOJd By} 0} pawile sBuly (Law) ysued] jeuoIZay Z8T Realy ssuly s3uly, uo1eZ1)13N quanedu *ssauyied Aq waysAs ay} Jo Nn ay) Ysnosy) payenjena sem ainyonsysesjul AS|IBMPaUUOD 94} JO ssauUaAIa}jo ay ‘sAep ssauisng om} UUM WaysAs Qs||a\}2aUUOD, dy} OJ! pasJajua aq pjnom sJiajuNnooua Ba]]O1Ua Jo %G¥ sea] Je Jey) [eos e Jas OBaIq ues uolez! ejeq OT6'T | —_ asiiempeuu0D o8aig ues WSdd 40 Asewuuins (shea) yy3ue] eweNn WSdd 201d IdM adAl Aioaye) WSdd 77Z0Z Jaqwiazeq WeJZ01g youeasay UOIeN|eAq pue s21WOUIJ Yea yoseasay Aotod YiJeaH 4OJ JajUaD VION woday uoenjenq jeuly a4e> UOsuad BJOYM | SYSCd IdM JO Sa|dwWex; aAije.jsni|| paya]as :f xIpueddy Ea *($z=u) suoday ysdd |enuuy 9 Jea, Wes801g pue jenuuy ¢ JeaA WeISOsd ‘JEaA-PIW ¢ JeaA WeIZOJd ‘|eNUUY 7 JeaA WeIZOId “IEdA-plIA| 7 JEBA WeIZOIg :291N0S ‘JdM aAosduu! 0} Japso Ul sanss! jeuo!}esado Y8NO1Yy 4JOM pue Jaa O} sauyed dM Ady 10} pauSisap a1am Jey) s8uljaaW suoljzesadQ suoljesado pue Suluueyd AjyjuoW pjay OuR|OS TSS‘T pue Suluue|d ouejos “ssad0Jd Siu} Jo Wed jesZajUl Ue SEM qujlogaseys ‘ydeadjNo pue uoNednpa AyuNWWOD Sulppe pue ssadoud jesajas ay} Bulzije4jUued wa3shs jeuajoy pue Suluijad Aq saysAs jeJajad 1184} JO UOISIAAL pue ela pue SulojiuowW 3ulosu0 pajuawajduui eyseys 88L'T eyseus “uo Ayea 4421S Jd Udy pue SJapUasJO uaamjag diysuolejas Buljsn4} e Susjeas9 Wels pue (skeq) adAy vSdd jo Asewuuns yysue7 aweNn WSGd 30]!d IdM Aioaye) WSdd WeJZo/d youeasay uoNenjenq pue so}WwoU0s] Yea yaueasay Adl|Od yIe3H JO Ja}UaD VION 7207 4aquiaseq UCLA CENTER FOR HEALTH POLICY RESEARCH HEALTH ECONOMICS AND EVALUATION RESEARCH Appendix K: Policy Brief Care Coordination Framework UCLA CENTER FOR ®:, HEALTH POLICY RESEARCH .6; e568) it Whole Person Care Final Evaluation Report | Appendix K: Policy Brief Care Coordination Framework UCLA CENTER FOR HEALTH POLICY RESEARCH Delivery of integrated services may improve the patient experience and reduce health care use and costs, Health Policy Brief October 2019 Whole Person Care Improves Care Coordination for Many Californians Emmeline Chuang, PhD, Brenna O’Masta, MPH, Elaine M. Albertson, MPH, Leigh Ann Haley, MPP, Connie Lu, MPH, Nadereh Pourat, PhD SUMMARY: California’s Whole Person Care (WPC) Pilots implemented under the Section 1115 Medicaid Waiver, “Medi-Cal 2020,” are designed to coordinate medical, behavioral, and social services to improve the health and well-being of Medicaid beneficiaries with complex needs. We examined literature on care coordination and developed a framework for assessing the progress of WPC Pilot implementation in eight key areas. Three years into the program, results show that WPC Pilots successfully implemented many essential care coordination processes, but they continued to further develop needed infrastructure. These findings highlight opportunities and challenges in implementing a cross-sector care coordination program for patients with complex health and social needs. T he U.S. health care delivery system has long been fraught with inefficiencies rooted in part in fragmentation of care and professional silos. Frequently, patients with chronic and complex needs must navigate between medical, behavioral health, and social service providers who are not prepared or equipped to provide them with holistic care. Preliminary evidence suggests that delivery of integrated services may improve the patient experience and reduce health care use and costs.!3 In 2016, California began implementing the WPC Pilot demonstration project to promote systematic delivery of coordinated care and evaluate its impact on health care costs and use for Medicaid (called Medi-Cal in California) beneficiaries.*> The WPC Pilot is part of California’s Section 1115 Medicaid waiver, known as “Medi-Cal 2020.” The aim of WPC is to improve coordination of medical, behavioral health, and social services for patients who use a high level of Medi-Cal services and ultimately improve patient health and reduce Medi-Cal expenditures. A total of 25 pilot programs in 26 selected counties* (hereafter referred to as WPC Pilots) were established by 2017. All WPC Pilots were led by a single, designated lead entity (LE), typically a county Health and Human Services Agency. These LEs partnered with health plans and other service providers to coordinate medical, behavioral health, and social services for targeted Medi-Cal beneficiaries. Specifically, WPC Pilots were expected to systematically identify target populations, share data, coordinate care, and evaluate improvements in the health of enrolled populations. a Twenty-seven counties initially implemented WPC Pilots, but Plumas County (part of the Small County WPC Collaborative with Mariposa and San Benito Counties) dropped out in September 2018. 2: Wer Vessel aa ele d cz. er) Effective cross- sector care coordination requires timely sharing of information among the care coordination team and providers. Acknowledging heterogeneity in how publicly funded services are structured and. delivered across California, WPC Pilots had considerable flexibility in the selection of target populations, outreach methods, services provided, and outcomes tracked. WPC Pilots also differed significantly in the amount of WPC funds requested and allocated to develop infrastructure for care coordination.° Information on specific characteristics of each WPC Pilot is provided in Appendix 1: https://healthpolicy.ucla. edu/publications/Documents/P DF/2019/wpc- appendix-datatable. pdf. What is Care Coordination? The Agency for Healthcare Research & Quality (AHRQ) defines care coordination as “deliberately organizing patient care activities and sharing information among all of the participants concerned with a patient’s care to achieve safer and more effective care.” Care coordination is distinct from care management, which is more focused on management of chronic medical and psychosocial conditions, and from case management, which includes services that help patients develop skills to access services and meet their basic needs.” We drew on elements of care coordination identified by AHRQ and an extensive review of the literature to develop a framework of elements critical for cross-sector care coordination. We then used this framework to assess care coordination under WPC. Cross-Sector Care Coordination Framework Cross-sector care coordination requires availability of infrastructure to support delivery of effective care coordination processes (Exhibit 1). Care coordination infrastructure elements include (1) care coordination staffing that meets patient needs, (2) data sharing capabilities to support care coordination, (3) standardized organizational protocols to support care coordination, and (4) financial incentives to promote cross-sector care coordination. Care coordination staffing that meets patient needs. To successfully coordinate care across sectors, staff must have sufficient capacity to effectively engage with patients to address a wide range of medical, behavioral, and social needs. Staffing levels appropriate for meeting patient needs include (1) developing a multidisciplinary team with relevant and diverse clinical expertise, (2) inclusion of peers with lived experience to build trust and promote compliance of complex patients, and (3) staff workload that ensures sufficient availability to meet patient needs.!°'? Data sharing capabilities to support care coordination. Effective cross-sector care coordination requires timely sharing of information among the care coordination team and providers. Data sharing infrastructure that facilitates this type of information exchange includes (1) formal agreements that define terms and conditions of data sharing with key partners; (2) a universal consent form to reduce barriers to sharing patient data; (3) use of an electronic data sharing platform that includes key information such as comprehensive care plans; (4) medical, behavioral health, and social service use data; and (5) capacity to track and report care coordination activities. Ideally, care coordinators can also access this data sharing system to (6) view and enter data (7) remotely (i.e., in the field) and (8) in real- time. > Standardized organizational protocols to support care coordination. Standardized protocols help minimize undesirable variation in delivery of care coordination services.'® These include protocols for (1) referring patients to needed medical, behavioral, and social services; and (2) monitoring receipt of services and tracking patient outcomes. Financial incentives to promote cross-sector care coordination. Financial incentives can facilitate organizational buy-in and accountability for cross-sector care coordination.*'7 Financial incentives that help align organizational priorities with these care coordination goals CLA CENTER FOR HEALTH POLICY RESEARCH Exhibit 1 Conceptual Framework of Cross-Sector Care Coordination fw Conduct needs assessments and develop comprehensive care plans Ti oe Cross-sector care coordination is built from the ground up, starting with a strong infrastructure that supports the care coordination team as they carry out care coordination processes. Actively link Ensure frequent patients to communication and needed services follow-up to across sectors engage enrollees juntability within the care coordination team Process Elements PR icra at ema uric} include use of payment mechanisms that (1) are risk-stratified and address financial risk assumed by providers and (2) reward better performance via incentive payments. Care coordination process elements include (1) ensuring frequent communication and follow-up to engage enrollees, (2) conducting needs assessments and developing comprehensive care plans, (3) linking patients to needed services and follow-up to ensure receipt of services, and (4) following protocols to promote accountability among care coordination teams. Ensure frequent communication and follow-up to engage patients. Effectively engaging complex patients in care coordination requires the adoption of patient-centered communication strategies. These include outreach or other contact with patients (1) in-person, at least initially, to build trust and engagement; (2) wherever and whenever they can be found, including in the field; and (3) frequent follow-up, i.e., more than once per month.!* Conduct needs assessments and develop comprehensive care plans. Full assessment of patient medical, behavioral, and social needs is essential to developing a comprehensive care plan. These care plans identify patient goals, the actions needed to achieve these goals, and resources or supports needed to ensure successful delivery of care.!#1519 Patients should have a single care plan shared across all providers that is updated regularly UCLA CENTER FOR HEALTH POLICY RESEARCH Exhibit 2 Care Coordination Infrastructure in WPC Pilots eee de ur Pag Cee ee aed er nec) Care coordination staffing that meets patient needs eye Neuer) Wen it eens Cee Perec RET) San Bernardino Pr Pare eer ne Pena eka bereits e =. = zl= Multidisciplinary care coordination 2 = = = = 8lS\a _ team composition* zl=z 2 =/3/5/8/35 & Sle Sle S1se]e s/=/ [2] /=/B/e]elele 2 zie] |sle z(S\g 216 Ss z/flSlzls @)= ele =\|= S|=/s =| U = z/¢ S | Se & zl/z=/S\5 e/2/S 2) 2|2|= =|¢la|elz/s =| SlZ|Z/=|Z2/2 a|Zl/2/z r|/Sle|a/2/4)=|/8/a]/e]a)/l,)/ 6/2 clZzlelzl|2/2 2 z/G 2/2/2/£/2]=]212/2|8|2/3121z| (2/822 |Z ele lelelale 5/2|2/8/8/2|2/2|2/2/2/2|2/2/2/E/2/5/2/5| 2 | 8/2/88 |2 Use of workers with lived experience | , | , wll. . wtelel. dl. wl. |. wt... ele le] 2 g 58 workiona** al8|Flelsislale g| |Slelol]_lals|2l/a/a]. is Be 2lelflslaislalalelelsislelalgials(slaisislelsisiale| so Data sharing capabilities to support care coordination Data sharing agreements among key a partners eit m 9 a "1 4 a | | | o| 438 blelSlelelelal& FlslBls/5/5 5/5/56 gle E/25s AlzlAlal alae} aelal ae] Al al Ala] ala ALAA . . ol-[-| 16 Standardized protocols for monitoring wlelefede élale ‘ wleJel. . -lel.| ov and follow-up Financial incentives to promote cross-sector care coordination Risk-stratified PMPM bundles* ole ole . . . . . «| 10 Contracted care coordination services : (aly ele) fe] fel fe elel felel fel Jelelelel |elel aes -|8/5/_/€/ |] s/f ~|8/5/-] 8) 8]_|€]/_] €] €] €] el] €] €| es z|2|2|2|a8| =| 2/28 Z| 2/2) 2/2] 2/ =| 8] <| a] a] a] 8] =| 8] a) <82 Financial incentives for contractors” | | _ | _ wlelele de _J_}.y} ef} e]. wlelelel. wlel ow + Pilots were identified as having risk-stratified PMPM bundles +f Financial incentives for contactors were assessed only when when enrollees were stratified into different PMPM bundles at care coordination services were contracted out rather than intake based on an assessment of risk. provided directly by the lead entity. to address changes in patient needs over time, and case conferences with care coordinators Ca re i.e., more frequently than once per year. or care teams to share expertise, negotiate coor din ation is differences in judgment, and define priorities : Actively link patients to needed services across for patient care.” most effective when sectors. Active referral strategies, e.g., through accountability directly arranging services on the patient’s behalf, are more effective in service uptake than informational referral strategies, such as giving patients information about available treatment options and leaving them to navigate the rest.' Successful care coordination includes active referral to needed medical and behavioral health, including mental health or substance abuse treatment, and social services such as housing or benefits assistance. Promote accountability within the care coordination team. Care coordination is most effective when accountability for different activities is clearly defined and monitored. Strategies that support accountability for care coordination could include regular meetings Evaluation of Care Coordination under WPC Data for the evaluation of care coordination under WPC was gathered between September 2018 to March 2019 using WPC applications, a structured survey, and follow up interviews with leaders, care coordinators, and other WPC Pilot staff. Additional details about care coordination efforts of individual WPC Pilots can be found here: Attps://healthpolicy.ucla.edu/publications/search/ pages/detail.aspx? PubID=1844, for different activities is clearly defined and monitored. Infrastructure WPC Pilots reported significant progress in establishing the infrastructure needed to coordinate the care of enrollees in the first 3 years of implementation (Exhibit 2). b See Data and Methodology section, 6 Over half of WPC Pilots reported successfully sharing comprehensive medical, behavioral health, and social services data with partners. Wal West ice ara an ena a a.\ 0] Pilots differed, however, in infrastructure investments, data sharing, and other infrastructure in place prior to WPC. Care coordination staffing that meets patient needs, Staffing varied across and within WPC Pilots based on target population(s) and identified needs. Care coordination services were often provided by non-clinical staff such as community health workers. Due to the complexity of enrollee care needs, however, all care coordination teams included at least some staff with clinical expertise (e.g., providers, nurses, social workers). Many WPC Pilots also used peers with lived experience (e.g., previously incarcerated or homeless peers) to help build trust and rapport with enrollees. Staff workload varied considerably across WPC Pilots depending on projected acuity of the target population and intensity of contact with enrollees. Data sharing capabilities to support care coordination. WPC Pilots were required to develop new data sharing capabilities. By 2018, all 25 WPC Pilots had at least some formal data sharing agreements with key partners. Many had developed universal consent forms for sharing patient data, and nearly all used an electronic data sharing platform that included information on comprehensive care plans. WPC Pilots that did not yet have these capabilities reported challenges such as vendor delays and difficulty obtaining partner buy-in. Yet they typically had temporary solutions to facilitate data sharing (e.g., ShareFile, SharePoint, Box) until more efficient and permanent systems could be procured or implemented. Over half of WPC Pilots reported successfully sharing comprehensive medical, behavioral health, and social services data with partners. Pilots that did not yet share behavioral health data typically identified federal confidentiality laws protecting the privacy of substance use disorder patient records (42 CFR Part 2) as a major barrier. Less than half of WPC Pilots reported providing frontline staff with real- time notifications about patient events, such as emergency department visits, but most WPC Pilots without this capability identified developing real-time notifications as a future priority. Standardized organizational protocols to support care coordination. Around half of WPC Pilots had standardized protocols in place for referring enrollees to needed services (e.g., checklists) and tracking or following up with enrollees to assess referral outcomes. Several WPC Pilots cited the heterogeneity of enrollee service needs as a barrier to developing standardized referral protocols, particularly when referral processes were not integrated with an existing electronic platform to facilitate tracking. Pilots that contracted out care coordination services to multiple partners also cited partner preferences for developing and maintaining their own internal protocols as a barrier to standardization. Financial incentives to promote cross-sector care coordination. Pilots were primarily reimbursed for care coordination under WPC using per- member, per-month (PMPM) payments for a bundle of services, though some received fee-for-service reimbursement to deliver additional services (e.g., outreach and engagement, assessments and screening). Eleven WPC Pilots stratified their PMPM bundles based on enrollee acuity or risk and tailored service intensity. The majority contracted with one or more external organizations (e.g., local health clinics or private social services providers) to supply some or all of their care coordination services. Of these, over half included financial incentives in contracts linked to the achievement of specific outcomes aligned with WPC goals (c.g., improving quality of documentation or scheduling a follow-up primary care visit within 7 days of hospital discharge). Care Coordination Processes WPC Pilots also reported significant progress in implementing key processes necessary UCLA CENTER FOR HEALTH POLICY RESEARCH Care Coordination Processes in WPC Pilots eT er ete ug Ceres c) eer U cd ren ate WET roc} Went eit Poe reed ec) eeu) ee rece} rer ea red eer ic] peeked ecru) Ensure frequent communication and follow-up to engage patients Enrolleecontactmorethanonceper |.) ./.}././.].j.iei.felele]e)e/elelelelele le le le le lel 26 month Field-based outreach olel-]eTel-[-[-]-]-[-]-]-]-]-]-]-]-][-][-]-]-]e]- [ee] 26 Frequent in-person, on-going of. efeflefefeleledade efelefele]ede ofe]eje de 2B communication with enrollees Conduct needs assessment and develop comprehensive care plan Needs assessment more than . wlele . wl. wlele sled. . . .| 46 once per year Single shared care plan ~[. ele[-]-[- e[-]-]-[-][- ~ |e . le]. ]- 7 2 Actively link patients to needed services across sectors Active referral to medical care olelelelele]e]elele]ele] el ]e]e]ele]ele]e lel lele >] 2 Active referral to behavioral slelelelelelelelelolelalelelalelelslslelelelelelelel a health care Active referral to social services efefedteletededteleleledetetedele ft etelele ted ede dele fe] 2 Promote accountability within the care coordination team Regular meetings with team to . ele ele . efelelelele . ele . ele ele ele ele . 25 promote accountability Data Source: WPC applications, mid-year and annual narrative reports submitted by WPC Pilots to the California Department of Health Care Services, interviews for effective cross-sector care coordination (Exhibit 3). Their specific approach to these processes varied largely due to their WPC Pilot’s target populations and the level of intensity of services they aimed to provide. Ensure frequent communication and follow-up to engage patients. Many WPC Pilots required care coordinators to contact enrollees at least once per month. However, care coordinators in nearly all WPC Pilots reported contacting enrollees more frequently based on patient need. Most also reported using and prioritizing in-person outreach in the field rather than contacting enrollees by telephone. WPC Pilots described field- based outreach as particularly important for identifying and engaging homeless enrollees. conducted with representatives of each Pilot from September 2018 to March 2019, and surveys of WPC organizations administered in the summer and fall of 2018. Field-based outreach was Assess patient needs and develop a comprehensive care plan. WPC Pilots were required to assess enrollee needs and develop . ] 1 a comprehensive care plan within 30 days of P articular ‘iy enrollment in WPC and, when appropriate, Important for to repeat this process at least once per year. In practice, most WPC Pilots required care coordinators to re-assess enrollee needs and update care plans more frequently. To assist with accurate identification of needs, many WPC Pilots reported the use of validated instruments such as the Vulnerability Index —Service Prioritization Decision Assistance Tool and the Patient Health Questionnaire-9. identifying and engaging homeless patients. Actively link patients to needed services across sectors. All WPC Pilots reported use of active referral processes such as accompanying enrollees to appointments or facilitating LA CENTER FOR HEALTH POLICY RESEARCH Continued investment in data sharing capabilities, staff training, and other infrastructure are needed to support effective cross-sector care coordination. warm hand-offs to medical, behavioral health, and social service providers. WPC Pilots reported perceived benefits of active referral to include the ability to ensure enrollees received important services, provide immediate follow-up after service receipt, and create additional opportunities for care coordinators to interact with enrollees and monitor enrollee needs and progress. Among WPC Pilots without standardized protocols for referral tracking and follow-up, active referral strategies were viewed as critical for helping informally “close the loop” on referrals. Promote accountability within the care coordination team. WPC Pilots were required to identify providers and staff responsible for care coordination. Almost all WPC Pilots reported use of regular team mectings to keep one another informed of enrollee progress and promote accountability for care coordination activities. A number of WPC Pilots also reported regular case conferences or other opportunities to share challenges and brainstorm potential solutions. Accountability was generally described as more challenging in WPC Pilots where responsibility for care coordination was distributed across many partners. In these WPC Pilots, challenges included lack of consistency in care coordination activities, the potential for enrollees to have multiple designated care coordinators across different organizations, and a greater need for careful communication during hand-offs across organizations. Future Steps Our interim examination showed many WPC Pilots made significant progress in building needed infrastructure and delivering cross- sector care coordination services. By mid- 2018, many WPC Pilots had successfully hired care coordinators, shared data across sectors despite multiple challenges, created standardized protocols to support care coordination activities, and built financial incentives for performance into contracts with providers. Many WPC Pilots also established care processes to engage enrollees in care, developed comprehensive care plans, actively linked patients to needed services, and promoted accountability among care coordination teams. All Pilots described WPC as an important opportunity to improve cross-sector relationships and build more effective systems of care within their communities. The implementation of WPC included significant and numerous challenges. Pilots acknowledged the need for further progress in multiple areas to achieve overarching WPC goals of better care, better health, and better efficiency. Our analyses identified specific strategies to address these challenges: Invest more time to further develop the infrastructure to support cross-sector care coordination. Many WPC Pilots had limited or no cross-sector data sharing capabilities prior to WPC. Pilots that successfully created this infrastructure reported investing a significant amount of time, typically more than originally anticipated, to accomplish their goals within the first few years of implementation. Universal consent forms facilitate information sharing, but WPC Pilots noted the need to plan significant time for review by legal counsel in different organizations. WPC Pilots located in counties in which the majority of services were contracted out to private agencies emphasized the importance of allocating sufficient time to ensure partner buy-in and to align financial incentives within contracts with WPC goals. All WPC Pilots reported the importance of continued investment in data sharing capabilities, staff training, and other infrastructure needed to support effective cross-sector care coordination, even mid-implementation. LA CENTER FOR HEALTH POLICY RESEARCH Promote person-centered practices that more effectively engage vulnerable patients in care. Pilots recognized the need for patient- centered outreach, communication, and referral strategies to engage enrollees in WPC services. Successful strategies reported by WPC Pilots to help foster enrollee self- efficacy included using case management in addition to care coordination to more effectively serve enrollees, the hiring of clinical staff that were only funded part- time by WPC to allow for direct provision of services as part of initial outreach and engagement efforts, and providing benefits assistance to help reduce Medi-Cal churn. All Pilots also reported ongoing adjustment of WPC programs (e.g., by reducing care coordinator caseloads or clarifying scope of work) to better meet enrollee needs. Leverage WPC resources and partnerships to help address structural problems outside of WPC Pilots’ control. Multiple WPC Pilots cited limited availability of long-term, permanent housing as a barrier. Similarly, several small and rural counties cited difficulties with recruitment and retention of staff and limited availability of private behavioral health providers accepting Medi- Cal as barriers to timely access to behavioral health services. Strategies used by some WPC Pilots to address this issue included leveraging WPC to ensure expedited access or priority placement for their enrollees and developing innovative partnerships to improve availability of services within the community, e.g., working with private homeowners to place people in new types of housing. Author Information Emmeline Chuang, PhD, is an associate professor of Health Policy and Management at the UCLA Fielding School of Public Health. Elaine M. Albertson, MPH, is a doctoral student in Health Policy and Management at the UCLA Fielding School of Public Health. Connie Lu is a project manager and research analyst at the UCLA Center for Health Policy Research. Leigh Ann Haley is a project manager and research analyst at the UCLA Center for Health Policy Research. Brenna O’Masta, MPH, is a project manager and research analyst at the UCLA Center for Health Policy Research. Nadereh Pourat, PhD, is associate director of the UCLA Center for Health Policy Research, director of the Center’s Health Economics and Evaluation Research Program, professor of Health Policy and Management at the UCLA Fielding School of Public Health, and professor at the UCLA School of Dentistry. Acknowledgments Funding for this project was provided by the California Department of Health Care Services. The authors thank Denisse Huerta and Kimberly de Dios for their assistance in developing case studies of WPC Pilots. We also thank WPC Pilot organizations for their time and effort spent completing reports, interviews, and surveys and reviewing our findings. Data and Methodology UCLA developed the care coordination framework following a systematic review of the literature on cros ctor care coordination. Screening of 1,694 articles identified 27 articles addressing interventions to coordinate health and social services for high-use patient populations. These articles were evaluated for key themes and trends and directly informed the conceptual framework used in this report. Qualitative data sources used to assess WPC Pilot care coordination activities included WPC applications, mid-year and annual narrative reports submitted by WPC Pilots to the California Department of Health Care Services, semi-structured interviews conducted with key informants from each Pilot between September 2018 to March 2019 (n=27), and web-based surveys administered from July 2018 to October 2018 to key program staff in WPC Pilot Lead Entities (n=27) and Partners (n=227). UCLA coded reports and interviews for themes by multiple coders to ensure validity. Analysis were completed using NVivo 12.0 software. Analysis of survey data was completed using Excel and Stata 13.1. Suggested Citation Chuang E, O’Masta B, Albertson EM, Haley LA, LuC, Pourat N. 2019. Whole Person Care Improves Care Coordination for Many Californians. Los Angeles, CA: UCLA Center for Health Policy Research. UCLA CENTER FOR HEALTH POLICY RESEARCH HEALTH ECONOMICS AND EVALUATION RESEARCH Appendix L: Policy Brief — A Snapshot of California’s Whole Person Care Pilot Program UCLA CENTER FOR &, HEALTH POLICY RESEARCH .o; Whole Person Care Final Evaluation Report | Appendix L: Policy Brief — A Snapshot of California’s Whole JZZ&} Person Care Pilot Program UCLA CENTER FOR HEALTH POLICY RESEARCH. «, © if The Whole Person Care Pilot program coordinates medical, behavioral, and social services to improve the health and well-being of Medi-Cal beneficiaries with complex needs. SIN L Mee Ved Health Policy Brief May 2021 A Snapshot of California’s Whole Person Care Pilot Program: Implementation Strategies and Enrollees Nadereh Pourat, Brenna O’Masta, Leigh Ann Haley, and Emmeline Chuang SUMMARY: The Whole Person Care (WPC) Pilot program implemented under California's Section 1115 Medicaid Waiver, "Medi-Cal 2020,” coordinates medical, behavioral, and social services to improve the health and well- being of Medi-Cal beneficiaries with complex needs. In this policy brief, we analyze data from the interim statewide evaluation of WPC to present a snapshot of the 25 participating pilots, based on key implementation strategies and enrollee characteristics. The data can be used by organizations that are developing population health management programs for high-need, high-risk Medi-Cal beneficiaries under the California Advancing and Innovating Medi-Cal (CalAIM) initiative, as well as by other programs providing care to low-income patients. small proportion of the insured population is responsible for a relatively large proportion of the health services used in the United States.’ Many of these individuals have complex medical, behavioral health, and social needs that require an integrated approach to care.’ In 2016, the California Department of Health Care Services (DHCS) began a demonstration program called Whole Person Care (WPC) to promote the integrated delivery of care for Medi-Cal beneficiaries who use acute and costly services in multiple care areas. Under WPC, eligible beneficiaries receive care coordination and other services not traditionally covered by Medi-Cal to address medical, behavioral health, and social needs, with the aim of improving their health outcomes and overall well-being. In 2017, 25 WPC pilots in 26 counties began enrolling eligible Medi-Cal beneficiaries. Pilots had flexibility in the specific target populations served and in how WPC was implemented.’ WPC was originally scheduled to end in December 2020 but was extended for a year due to the COVID-19 pandemic. Some of the services provided under WPC will be incorporated into CalAIM, a multiyear initiative planned by DHCS that is designed to use WPC approaches to improve beneficiaries’ health outcomes. Under CalAIM, Medi-Cal managed care plans are expected to provide Enhanced Care Management (ECM) and In Lieu of Services (ILOS) through contracts with community-based providers, including organizations participating in WPC.‘ CalAIM is expected to begin implementation in January 2022. This policy brief provides a snapshot of each pilot’s implementation strategies and enrollee characteristics to inform CalAIM transition planning. Data are drawn from the statewide evaluation of WPC conducted by the UCLA Center for Health Policy Research.*® The data indicate the importance of tailoring future efforts to the unique needs of various subgroups of Medi-Cal enrollees. WPC Program Implementation Strategies Exhibit 1 provides insight into similarities and differences by county across pilots in the target populations served, strategies used to identify and enroll eligible beneficiaries, care coordination approaches, other WPC services offered, and engagement of social service providers as partners. For example, data show that 16 pilots provided services to more than one target population, and 16 used street- or shelter-based outreach to identify eligible enrollees. Thirteen pilots used a single dedicated care coordinator to follow enrollees across all WPC-participating care settings, and 17 used co-located staff from different service sectors to facilitate access to care. Care coordinators’ caseloads varied significantly across pilots (from 10 to 300), reflecting differing levels of enrollee need and intensity of services provided. Highlighting the importance of housing support to enrollees, 12 pilots offered tenancy support, landlord incentives, and funds to support housing placement. Many provided medical respite (18) and sobering centers (14). WPC Enrollee Characteristics Exhibit 2 provides insight into the WPC enrollee profile by county, including enrollment information, the demographics and health status of enrollees, and the utilization of services by these individuals prior to WPC enrollment. Pilots differed in multiple elements, such as the number of enrollees served (from fewer than 300 to more than 10,000); average length of enrollment (3-17 months); inclusion of adults 65 years of age or older (1%-22%), individuals experiencing homelessness (4%—100%); those affected by mental health conditions (30%-87%) or substance use disorders (12%—-67%); and those ever involved with the justice system during enrollment (0%-100%). Data showed considerable variation across pilots in the average use of services pre-W PC (per enrollee, per year) for outpatient services (7.4—50.4), ED visits (15.8), and hospitalization rates (0.3-2.2). We ian e) eal e) Mek Neal WPC Pilot Profiles Collectively, these data demonstrate how individual pilots tailored their approaches to address community-specific needs. For example, Los Angeles County's WPC pilot focused on all six target populations and used multiple programs and forms of outreach to identify and enroll eligible beneficiaries A diverse care coordination team that included peer staff helped link enrollees to a medical home and services such as housing and medical respite. In another example, Riverside County’s WPC pilot focused on serving the justice-involved population; co- located WPC enrollment staff with probation staff to enroll individuals in jails and prisons prior to release; and used a single dedicated care coordinator (typically, a registered nurse) to connect enrollees to a medical home and services, including employment assistance. Implications for Transition to CalAIM This snapshot is intended to inform efforts to transition the WPC program into ECM and ILOS components of CalAIM. Heterogeneity across pilots in program implementation and enrollee characteristics highlights the importance of tailoring future efforts to the unique needs of various subgroups of Medi- Cal enrollees with high utilization of services. In some counties, a narrower focus on specific target populations or smaller enrollment indicate that additional work is needed to expand enrollment to everyone with high levels of need and service use. The data also reflect the level of effort necessary to establish a specific infrastructure for effectively serving identified target populations WMG ts aoe) alae Molt ela 3 Exhibit 1 WPC Program Implementation Elements by Pilots as of July 2020 4 3| 2) > q ry oes ry 3 See s FS SS eal | Ey s i Sle | 2 5 < & S4/s/s z Primary target population 1. High utilizer 2. Homeless 3. At risk of homelessness 4. Chronic physical conditions 5. Severe mental illness/substance use disorders (SMI/SUD) 6. Justice-involved Enrollment Strategies Identification approach 1. Street- or shelter-based outreach 1) 1)1 41 1 1) 14 2. Health care facility outreach 2,2)2)2)2 2/2] _]2] 2 3. Referrals 3/3 ]/3)3]3)3)3)3 3) 3/3 4. Administrative data (e.g., health plan eligibility lists) 4 4 4/4] 4)4 5, Predictive modeling based on program criteria 5 5 Enrollment approach 1. At health care facilities r}a}afryr. da 1}4]4 2. Warm handoff at co-located organization 2]2]/2]2/]2)2);2)2)_)2)2]2 3. On street, at shelter, or other community-based location 3 i 3 3 3 8 3 3 > 4. By telephone 4)4)4 4] 4 4 5. Auto-enrollment and opt out 5 5 5 Care Coordination Approach Organization of care coordinators (CC) 1. Single CC ee 141 - 1] 1 2. Multiple CCs 2] 2 2 2 (25, Average CC caseload (by tier) 80,} — | Go, (17, 15 | 300) 75) | 25 | 30) | 10 | 19 | 43 | 40 | 35 | 20 | 50 Selected types of staff included in care coordination team 1. CHW or staff with lived experience 1/1 1}a]a]a]a4 ~fafada 2. Licensed social worker or psychologist 2] 2]2 2/2)2)2)2 2) 2] 2 3. Physician or nurse practitioner 3 | 3 | 3 3 | 3] 3] 3 3 3 Type of co-located staff to facilitate access to services and resources 1. Medical aJrfarsyrtapaya 1 1 1 2. Mental health 2 2/2)2)2) 2 2] _]2 2 3. Housing 3 3 3] 3 4. Non-housing social services 4/4} 4])4]4] 4] 4 4) 4 5. Substance abuse 515 5 5 5 CCs have real-time access to at least some of the following data: 1. Medical tyr} ry r} ttt inet t]nuof - | 1) 144 2. Behavioral health 2) 2 2 2 |%° ° 2) 2 3. Social services 3 [3 | 3 | 3 3 3 3 Care coordinators can access needs assessment, comprehensive care plan, and referrals in the same system CMA Oe ee wee nl Selected WPC Services Offered Housing-related services 1. Housing navigation, tenancy support afarfatartapapatata a}4]4 2. Landlord incentives 2 2 2)2)2 2]; 7] 2)2]2 3. Funds (e.g., security deposit, utilities) 3 [3 | 3 3/3 [3] 3 [3 3 | 3 Selected other services 1, Employment assistance t}a}afafrpryafayofrajpradya 2. Sobering center 2 2) 2 2) 2 2 3. Recuperative care (medical respite) 3 3 3/31/31] 3 3/3 Partnership Characteristics Total number of organizations participating in WPC pilot 42 | 12] 15 | 8 | 50] 39] 11] 8 | 17] 12 | 34] 24] 14 Types of partners with highest engagement with WPC administration ° 1. Housing 1 1 1} 1 efa;a;afa)4 2. Justice 2/2)]2);2/2)/2)2)3)2) 2 2/2 3. Other social services 3 3 3 3 3 3 3 3 3 3 3 3 Note: Unavailable data are indicated by a dash (-). (Exhibit 1 continues on next page) 4 UCLA CENTER FOR HEALTH POLICY RESEARCH WPC Program Implementation Elements by Pilots as of July 2020 Exhibit 1 & re o is S 5 i eA Pe (SCWPCC) Cena UerEC iy eae PeeKer te] Santa Cruz San eee ri ie atts) RTT) Primary target population 1. High utilizer 2. Homeless 3: At risk of homelessness 4. Chronic physical conditions 5. Severe mental illness/substance use disorders (SMI/SUD) 6. Justice-involved Enrollment Strategies Identification approach 1. Street- or shelter-based outreach 1/4 afa]a)4 1 1/4 2. Health care facility outreach 2) 2 2}/2/2]2]2 2) 2 2 3. Referrals 3 | 3 3/3/3]/3]3]3]/3]3)3 43 4. Administrative data (e.g., health plan eligibility lists) a4}/4a]4 4}/4]4]a]a4 4)/4]4 5. Predictive modeling based on program criteria 5 5 5 5 Enrollment approach 1. At health care facilities 1/a]a]4 a}afatafarfajada 2. Warm handoff at co-located organization 2)/2)]2 42 2}2|2 2)/2/2 3. On street, at shelter, or other community-based location 3] 3 | 3 3/3]/3]/3]/3]3]3]3 4. By telephone 4 4]4a]4 4 | 4 5. Auto-enrollment and opt out 5 Care Coordination Approach Organization of care coordinators (CC) 1. Single CC 1 1 1] 4 1] - 2. Multiple CCs 2 2|2 | 2 2 2 2 . (55, 6, (6, (60, Average CC caseload (overall and by tier) 75 | 13 | ss | 73) | 176| 75 | 31) | 30 | 30 | 23 | a5 | 20 | 100) Selected types of staff included in care coordination team 4: | as 1. CHW or staff with lived experience 2}, e}a}afarjarfaryada 1/441 2. Licensed social worker or psychologist 3 | 2 2)/2|/2]/2]2]2]2]2)2]2 3. Physician or nurse practitioner 3/3 |3]3]3 ]3 3 | 3 Type of co-located staff to facilitate access to services and resources 1. Medical 1 a | w| 1 a | lw) we! «| « 2. Mental health 2 c c S c < € < 3. Housing 3/3|/3/3/3]/7-)]3 2)/2/2)/2)3 4. Non-housing social services 4 4 4 4 5. Substance abuse 5 CCs have real-time access to at least some of the following data: 1. Medical 1/4 1 1/4 1 1 2. Behavioral health 2 | 2 |'Ne Ne 2 | No No} 2]-|a2 3. Social services 3 | 3 3 3 | 3 3 3 Care coordinators can access needs assessment, comprehensive care plan, and referrals in the same system we) ele | wee |) a |# Selected WPC Services Offered Housing-related services 1, Housing navigation, tenancy support 1} a]}afarfada af a]}a.]4 1 2. Landlord incentives 2) 2 2 - 3. Funds (e.g., security deposit, utilities) 3 | 3 3 | 3 3] 3 [3] 3 Selected other services 1. Employment assistance 1/4 1 1 1} a]a.]4 2. Sobering center 2] 2 2)/2]2]|2 2 2 3. Recuperative care (medical respite) 3{/3|3]3 [13] 3 3 | 3 3 3 Partnership Characteristics Total number of organizations participating in WPC pilot 31 | 10| 9 | 20] 9 | 25] 8 | 43} 18 | 15 | 11 | 16 | 46 Types of partners with highest engagement with WPC administration ° 1. Housing 1/1 rfafartarye|a 1/4 2. Justice 2 2 | 2 2 2 2 3. Other social services 3/3 ]3]|3 [3 3 3] 3 [3 | 3 Note: Unavailable data are indicated by a dash (-) WMG ht aoe) alae Molt ela 5 Exhibit 2 WPC Enrollment Profile by Pilots for the First Two Program Years, 2017-2018 cs ol |les _ cd Prous ° 3 8 ae 3 iS bs =| | 2 & 5 i aes 3 z g Pas Fa Primary target population 4. High utilizer 2. Homeless 3. At risk of homelessness 4. Chronic physical conditions 5. Severe mental illness/substance use disorders (SMI/SUD) 6. Justice-involved Enrollment Characteristics Total enrollment 1. Up to 300 1 rfafada 1 2, 301-1,000 2 2 3. 1,001-10,000 3 3 3 4, >10,000 4 4 Ever disenrolled (%) to] s6| 4] 49] 6] 2{ -| 15 | 44] 38] 57] 63 | 15 Mean length of overall enrollment, in months 7[/43[ s| 7[ 1] 3[ s[ 9/14] of afi] eo Enrollee Demographics Age 0-20 at enrollment (%) 3] s| 2[-]1]-] olf -]| of -[ 3] of 3 Age 45-64 at enrollment (%) 48 | 38 | 41 | 33 | 48 | 53 | 63 | 50 | 62 | 48 | 50 | 63 | 21 Age 65 years or older at enrollment (%) 6] as{ 4[- [| sfa2]- [alfa] 5] 7] 10] 1 Male (%) 56 | 40 | 53 | 55 | 62 | 63 | 52 | 50] 48 | 61 | 59 | 58 | 76 White (%) 22 | 27 | 34 | 37 | 21 | 61 | 85 | 76 | 34 | 69 | 48 | 75 | 33 African American or Black (%) a4 | 22[ 13] [asf 16] of -|-|- | 6[ - [15 Latinx (%) 12 | 24[ a1] 43 | 28[ 10 - | 7 | 34] 19] a5] 7 | 43 Ever homeless during enrollment (%) 19] 4| 31] 15] 51 | 64 | - | 46 | 95 [100 | 100 | 97 | 27 Ever justice-involved during enrollment (%) - | - [42] 30] 2[ of - [as] - [ 0] 0[ 20 [100 Enrollee Health Status at Enrollment (Light Orange = Lowest %; Dark Orange = Highest %) Any chronic physical health condition (%) 73 | 59 | 53 | 64 | 69 | 69 72 | 37 Hypertension (%) 2a | 21 | 22 | 15 | 20 | 20 als Diabetes (%) [is | 12] 13] 12/8 12] 2 Any chronic mental health condition (%) 65 | 33 | 30 | 54 | 58 | 62 66 | 33, Any substance use disorder (%) 38 | 12 [15 | 22 | 24 | 37 44 | 23 Pre-WPC Utilization per Enrollee per Year (Light Orange = Lowest Quartile; Dark Orange = Highest Quartile) Number of outpatient services 22 | 10 | 20 | 15] 20] 19 23] 7 Number of outpatient mental health services 3] 3/5 6 a| 3 Number of outpatient substance use disorder services 4[ 7 2] 3/2 All A Number of emergency department visits 23 [1.0 | 15 [20] 23 [20 26 [13 Number of hospitalizations 10 [os [os [os Ros 06 [03 Notes: Unavailable ot sparse data are indicated by a dash (-) (Exhibit 2 continues on next page) Health status conditions are based on CMS' Chronic Condition ‘Warchouse condition categories. Utilization is measured during two years pre-WPC enrollment. Outpatient services include any service nor provided in an inpatient setting, at the emergency department, or through long-term care 6 UCLA CENTER FOR HEALTH POLICY RESEARCH WPC Enrollment Profile by Pilots for the First Two Program Years, 2017-2018 Exhibit 2 2 = o E £ 5 3 Cf ne (Savina) Cee eet) eres Peeneriey Santa Cruz San erry aU ie uaricey TTT) Primary target population 41. High utilizer 2. Homeless 3. At risk of homelessness 4. Chronic physical conditions 5. Severe mental illness/substance use disorders (SMI/SUD) 6. Justice-involved Enrollment Characteristics Total enrollment 1. Up to 300 1 1 1] 4 2, 301-1,000 2 2 2 2 2 3. 1,001—-10,000 3 3 3 4, >10, 000 4 Ever disenrolled (%) 31 53 | 28 5 | 43 | 13 | 40 | 17 | 10 | 74} 43 | 38} 13 Mean length of overall enrollment, in months a] stu] sl] il 7[i6[ 17] 13] 12] 13] sf Enrollee Demographics Age 0-20 at enrollment (%) - 0 8 0 0 0 1 1 - - 0 4 1 Age 45-64 at enrollment (%) 61 74 | 44 | 78 | 50 | 50 | 44 | 59 | 49 | 67 | 58 | 42 | 57 Age 65 years or older at enrollment (%) 8 - 6 - 8 4 | 22 8 | 22 - 9] 11 3 Male (%) 57 | 52 | 45 | 58 | 72 | 52 | 52 | 49 | 60 | 50 | 48] 50 | 46 White (%) 38 | 56} 22 | 50 | 29 | 40 | 34 | 29 | 57 | 77 | 32 | 58] 42 African American or Black (%) 31 O} 18] 15 | 31 18 7 8] - - 35 5 4 Latinx (%) 9] 41) 46] 11 11 | 26 | 27 | 34] 11 5 | 10} 12 | 38 Ever homeless during enrollment (%) 98 | 97 4 | 61/100 | 47 | 34 | 41 54 | 98 | 50 ot 59 Ever justice-involved during enrollment (%) ofe{ of 9] - [14] of of 1s] of -] of o Enrollee Health Status at Enrollment (Light Orange = Lowest %; Dark Orange = Highest %) 61 24 14 82 64 15 6 74 28 14 49 57 | 63 | 62 | 53 33 24 42 | 38 | 31 | 28 | 35 ion per Enrollee per Year (Light Orange = Lowest Quartile; Dark Orange = Highest Quartile) 81 82 Any chronic physical health condition (%) 31 27 14 Hypertension (%) Diabetes (%) Any chronic mental health condition (%) 19 70 43 Any substance use disorder (%) Pre-WPC Util Number of outpatient services Number of outpatient mental health services Number of outpatient substance use disorder services Number of emergency department visits Number of hospitalizations Notes: Unavailable or sparse daca are indicated by a dash (~), Health status conditions are based on CMS’ Chronic Condition ‘Warehouse condition categories. Utilization is measured during two years pre-WPC enrollment. Outpatient services include any service not provided in an inpatient setting, at the emergency department, or through long-term care UCLA CENTER FOR HEALTH POLICY RESEARCH HEALTH ECONOMICS AND EVALUATION RESEARCH Appendix M: Policy Brief - COVID-19 UCLA CENTER FOR ::. HEALTH POLICY RESEARCH 6: Appendix M: Policy Brief - COVID-19 | Whole Person Care Final Evaluation Report UCLA CENTER FOR HEALTH POLICY RESEARCH. «, © if The state’s five- year Whole Person Care (WPC) program was extended to December 2021 due to the COVID-19 pandemic. SIN L Mee Ved Health Policy Brief January 2022 Whole Person Care Program Successfully Navigated Around COVID-19 Challenges in 2020 Nadereh Pourat, Ph.D., Brenna O’Masta, MPH, Leigh Ann Haley, MPP, Weihao Zhou, MS, and Emmeline Chuang, Ph.D. SUMMARY: California implemented the Whole Person Care (WPC) Pilot program under “Medi-Cal 2020,” a Section 1115 Medicaid Waiver program designed to coordinate the care of high-utilizing Medi-Cal beneficiaries across medical, behavioral health, and social service sectors. The COVID-19 pandemic stay-at-home orders began in mid-March 2020, during the last year of WPC implementation, and disrupted California's plans to transition WPC enrollees into a new program under the California Advancing and Innovating Medi- Cal (CalAIM) initiative. In this policy brief, we examine the impact of the pandemic on WPC implementation, enrollment, and health service utilization. We found that all WPC Pilots reported at least some pandemic-related alterations to WPC implementation. Total enrollment increased in 2020, with lower rates of new enrollment and disenrollment. The mid-March shutdown also resulted in an initial decline in enrollee health service utilization. However, by the end of 2020, primary care and specialty services had reverted to pre-pandemic patterns, while emergency department and hospitalization rates remained lower than pre-pandemic rates. In this policy brief, we discuss the implications of these findings for the transition to CalAIM and WPC evaluation. t the request of the California Department of Health Care Services (DHCS), the state’s five-year Whole Person Care (WPC) program was extended to December 2021 due to the COVID-19 pandemic. The extension was intended to prevent disruption of WPC services to enrolled beneficiaries while the state prepared for their transition to other programs planned under the CalAIM initiative. A statewide shelter-in-place order was enacted in California on March 20, 2020. The impact of the pandemic and its associated consequences —such as recession, job loss, and reduction in utilization of health care—are well documented and suggest a disproportionate impact on many WPC enrollees. In this policy brief, we examine the progress of the COVID-19 pandemic in California and its effects on WPC implementation and enrollee health care utilization. Our findings illustrate changes during the pandemic in WPC implementation and enrollment and in four categories of health service utilization. We also discuss potential implications for the evaluation of WPC and the implementation of CalAIM. Spread of COVID-19 in California and WPC Counties Nearly 2.3 million confirmed COVID-19 cases and 25,986 resulting deaths were reported in California in 2020. Our analysis of confirmed cases in WPC counties showed Most pilots said that pandemic- related shutdowns and social distancing requirements limited their ability to deliver WPC services in person. UCLA CENTER FOR HEALTH POLICY RESEARCH a cumulative rate for that year of 5,844 confirmed cases per 100,000 residents, relatively similar to the statewide rate of 5,822. When examining the 14-day average daily case rate, we found two distinct peaks among WPC rates in late July (24 confirmed cases per 100,000) and late December (108 confirmed cases per 100,000; data not shown). Most WPC counties had peaks in the same time frame, but there were variations in the magnitudes of these peaks by county (data not shown). Trends in daily hospitalizations from COVID-19 mirrored trends in confirmed cases, peaking at 18 and 52 hospitalizations per 100,000 in July and December, respectively. The Impact of the COVID-19 Pandemic on WPC Implementation WPC Pilots reported the impact of the COVID-19 pandemic on WPC infrastructure and service delivery. Most (20 of 25) pilots said that pandemic-related shutdowns and social distancing requirements limited their ability to deliver WPC services in person. While many providers transitioned to care delivery through telehealth, pilots explained that it was difficult to make meaningful progress toward care management goals when enrollees frequently had inadequate access to cell phones, computers, the internet, or electricity. “For many of our patients ... {without} access to a smartphone ... delivering telehealth services was virtually impossible. We ... create{d} a room ... and set up telehealth equipment ... {for} our provider {to} see the patients from another room in the clinic.” — WPC Pilot, Kern County More than two-thirds of pilots (17 of 25) reported limited capacity to deliver WPC services due to hiring freezes, staff safety concerns, or reassignment of staff to support other urgent COVID-response activities. For more than half of the pilots (16 of 25), pandemic-related restrictions also limited the ability of staff to engage in field-based outreach and provide warm handoffs or other supports needed to effectively engage certain CALIFORNIA'S WPC AT A GLANCE: Purpose WPC was a Medicaid Section 115 Waiver demonstration project designed to coordinate medical, behavioral health, and social services for high- utilizing beneficiaries with complex needs. Enrollees Those enrolled were Medicaid beneficiaries with high service utilization, multiple chronic conditions, mental health conditions or substance use disorders, experiencing or at risk of homelessness, or recently incarcerated. Pilots Twenty-five entities from 26 of the 58 California counties provided WPC services using local partners. All pilots provided care coordination and housing support but varied in other services and enrollees targeted. Timeline The WPC Pilot program, begun in January 2016, was extended by one year beyond its original end date of December 2020 due to the COVID-19 pandemic. enrollees in care, particularly those living on the streets or in homeless encampments. In some pilots (11 of 25), frontline staff also experienced challenges with the remote work environment, which impacted their ability to effectively collaborate with their care team and other WPC partners. Pilots met these challenges by capitalizing on existing WPC infrastructure and, when possible, finding synergies with COVID-19 response activities. Many pilots (18 of 25) reported increased engagement of enrollees, as people could be reached more easily at home or shelters due to the shutdown. UCLA CENTER FOR HEALTH POLICY RE: H cere Monthly Enrollment and Total Quarterly New Enrollment in WPC, January 2017 to December 2020 === Total New Enrollment per Quarter Monthly Enrollment Exhibit 1 COVID-19 Pandemic 96,563 100,000 80,000 62,109 60,000 40,000 = wo o 2 ° g ra) n g es gre -. § $ $= Ff 8B & S$ FB FS & FB BV So N 2 a 2 nN o oo s Y Nn N in a 20,000 = a wo < < I 2 2 8 ms = . | | I | | | | | [ | | | I ° JT ttt ttt SPs ssfrPZHOZaglslslessPZsoZagiselsesspP ZHAozZaiselslesspPrgsloZza ai a2 | a3 | aa | at | a2 | a3! aa | at | a2! a3! aa | at! a2! a3! as 2017 2018 2019 2020 Source: UCLA analyses of WPC Quarterly Enrollment and Note: 23 of 25 pilots started enrolling throughout 2017, and two. Utilization Reports from January 2017 to December 2020 Some pilots (17 of 25) that partnered with short-term housing programs, such as Project Roomkey, were able to better identify eligible enrollees, engage them, and enroll them in WPC services, while also making progress toward care plan goals and increasing short- term housing opportunities. One pilot noted: “Housing {homeless} individuals in hotels not only helped reduce the spread of COVID-19, it allowed for co-location of physical health, mental health, substance use, {and} housing services.” - WPC Pilot, Kings County Several pilots (15 of 25) succeeded in improving collaboration in emergency operations and structures among county partners, as well as establishing closer collaboration with provider networks. Fewer than half of pilots (12 of 25) utilized centralized data systems to find and deliver WPC services to enrollees who were at higher risk from COVID-19. pilots started enrolling in early 2018 The Impact of the COVID-19 Pandemic on WPC Enrollment Exhibit 1 illustrates the trends in monthly enrollment and the total new enrollment per quarter during WPC, including the pandemic. A total of 96,563 Medi-Cal beneficiaries were enrolled in WPC in December 2020, an increase from 77,198 in December 2019. Total new enrollment in the last three quarters of 2020 was lower than it had been in the same quarters in 2019. There was also a 20% decline in average monthly disenrollment in 2020 compared to 2019 (data not shown). Estimated Prevalence of COVID-19 The diagnosis code for COVID-19 was developed and utilized by providers starting in late March 2020. To estimate the prevalence of COVID-19 among WPC enrollees, we analyzed Medi-Cal claims starting in March 2020 and identified enrollees with services for which COVID-19 was the primary or Many pilots reported increased engagement of some enrollees because they could be reached more easily at home or shelters due to the shutdown. We ian e) eal e) Mek Neal Exhibit 2 WPC enrollees with a COVID-19 diagnosis were more often female, ages 50 to 64, and Latinx. Proportion of WPC Enrollees With a COVID-19 Diagnosis From March to December 2020, by WPC Target Populations Severe Mental Illness/Substance Use Disorder Chronic Physical Conditions At Risk of Homelessness Homeless All WPC Justice Involved High Utilizers Source: UCLA analyses of Medi-Cal enrollment and claims data from March 2020 to December 2020, and WPC Quarterly Enrollment and Utilization Reports from January 2017 to December 2020 secondary diagnosis. A total of 8,738 WPC enrollees (4.1%) had at least one service with COVID-19 as the primary or secondary diagnosis (Exhibit 2). This proportion was highest for enrollees identified by the pilots as having severe mental illness or substance use disorders. UCLA compared the demographics of WPC enrollees who had a COVID-19 diagnosis with the demographics of those who did not have this diagnosis (data not shown). WPC enrollees with a COVID-19 diagnosis were more often female (47% vs. 44%), ages 50 to 64 (35% vs. 31%), and Latinx (42% vs. 26%). 5.5% 5.4% Notes: COVID-19 diagnosis was identified using ICD code U07.1 in primary or secondary diagnosis per claim. Enrollees can be reported in more than one target population. COVID-19-Related Service Use of WPC Enrollees ‘We examined the types of health services for COVID-19-related care utilized by WPC enrollees with a COVID-19 diagnosis in 2020. Enrollees most frequently used primary care services (42%) and emergency department visits (33%), followed by hospitalizations (22%), lab tests (18%), specialty services (12%), and stays in long- term care facilities, such as nursing homes and assisted living (8%) (Exhibit 3). The median length of hospitalization for those with a COVID-19 diagnosis was five days (maximum of 114 days; data not shown). Exhibit 3 Proportion of Enrollees With a COVID-19 Diagnosis Who Received Specific COVID-19- Related Services Primary Care Services Emergency Department Visits Hospitalizations Lab Tests Specialty Services Long-Term Care Stays Source: UCLA analyses of Medi-Cal enrollment and claims dara from March 2020 to December 2020. Notes: Setvices with COVID-19 as the primary ot secondary diagnosis (identified using ICD code UO7.1) only. Emergency department visits only include visits that did not result in hospitalization. UCLA CENTER FOR HEALTH POLICY RE: Monthly Utilization of Health Services per 1,000 Member-Months Among WPC Enrollees, Exhibit 4 2019 Compared to 2020 2019, ——=== 2020 COVID-19 Pandemic in 2020 800 ;— 700 |— Primary Care goo: 586 Services 497 500 |— 504 495 400 400 ;— 300 [— Specialty C. Specialty Care 225 207 56° |_ —== == wTrreTs= 209 177 100 250 -— 202 198 200 ;— Emergency Department 150 Visits 4 100 | 143 141 50 The number of ED visits eo declined in April a 2020 relative to Bi 62 April 2019, and Hospitalizations. 50 |- om remained lower os 50 42 through December g 2020 relative to Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Deceniber 2019 Source: UCLA analyses of Medi-Cal enrollment and claims data Note: Member-months were based on Medi-Cal enrollment from March 2020 to December 2020. Changes in the Use of Health Services April 2020 compared to April 2019 for all Before and During the COVID-19 Pandemic service types (Exhibit 4). By December 2020, ‘We assessed service utilization patterns however, rates of primary care and specialty among WPC enrollees before and during service utilization were similar to those in the pandemic, and we found a decline in December 2019. In contrast, the number of 6 UCLA CENTER FOR HEALTH POLICY RESEARCH The rate of services delivered through telehealth increased from fewer than 0.1% of primary and specialty services prior to the pandemic to 9% of primary and 10% of specialty Services in December 2020. ED visits declined in April 2020 relative to April 2019, and the number remained lower in December 2020 relative to December 2019. A similar pattern was observed for hospitalizations. Further analyses (data not shown) found that fewer than 0.1% of primary care and specialty services were delivered by telehealth prior to the pandemic. This rate changed to 2% of primary and 3% of specialty services in March 2020, and to 9% of primary and 10% of specialty services in December 2020. Implications Our analyses indicated that the COVID-19 pandemic altered the type and modality of WPC services and the patterns of WPC enrollment and health service utilization in 2020, which was the last planned year of WPC implementation. The ability of pilots to rely on WPC infrastructure and continue to deliver care coordination and housing support services may have mitigated the impact of the pandemic on enrollees. These findings highlight the value of having future Medi-Cal programs incorporate an infrastructure similar to that of WPC, integrating elements such as partnerships with community-based organizations and data-sharing capabilities. This evidence supports CalAIM’s intent to sustain and strengthen such infrastructure statewide. The ability of pilots to maintain continuity of care coordination and housing support services during the pandemic is likely to have maintained the positive WPC outcomes— for instance, by improving access to needed services and preventing a high use of acute care. Therefore, WPC enrollees might be less likely to have pent-up demand for care coordination and housing support services than Medi-Cal beneficiaries not enrolled in WPC. These advantages are likely to continue after enrollees are transitioned to CalAIM in January 2022. The low proportion of enrollees with a COVID-19 diagnosis reflects the subset of enrollees who received care for this condition rather than reflecting the prevalence of COVID-19. Nevertheless, the findings likely indicate the limited impact of COVID-19— related service use on our evaluation of WPC. The pandemic’s limited impact on the utilization of primary care and specialty services is likely due to the rapid increase in the provision of care using telehealth under emergency Medicaid waivers that allowed for the reimbursement of such visits on par with in-person visits. These findings further support the need to address digital access barriers to telehealth for WPC enrollees and other Medi-Cal beneficiaries. Lower use of ED visits and hospitalizations from pre- to post- pandemic rates also indicate the importance of addressing these changes in our evaluation of WPC. UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 Introduction and Instructions The UCLA Center for Health Policy Research was selected by California Department of Health Care Ser- vices to evaluate the Whole Person Care (WPC) pilot program. This questionnaire is intended to assess how participating Lead Entities (LEs) have implemented the Pilot and to understand your efforts towards achieving WPC program goals. This questionnaire is comprised of a mix of closed- and open-ended questions, and is divided into the following domains: Respondent Information The Local Context Motivation for WPC WPC Infrastructure and Resources WPC Implementation WPC Leadership, Communication, and Decision-Making Processes Inter-agency Collaboration Identifying and Retaining Eligible Beneficiaries |. Perceived Impact of WPC 10. WPC Program Monitoring, Feedback, and Performance Improvement 11. WPC Learning Collaborative PEN ARHAwWNE This questionnaire is to be completed by the individual(s) most knowledgeable in implementing the WPC program within the LE institution, which may include one or more persons depending on the LE. The questions are intended to be distinct from LEs mid-year and annual reports to DHCS and narrowly fo cused on specific issues. In completing this questionnaire, please focus on the LE perspective. A sepa- rate companion questionnaire will solicit partner perspectives. You can distribute the PDF version of this questionnaire to the most knowledgeable individual(s) within the LE institution to complete the relevant sections of the survey. However, we ask that all responses are entered online by one individual due to limitations of our online data system (SurveyMonkey). We anticipate that this questionnaire will take about 2-3 hours to complete. For ease, please enable cookies on your browser. With cookies enabled, responses will be saved prior to submission of the questionnaire as long as the respondent uses the same computer and browser. Confidentiality. Your responses on this questionnaire will be confidential. Only the UCLA evaluation team will have access to your individual responses. Only aggregated data will be included in evaluation reports and publications. Your responses to this survey will not impact your WPC funding from DHCS. The evaluation team are available to answer your questions if needed. Please contact the UCLA evalua- tion team at wpc@chpr.em.ucla.edu with questions. Whole Person Care Final Evaluation Report | Appendix N: Lead Entity Survey Instruments JES} Methods Population-level COVID-19 data were created using the Los Angeles Times and the July 2019 U.S. Census population estimates. Subsequent COVID-19 rates were likely underreported at the start of the pandemic. WPC enrollment data were based on an analysis of WPC Quarterly Enrollment and Utilization reports from January 2017 to December 2020. The data on the effects of the COVID-19 pandemic on infrastructure and service delivery, and associated challenges and successes, were based on an analysis of WPC Program Year 5 Annual Narrative Reports from July to December 2020. Identification of enrollees with a COVID-19 diagnosis was based on a primary or secondary diagnosis of COVID-19 (ICD codes U07.1) in Medi-Cal claims data. Health service utilization data were based on an analysis of Medi-Cal enrollment and claims data from January 2019 to December 2020. Utilization rates were not adjusted for patient characteristics. Author Information Nadereh Pourat, Ph.D., is a professor of health policy and management at the UCLA Fielding School of Public Health, associate director of the UCLA Center for Health Policy Research, and director of the center's Health Economics and Evaluation Research Program. Brenna O’Masta, MPH, and Leigh Ann Haley, MPP, are project managers and research analysts at che UCLA Center for Health Policy Research. Weihao Zhou, MS, is a senior public administration analyst at the UCLA Center for Health Policy Research. Emmeline Chuang, Ph.D., is director of the Mack Center on Nonprofit and Public Sector Management in the Human Services at the University of California, Berkeley, and an associate professor in UC Berkeley's School of Social Welfare Funder Information Funding for this project was provided by the California Department of Health Care Services. Acknowledgments Sincere thanks to Ninez Ponce and Ana Martinez for reviewing this policy brief and providing helpful feedback. Suggested Citation Pourat N, O’Masta B, Haley LA, Zhou W, Chuang E, 2022. Whole Person Care Program Successfully Navigated Around COVID-19 Challenges in 2020. Los Angeles, CA: UCLA Center for Health Policy Research. UCLA CENTER FOR HEALTH POLICY RESEARCH HEALTH ECONOMICS AND EVALUATION RESEARCH Appendix N: Lead Entity Survey Instruments UCLA CENTER FOR &:. HEALTH POLICY RESEARCH ,o; es on fit Appendix N: Lead Entity Survey Instruments | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research py ber 2022 . " fea at Health Economics and Evaluation Research Program Domain 1; Respondent Information 2) Name of your LE organization: This survey is focused on the LE perspective, and should be filled out by the individual(s) within the LE organization that are most knowledgeable about WPC. We realize there may be considerable variation across LEs in who these individual(s) may be. To provide context for survey re- sponses, please provide the names of all individual(s) within the LE organization that completed the survey, their title and (if applicable) the LE department or division in which they are located, and their role in WPC (e.g., WPC program manager). 2), Names of individual(s) within the LE completing this survey: Name Title Department/Division | Role in WPC {ifapplicable) Email/Contact info ‘Questionnaire Domain(s) Addressed 3) On average, how often has your LE organization participated in meetings with WPC partners about the WPC pilot program during plan- ring and implementation phases of WPC? We understand that each pilot will have different workgroup compositions and titles, but please try to fit your partner meetings into the categories described below. Any concerns can be noted in the comment section, Planning phase implementation phase | Meeting type. Executive / steering committees ClWeekly Weekly Cl Biweekly CO Biweekly C1 Monthly. Cl Monthly Clauarterty C Quarterly Clother (please specify _) ‘Clther (please specify 1 Does not apply Si Does not apply Data governance and sharing committees Clweekly Oi Weekly CBiweekly Ol Biweekly Cl Monthly Cl Monthly Appendix N: Lead Entity Survey Instruments | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 Planning phase Implementation phase Ci Quarterly D Quarterly other (please specify__) Cl Other (please specify Does not apply Does not apply Operation committees CWeekly Weekly Biweekly Cl Biweekly Monthy 5 Monthiy Cawarteny Quarterly other (please specify Other (please specify Does not apply Ci Does not apply ‘if you would like to comment on any of the items above, please specify and do so here: Whole Person Care Final Evaluation Report | Appendix N: Lead Entity Survey Instruments Ge December 2022 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Domain 2: The Local Context This section asks questions about the environment under which WPC is being implemented, in particular which initiatives your LE was already participating in prior to or during WPC. 1)_ Is your LE participating in any other initiatives similar to WPC (e.g,, similar goals, services, and/or clients/patients served)? [1f no, skip to Domain 3]. No Ces 1a. [if yes] Please provide the name of the initiative, funding sources (if applicable), approximate time frame (start and end dates), and extent to which there is synergy between this initiative and WPC. Examples of initiatives that could be similar to WPC: PRIME, Health Homes, and Full Service Partnerships. Name of initiative Source(s) of funding: Approximate time frame (start and end date}: Ona scale from 0 to 10, where O=No synergy and 10=£xtremely high synergy, please indicate the ex- ‘tent to which there is synergy between this initiative and WPC? Appendix N: Lead Entity Survey Instruments | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 Domain 3: Motivation for WPC The following questions relate to perceived benefits of participating in the WPC program and how WPC fits with your LE’s mission and overall strategic goals. 1) Please rate on a scale of 0 to 10, where O=Not at all important and 10=Very important, the importance of the fallowing to your LE’s decision to participate in WPC. Ifa particular element is not applicable, please select N/A and explain in the comment section. omnes! » | 2 | 3 | 4 10 = very e}/7]e|o comment Important et ee ‘important a. Synergy with existing programs b. Consisteney with or- ‘ganizational goals Improve integration of care for li- ents/patients with, multiple needs 4. Develop collabora- tive relationships with participating WC entities & Continue/maintain existing relationships with participating WC entities f Getting necessary services for elt. o a ojoj/oala a olajoja o ents/patients Getting cient/pa- tient referrals from participating WPC entities Ease of mplementa- tion (eg, due to o o oj/o/aja o olajaja a concordance with o]| o |ojolofo} a J|olofa}fa}] a o| ao }ojolofo} oa Jolofoa}a}] a Whole Person Care Final Evaluation Report | Appendix N: Lead Entity Survey Instruments [G&} December 2022 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Se har = nono 10=ven i. tant nor mmer MA apotent Lk important | Comment nimort wasting processes of care) 17 tow esoure re quirements (e.8., lowest cost, least a a o a stafftime to imple ment) 7 Redewwtorare | |g a a Te improve quality of = of a a cy Dena eenespesTy o o o 2) Onasscale from 0 to 10, where O=Very low and 10=Very high, please indicate the extent to which each of the following WPC pilot program goals and/or program components fits with your Le's overall strategic priorities. Ifa particular element is not applicable, please select N/A and explain in the comment section. oe ae ES Manage the eare of high riskand high utiizng | | Qo o populations Use of case management tomanageheattheare | ch | Ol oO a utilization © _ Earlier identification of patient/client needs a | 8 a e Identify clients/patients recelvingservicesfrom | | o o ‘more than 1 system Appendix N: Lead Entity Survey Instruments | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 NA O= Very tow 2 |/3)\| ce) 2eheee |e 7 |\a. | We ow nor hgh 10= ery high Reduce inappropriate ‘emergeney department visits and hospitaliza- tions Improve quality of care Coordinate health, be- havioral health and so- cial services Sharing data with exter- nal partners Increase clent/patient ‘access to housing and supportive services (eg, housing navigation, ten- ‘ancy support) Increase clent/patient ‘access to other social services (e.g., employ- ment assistance, TANF, etc) Increase dlent/patient ‘access to mental health ‘and/or substance abuse treatment Whole Person Care Final Evaluation Report | Appendix N: Lead Entity Survey Instruments 465 UCLA Center for Health Policy Research py | grAeyrd . " fea at Health Economics and Evaluation Research Program 3) Ona scale from 0 to 10, where 0=Very low and 10=Very high, please indicate the extent to which WPC program implementation is a priority for your organizatio = verviow 1 2 3 4) | ages ||| 5 7 8 9 | 10=Vveryhieh comment o o o a o a a o o oO a Appendix N: Lead Entity Survey Instruments | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 Domain 4 astructure and Resources This section asks questions around infrastructure and resources related to WPC activities. We are interested in learning about infrastructure and resources in place prior to WPC as well as efforts to develop additional infrastructure as part of WPC. 4) Please indicate whether your LE organization participated in any of the following activities with INTERNAL WPC partners prior to WPC and/or whether you are planning to implement any of these activities as part of WPC. Internal partners are organizations that work under the same umbrella agency as yours such as county hospital or county mental health department. Ifa particular element is not applicable, please select N/A. (Select all that apply) [Prior twee | Panofwrc [N/A | Comment Health information technology and data sharing 4. Business associate agreements or memorandum of understanding a Oo a b. Date use or sharing agreements a a o c _ Electronic sharing of client/patient information via a centralized data ware- house and/or a query-based record locator (health information ex- a a a change) Bidirectional electronic referral a a a @.__ Shared electronic sytem for tracking care management services a a a #.__ Standardized electronic intake forms Oo a a & — Standardized diagnostic and/or evaluation or assessment tools a a o nh. Standardized clent/patientreferal protocols a a a i. Real-time access to client/patient data by providers/staff a a o Care coordination a. Shared coordinated assessment system to identify high rak/need ee ents/patients and prioritize receipt of services a a a 3. Use af shared care navigators o cae coordinators to guide clients/patients d ig or a s{patient: in io 0 receiving care & Godocation of providers or salto faitate acess to services and/or re sources i o a a a. Maiteiscplinary teams comprised of prowders and/or staf from ruple disciplinary pr Pi nd pl 5 a a organizations Whole Person Care Final Evaluation Report | Appendix N: Lead Entity Survey Instruments Gy UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 Prior toWPC | Part of WPC N/A ‘Comment &._ Warm hand-offs of clients/patients to partners o a o te yaferences including multidisciplinary providers and staf to discuss —— 1ces including multidisciplinary providers and staff to discuss a 5 a Other (please specify o a o Please indicate whether your LE participated in any of the following activities with EXTERNAL WPC partners prior to WPC and/or whether you are planning to implement any of these activities as part of WPC. External partners are organization outside your umbrella agency such as health plans, community clinics, county probation/law enforcement, housing service providers, etc. a particular element is not applica- ble, please select N/A. (Select all that apply) [_Priortewec | Partofwec | N/A | Comment Health information technology and data sharing 2. Business associate agreements or memorandum of understanding o a a ®,__Date use or sharing agreements a Oo a Electronic sharing of clent/patient information via @ centralized data ware house and/or @ query-based record locator (e.g., health information ex- o a a change) d. Bi-directional electronic referral oO a a Shared electron system for tracking care management services a a a Standardized electronic intake forms oD oO a & Standardized diagnostic and/or evaluation or aesesoment tool a a a Th Standardized cient/patient referal protocols o oO a i Real-time access to client/patient data by providers/staff o a a Care coordination a. Shared coordinated assessment system to identify high risk/need cli- ents/patients and prioritize receipt of services a A 2 3. Use of shared care nvigators or care coordinators to guide clents/paients receiving care " : ii o a 5 Appendix N: Lead Entity Survey Instruments | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 Porto WeC_| Pan of WoC WA en . Cordocation of providers or staff ofeltae acess to services aor re T e ue a a a sources d. Maltiiscilinary teams comprised of providers and/or staff rom matiple 5 5 organizations Warr tand-offs of cianta/paton To partners a a a f. Case conferences including muliiscpinary provides and staf to discuss 5 3 multiiscpinay 4 a 4 iointcare Other (nleasespeaiy a a a 6) Do you participate in a health information exchange? [IF no, skip to Domain 5]. a. Ces b. ONo 7) if you have participated in a health information exchange (HIE) prior to WPC and/or will participate in an HIE as part of WPC, please answer the following questions, ‘a. Please specify the names of the health information exchange: b. Please indicate which agencies in your local government participate in the HIE (Select all that apply) Health services agency Ci Mental health agency i Substance abuse agency C Human service agency (e.8., housing) 1 Probation/law enforcement Other (please specify: c. Please provide the year when your lead entity first began participating in the HIE (or anticipated start date if planned): Date: Month Year 4. Please indicate the type of data architecture model of this HIE: CO centralized 1: Centralized via County infrastructure/EHR Whole Person Care Final Evaluation Report | Appendix N: Lead Entity Survey Instruments JG} UCLA Center for Health Policy Research By ber 2022 . " a at Health Economics and Evaluation Research Program centralized 2: Centralized via third party organization CO Federated/decentralized (i.., client/patient data owned and stored locally at point of service) Ci Hybrid model (a cross between the centralized and federated architecture, e.g,, where some data stored in a central- ized data repository) Dother (please sp Please specify what type of data is currently shared in your HIE (Select all that apply): Demographic data U Medication history (e.g., medication prescribed) DLab and imaging results Cealth care encounter/visit data ‘CiMental health treatment encounter/visit data Disubstance abuse treatment encounter/visit data Dther service encounter/visit data (e.., social services) CClient/patient medical history Dther data on social determinants of health (e.g., income, employment, housing) C1 Event-based notifications/alerts Cther (please specify: _) f. Does the HIE under WPC have the following functionalities (select all that apply)? Ci Aggregating data and reporting Track eligibility and enraliment CO Event notifications/alerts (e.g., to PCP upon hospital discharge) tracking enrollees across various systems IF you would like to comment on any of the items above, please specify and do so here: Appendix N: Lead Entity Survey Instruments | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 Domain 5: WPC Implementation The questions in this section asks about implementation of the core components (as outlined in Attachment HH to the WPC Special Terms and Conditions) and overall implementation strategies as outlined in your LE’s WPC application. Please answer these questions from the perspective ofthe LE, 1) Overall, on a scale from 0 to 10 where O=Not at all and 10=Very much, how much have you had to change organizational policies and prac- tices in order to implement WPC? aznenn | a | 2 | a | a | Scmme | | 7 | ao | 9 | soevenmun | comment a «(ee | a a fala] o Q 2) Please rate the overall level of effort required of your LE to implement the following WPC program activities on a scale where 0 =Very low and 10 =Very high. if you are not engaged in a specific activity, please select N/A. wa] over [a [a] | 4 [Stumm] c | 7 | a | a [2%] comment 2, WOC data governance ve, mane agement of data being shared as | C1 a oj;a;o;a a o;o/;a a part of WPC) B. Other WPC program governance [e.8., participation in committee: o oa ee | a) o o;a;o/;oa a meetings) Recruiting or hiring provid a ers/staff to deliver WPC services a a ey a a 5{O)o a a. Ensuring suficient physical space ndjorotrersdmins: | | on talalolol a alal oa tive infrastructure necessary to implement WPC Whole Person Care Final Evaluation Report | Appendix N: Lead Entity Survey Instruments [WAR December 2022 UCLA Center for Health Policy Research Health Economics and Evaluation Research Program N/A On Vey 5 weiter low nor high @. Executing Data Use Agreements [DUA or Business Associate Agreements (BAAS) with LE and/or other WPC partners %_ Datasharing with LE and/or ‘other WPC partners for commu nity needs assessment and pro: gram planning & Data sharing with LE and/or ‘other WPC partners to track WPC program results/autcomes h. Data sharing with LE and/or ‘other WPC partners to identify ‘opportunities to improve the \WPC program Coordinating or integrating WPC activities with health plan part- ]_ Delivering WPC services (eg, ‘case management, housing navi- ation and tenancy support, linkage to re-entry, substance use disorder or mental health ‘treatment, or other support ser- vices) ke (dentifying eligible beneficiaries IC Engaging eligible beneficiaries Tm, Meeting WPC reporting require: ments and timelines Appendix N: Lead Entity Survey Instruments | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 On a scale from 0 to 10 where 0=Very low and 10=Very high, please rate the extent to which turnover or other changes to leadership within your LE has posed challenges to implementing WPC? — 6 7 a 9 110 = very high ‘Comment = Verviow 1 2 3 4a cetane. o o o o Ona scale from 0 to 10 where O=Very low and 10=Very high, please rate the extent to which turnover or other staffing changes within your LE has posed challenges to implementing WPC? 5 = Nether y mane 1 2 3 4 ooo 6 ri 8 9 10 = very high Comment a o g a a o o o a a a We are interested in learning about the ways in which your WPC program has changed from what was proposed in your original WPC appli- cation, Please rate the extent to which each of the following have changed over time on a scale of 0 =Not at all and 10 =Very much. If not applicable to your WPC program, please select N/A. Wal) net || a lz | 8) ||| @ 6: |= ||| nee aren Comment stall much WEE epuram ena o/ajfojajajfa a ojofjalo o aa ni a tal o}ojajaj/oa o oja/oaja Services delivered (e.g, case manage- menthousingasss- | O | O |o|aj|ala o o}o/;a/a o tance, other support services) Whole Person Care Final Evaluation Report | Appendix N: Lead Entity Survey Instruments [Wey UCLA Center for Health Policy Research By ber 2022 . " a at Health Economics and Evaluation Research Program 9 | ever 5 = Nether wns RENE a | oz) [sell ar ire | 8)|l) 71] 8 soe Comment. d.Process(es) for sharing datawithweceat- | o | o |ol|oj}ajlo o BE || Gh) a o fe. Process(es) for identi- fying or enrolling eligi- ble beneficiaries in wee f. Process(es) for engag: ing and retaining elig- ble beneficiaries in WPC programs) 8 Universal or adminis trative metrics used to track and report WPC outcomes fh. Other (please specify 6) Ona scale from 0 to 10, where O=Very low and 10=Very high, how would you characterize overall buy-in for data sharing and/or care coordi- nation activities among each of the following categories of partners? If nat applicable to your WPC program, please select N/A. AN See ee (ea |e en linen e Comment vervow ee very ith Sr ae es a/oafojlo/ofo}] a afojoj/o/oa Hospitals a/oafalefefol o fofafafofoa & Other health care pro- vderseg.comm- | a | o fofofojo]/ a jolaojojlo] oa nity health centers) Appendix N: Lead Entity Survey Instruments | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 oe Sznme i= wa ]et + [2 [2] 4 [sume] s [oe [|e [oo [2] comment Mentalhesther® |) ol oo lololjojo)/ o jolajola]o Substance abuse weamentprovdes | 2 | FPSPP;e {S| 8 |F)s/e}e| se Howsingprovdes | o | o f[ofalo{/a| ao |alaoloalafa Justice system of/ofafofofof a fofafefofa Other soda servi omelet | o folalalo| a fololalfol| a (Other (please spel seer | a} ao fofelo/o; o folal/olo|o How is your LE using shared data as part of the WPC program (Select all that apply)? Inform collaborative community needs assessment with partners Inform collaborative program planning with partners O identify target populations Identity eligible Medi-Cal beneficiaries Provide real-time data access for providers/staff to use in developing care plans and/or coordinating care for clients/patients Ci Support workflows for care transitions across different service settings inform quality improvement efforts with partners Ci Track and provide feedback to partners Cl Other (please specify ) Whole Person Care Final Evaluation Report | Appendix N: Lead Entity Survey Instruments [AWAY UCLA Center for Health Policy Research By ber 2022 . " a at Health Economics and Evaluation Research Program suonevasaeee | oroaree | *ttberaee | pyres | strona earee | commen & AIWPC partners aren aareement about pe a o a o a epawely [] *(Ajdde jeu |e 39a/aS) YM BuljOM aie NOA (sal) Ay3UZ Ped] DdM Y!YM Ajloads aseald [SAA 4] “eS ON 2 S9A [J [9 uonsanb 03 dpyjs ‘ou yy] ¢(37) Aiuy pee7 dM auO UeY} aJoW YM BuUlJaujed UO!}eZIUeZIO INOA S| “¢ (—_____ Ajioads aseajd) sapinoud Ayunuwos 43430 [ (Aqunod jou 93a ‘Bulsnoy ‘ple jeda| “3"a) Japiaosd sadiasas jelso0s / ueWNH [] (Aqunod jou) Aduage yuaw}ea] asnge adueysqns Jo yyeay jeWUa| [) JOMIAU IJUI|9 JO DIUI]9 AUNWWOD lewdsoy uejd yijeaH CF] ("___ Ajloads aseajd) Aouage a1)qnd wayi0 juawaoiojua me| / uoleqold [] Aduase Sulsnoy AyunoD F] Anuagde uawijead) asnge aauejsqns AyunoD [] Aduage yyeay jequaw Ayuno9 [) *(Ajdde yeu |e Ja/aS) “adA} UOIJeZIUeZIO INOA aYedIpU! asea|q “7 éaaey uolzeziuesso INOA saop sj Auew Moy Ajajyewixouddy “¢ ("__ :Ajipads aseajd) 48419 (4adeuew ased ‘4o}euIpsOOd ased “3"a) yyeys BUI]}UOI4 dosiAsadns auljjuos4 5] (40aJIP WesSOJd Jo WaBeUeW JdM “3'a) JUBWAaseUeW jana] WeISO1d [] (40}9841g 8AIINIaXxJ ‘OOD ‘03D “8'a) diyssapea| soluas ~~ uoqeziuesiQ a4} UIYUM ajOYyINOA ‘7 WEN $,UOIeZIUEBIO INOA “T UO!JEWJOJU] JUapUOdsay WeiZolg Yoseasay UO!eNjeAq pue s2!WoUCs] YyyeaH youeasay Adijog Yye9H JJ 493Ua WION radeyaee lets Ta] fxcioqy JUAWUNJ}SU| Aanuns JaULed :O xIpuaddy | Woday uoenjenq jeuly aseD UOSJdad BJOUM sJauyed JdM JA4jO JO 3] YIM Buleys e}eq [] uole}Uawa|duu! JO ‘WYy8Isuano ‘juaWaseueW 3a[o1d JdM 404 ajqisuodsas dnos3yJ0M JO aajzIWWOD BulJa—ajs Jo aqua! F] (Ajdde yeu) |e 198/25) -ddM U! paAjoAU! AjjUaINd SI UOIeZ!UeZIO INOA YdIYM UI! SAEM AY} 9LI!PU! aSea|q Aouasy ased yyeaH Ajuno> einqua, [] Sad|AJaS [2190 pue YIJeaH AjuNOD OUe|OS [] Anuady sadiasas UeWN} pue YyeaH Aqunod eyseys Fj BAI}EIOGe||OD B1eD UOSJag BJOYM AjuNOD [JEWS CF] Wwa}shs jey!dsoH pue YyyeaH Aayje, ese[D eJUeS [] wayshs yyeaH AjunoD oajeW ues [] Aduasy saa|Asas ase y3eayH Ajuno3 uinbeor ues [] Yyeay GN Jo JUaWyJeda|aq ODsIdUeJY URS [] Jajuad |ea|pa\| |euoIZay peaymouy AjunoD oulpseuag ues [) yyeaH [esolaeyag Wayshs YyYedaH Aysuarluf aplssarly [] quawyedag sao|Alasg UeWN} pue Yyeay AjunoD Ja2e|g Fj Aquno9 eden [] quawyedag yyeay Ajuno Aasajuoy| Fj Aduasy sadiAuas uewNnY pue YyyeaH AjuNoD OUlsopuaY| [] SadIMIdS YEAH JO JuaWedaqg AjUNOD sajaBuy so7 [] Aduasy sadialas uewny Ajuno3 ssuly [] Jaquad |e2|pay| Way 4 IQ YedaH [eJOIAeYyag Sad!AJas YIJeaH Jo JUaWedaq ‘eWoUOS Jo AjuNOD [] Anuasdy saoias yijeaH ‘25 ejues Jo Aquno) [] Aduasy sadiasas ueWNH pue YijeaH ‘o8aig ues Jo AyunoD [] Aguasy ase5 yyeaH ‘asueso Jo AqunoD F] sad|AJasg URN pue YJeaH Jo JuaWedag ‘ule Jo AyuNOD F] SADIAJBS YIJPBH &ISOD e4}UOD [] ojuaweses Jo AID Fh WeJZ01g Youeasay UOlen|enq puke so!WoU0s3 YyIeaHq yoseasay Aatjod YyeaH 40J JaqUaD YIN radeyapelel OC eTTa] woday uonenjenq jeuly a1eD UOssad BjOyM | JUaWNIIsU] AaAINS JaULed :O xIpuaddy Ea ("_______ :Ajigads aseayd) sayi9 ("_____ :Ajigads aseajd) saajjouua Dg 0} SadIAJas J94}0 JaAI|ag F $9a||01Ua JdM JOJ UO!JEUIPIOOD aJed JO JUaLUaSeUeW ased apIAOJd [] ssauyed JdM Jay}0 JO/pue 3] WO sjeuajad aAladay F] juaw||Osua JOJ syual|a/sjuaijed 3/qisija Jajes pue Ayuap] WeJZo1g youeasay Uoljen|enq pue sowioUucs] YyIeaH yourasay Adljod YeaH 40} Ja3UaD YIN 2202 Jequissaq (fom) JUaLUNJ}sU] Aavuns JaULed :O Xipuaddy | woday uonenjeng jeuly aseD UOSJad BJOUM Oo a Oo Q oO o Oo oO o oO o Oo oO Oo oO oO oO oO Oo oO oO Oo Oo oO oO a Oo Oo oO a oO oO Oo o oO oO oO o Qg oO oO oO (uonte0o (3804 ay3 405 AquNWWOD yesjuo2 pue ejo (aweu | aoinsase jo | aued 10/pue (juawssasse | qed se “3'a) youyed | yed sanyap spaau ju Spaou ‘aie2 Sujuueyd Sulpnjour noA “S'a) | aned/juaro quaned/ju quiof syUBUWOd Auanyjap ynoge uoly syessajyos 404 “3'a) | 40 Ases0ape ur Ayioads | aoiasas quiog | eajunuiwio3 | jualed, (/yual|> Suyeys eyeq quior aseajd) 43410 Ajanijaq aainias | uonesjsiuiupy Sujuue|d auoN suo}}ez|UeZI0 Jaued rsuauyed JqM BUIMO]|OJ a4} JO YOR YIM S}9eJ91U! ATLNAYWYND VO!WOZ/UDBsO ANOA YIYM Ul SARM AY} A}eIIPU! aSea|q “LZ *sdaULed JM JAUIO YIM SUOIDeJaIU! Ue UO!}e1Oqe||09 AdUaZe-JaU! Ssauppe SUO!SANb BuIMo||OJ YL. areola) uolyesoge||op Aduase-19}u| WweJZo1g youeasay Uoljenjenq pue so!woU0s] YyeaH Yyoseasay Aatjod YiJeaH JOJ JaqUaD VION Moday uoljenjeng jeuly aseD uossag BjoyM | JUawNsysu| AaAins JaUed °O xIpuaddy Ea :(s)juawwu0) oO oOo Oo oO Oo oO Oo (uonyeos (3804 a4} 10} Ayunwwos yesjuOd pue eyo (aweu | aoinsase yo | aued 40/pue (quawssasse | wed se “3-a) yauyed | ued sanyap spaou qu spaeu ‘a1e2 Sujuueld Suipnpur noA “3-a) | anyed/quay> quaned/quayj2 quiof syUsWWO0D Auanyap ynoge uo}. sjessajyou 404 ““*a) | 40 Avevoape uy Ayioads @o1Auas JUIOF | ~edUNWWOD Suueys e3eq quior aseajd) 43410 Ajanijeq aainias | uonesjsiuiupy Sujuueld 3UuON suojjezjuesJ0 Jauyed WweJZold youeasay uoNen|enq pue so1woU0s] yea yoseasay Arlo YijeaH 404 JaqUaD YIN areola) Vas) JUBWUNIYSU] AaAINS JaULed 0 xIpuaddy | Woday uo enjenq jeuly a4e) UOSIad BJOUM Bujuleyas Oo o o o o o o oOo; a Oo Qo Aainoyyia 7 _ ~ _ ‘Bulynsoe4 o o a o a o Oo Oo; a oO o Ainoyyia YNIyyIP Aynoyyip unoWyp quawwi0D Ajouianx3, 8 ic 9 aerepow=g| ¥ £ z T We2€10N =9 v/N =0T ade e la) “IdM JO4 UO!eZIUeTIO INOA UIYIIM Jjejs Ulead pue Unidad 0} Uaag sey } YNIYyIP Moy ayes aseajd ‘NdyyIpP AjaWias3XZ = OT pue YNdYJIP [Je Ie JON = 0 BBYM ‘OT 010 WOYajedseUuQ “gs BUJeIS Dd M WeJZ01g youeasay Uolenjenq puke so1woUds] Yea Yyoseasay Aatjod YiJeaH JOJ JaqUaD VION Moday uolenjeng jeuly aseD uossag BjoyM | JUawWNsysu| AaAINs JaUed °O xIpuaddy Ea a 5 A Qo Qo L = a Qo o o Qo oO yum Sueus ejep paseaiouy “q suoljejndod 8uiz1|!3n Qo Qo Oo o Oo i O;o;o;a0 i YBiy pue ysl YyBly Jo a1e> ao jO Juawagseuew SujAoidu] “2 annoaya Jenan aanzaya quawiwoy | Ajewanxg = oT 6 8 é 9 =S 0 e e T | on=0 via “W/N Palas aseajd ‘wesZ01d Jd M ay} JO [203 e aq 0} paAladad Jou JO UMOUAUN JI ‘sjeOS BUIMO||OJ 94} BuIA Ue Je SSUDAIPAO 5A0]!d dM [[e4eN0 BY} 9}e9!pul aseajd ‘aAIDayja AjaWas]XJ = OT PUL BAIIDAJJa JON = CO AJAYM ‘OT 0} O WO. ajedS e UO “OT s9Wodjno Oo go oO oO oO Oo o;oO};o0;0 o g aajjouue paroidw| ‘p a Oo a Oo Oo a o}a/;a/a Qa d ysoo paseasvag «2 o a o Qo a o o}oa};aja Qo g Ayjenb ases parosdwyy-q ose2 Oo a Qo im a Qa o/;/o}oa);oa Qo OO | jouonessaqui paroudwy -e yaiy Aon eae me MN - 6 8 L 9 € z T = vIn quawiwo) =0T 4eyEN = ¢ v ae ‘Y/N Palas aseajd ‘ajqesijdde jou s! yuawWaja JejNd!Wed e 4] “s}eo8 BulyIseJaAO asay} JO yes UO PedUll $,10]Id JdM [|€49A0 9} JO UONdadJed ANOKA aje2!pul aseajd ‘YyBiy Asa, = OT pue Mo] Asa = 0 B4aYM ‘OT O10 WO ajeIseUQ ‘6 “aanedsied 5,UONeZIUeSIO INOA WO) UOI]sanb ysea JaMsue aseajd ‘SAWOIINO |eUO!}eZIULTIO JAY} SUIAOJdUU! JO/pue ‘squaljed/syu: Jo} ased BulAOsdw! ‘sjeos d1yeWWeIZOId SulAaiyse UO DdM JO Jed! Panladiad ay} INoge yse UO!AS SIy} UI SUOI}saNb aYL 2 DdM JO Ped] panlaoiad WweJZo/d youeasay uoNen|eng pue so1woUcs] Yea yourasay Adljod yye2aH{ JO} 4ajUaD YION ade aed ela (ify) JuaWUNJ}sU| Aanuns JaULed :O xipuaddy | woday uoenjenq jeuly aseD UOSJag BJOUM “Y/N Palas aseaid ‘ajqeaijdde jou si yuawaya Jejnoried e 4] ‘payqWWos AjawalxJ = ¢ pue payJIWWOd |/e e JON = T Baym ‘g 0} T JO ajeos 2 UO a}eI aSea|q EIdM JO PUud AU] Jae UAAd SjeO3 BUIMO}|OJ BY} Sululeysns 0} pay!WWO UONEZIUETIO ANOA SI JUa}Xa JeYM O| “TT 0Z0Z Jaquiazaq Ul Spua BulpuNy Jaye y UOdn pling pue ssau80Jd JdM uleqUleW 0} (S97 JO AjjUapUadapu! JO UO!JeJOGe|/O9 Ul) SJUOJJa SUOI}EZIUeZIO JaUeEd UO pasndo} SI UO!DAS SIU] Ayyjiqeureysns saaiauas je!20s pue yyeay 5 Qo Qo Qo o B o o 0 o 0 o 4O uol}e139jU! parosdu| “3 squal|a/sjuaied Joy aseo o Qo Qo o L Qo O/;O/0/0 Qo o JO uoeUIps009 panosduy| +} suonezijeydsoy pue o oO oO a Oo oO o o ag oO a quawyedap Aduasiawa ayetudoiddeu! Suisnpay = “a uonejuawajdui oO Oo Oo o Oo Oo obDlololoa a wesZ01d 104 sdiysiaujied Oo SAIJesOge||OD Bulaosdw] “pp (saaiasas |2190S “yyeay jesolneyag ‘jedipaw o o o oO o oO oO oO o oO oO “'3) waysAs auo ueY} oO as0W WO, Sa2Aas BUIAIaIa4 squaied/sjualj> SulAynuap| = "9 annoaya 1en2N aanney2 quauiwoy | Ajwenxa= or 6 8 “£ 9 =s§ v | & | @ T | son=0 vIN WweJZ01g youeasay Uoljen|enq pue sowoUucs] Yea Ae ate OE) Yyoseasay Aatjod YiJeaH JOJ JaqUaD VION Moday uoljenjeng jeuly aseD uossag BjoyM | JUawNsysu| AaAins JaUed °O xIpuaddy 095 | ("_ :Ajipads aseajd) 49410 0Z0Z Jaquiacag Jaye saljiaize JdM Buljsixa uleysns 0} Bulpuny jeuoippe Bulunsas F uejd Ayjiqeuleysns jewWJ0J e OJ UOIINGIJJUOD JO Jo UOI}eaID FY suoljeziuedio Jauyed Jayjo YM sBuljaaw BuluUeld Ayiqeuseysns [J Aqquy pea] ay} yim ssuijaaw Sujuuejd Apjiqeujeysns [J ‘Ajdde yeu} |e 399]aS) SuImo}|O} BYy3 JO Aue Ul payedioiwied aAey NOA J! a}ed!pul asea|q “ZT (Bulsnoy jou “2'1) spaau jedipauw-uou Oo oO o ao Oo Oo Jayjo squalp ssauppy ‘3 S@DIAIaS aalyoddns pue Bulsnoy Oo i Qo o oOo oO 0} ssa02e ualjo aseasou| +} sadiAsas Juswaseuew 2429 40 UO!}eUIps009 oO Oo Oo oO oO oO ued YUM squ -YBIY ASU-YBIY BPIACId “8 SODIAIAS OF ssao0e pue Uo!}eUIpso0o oO a Oo Qo Oo 0 aued aseasu| ‘Pp suieys pue U01393)/09 eyep |e90] a gO Oo oO Oo Oo pue apim-wayshs aaoudw “2 quaijedu! pue Asuasiawa oO oO o o Oo o ayelidoiddeul aanpay -q uolesoge|joo pue uoljes3aju! |290) o oO Oo o Oo Oo pue apim-wayshs aseauu) “e paniwwo5 payiwwo05 quawiwio) v parywwos = ¢ Zz v/N Ajawiasyxq = ¢ 2 38 ION = T WweJZold youeasay uoNen|enq pue so1woU0s] yea yoseasay Adljog Y22H JOJ 493Ua VION ra deyaer eT Ta] baclomy UDLUNJ}SU] AaAuNns JaULe :O XIpuaddy | woday uoenjenq jeuly aseD UOSJad BJOUM :(sdiysuauyed ‘sjapow uoljeulpsood aed ‘aunyonujsedjul Sueys eyep “3'a) aAoge paylejap saigajzejs |Aj|y-|2D 24} JO} ssaulpeas InOA padeys yey} sjuawaja Ajuap! asea|d “eyT (Aqiqua ased paseuew payses}uod aj/8uls e ul yeay | uonediniqed JeJo pue ‘yyeay jesoineyaq ‘jea1sAyd ayes3a}U! Ajjn} OF JUazU!) Ue|q UO!}e439zY] ||NJ yueaH [eso1neyag juawaseue YyyeaH Uolejndog (S11) sa91das Jo nal|-uj (IND3) JUaWaseUeW| a1eD padueyUR (ot -0 JO a]e9S) ssaulpeay urewog “V/N 21M aseajd ‘jesodoud \l'y-|eD JNOA 0} ajqeaijdde jou si ulewop Jejnsied e 4] “suleWOp SuImo}|oJ asau} Jo Yea UI SalIAIIIe/salZajess W\|y-|eD pauue|d padeys arey Aew Jey} JdM JO SjuaWwala Aj; UAap! pue SUIEWOP |A|¥-1VD SuIMoO}|OJ ay} U! ajed!9IWed 0 ssaulpea sue|d UNOA AjUAP! asea|q “ET {,,ueld yyjeay,, se paynuapl jl suonsenb esey} Jamsue Ajuo) Ajuo suejg aie) poseuey] 104 WeJZ01g youeasay Uolenjenq pue sowoUu0s] Yea Ae ate OE) Yyoseasay Aatjod YiJeaH JOJ JaqUaD VION Moday uolenjeng |e! sed UOSJad BJOYM | JUaWNJIsU] AaAINS JaULed :O XIpuaddy Ea ( :Ajlpads aseajd) sayi0 juawaseuew yyeay uoljeindog [] SOdIAJaS JO Nal] Ul sJaplAosd JUaWaseueW| ase pasueyug F] (Ajdde yey |e y2eyD) :suleWwop WIV-Tv9 BuIMO}||Oj 34} JO Aue JoJ Suaued AdUaZe AyUNOD 49Y}0 JO $3] DdM Aue YUM 3924]U0D 0} Sued |eLUIOJ BAeY NOA Op ‘aWI} SIU} IY “OT :(UoleJOge}/09 0} Sualeg ‘UOlJesOge]/09 pazezyioej Jey) Su0}de4 “B'a) Suljes ulejdxa Ayjaiug aseald “eT o oOo o o o (ey o oO Oo o o wonesoge}|o2 Jo sjana| 1eanan uonesogeljoa quawwoD y3iy Ajawasyxa = OT & c Z 9 =s Y G c u oN =0 (uoijesogel}oo yo sjana| YBly AjawiasyxXg = OT pue UO!JeJOge||09 ON = 0 2484M ‘OT 01.0 JO 8129S) EWIV-TVD 404 Suluue|ld ul (37 JOU “3"!) suaUed gM 4J94IO YM pajesoge|joo NOA aaey jUAa}xAa JeYM O| “ST o oO o o Oo o o oO Oo o o uoneroge}joo Jo sjanay 1eaynan uonesoge}jo2 quawwo) yay Ajauia.yx3 = OT 6 8 “ 9 2g v & z Tv ON=0 (uoNesoge||od Jo sjana] ysiy Ajawasx9 payesoqe||oo nod aney JUa}xa JEYM OL “YT. = OT pue uoreJoge||o2 ON = 0 a4aYM ‘OT 03 0 JO a[e9S) éWIV-TV9 404 Suluue|d Ul S37 YP WeJZo/d youeasay UoNen|eng pue so1woUcs] yea yoseasay Adijog Y22H JOJ 493Ua VION ra deyder eT Ia] fclomy JuawuNs}su| Aanuns JauLed :O xipuaddy | woday uonenjeng jeuly aseD UOSJad BJOUM a4ay apnjoul aseajd ‘sjuauodwod weisol1d JqM Aay Jo Ayjiqeuleysns 03 pajzejas syyZnouj jeuolippe Aue aaey NOA 4 ‘gt “WIv-Tv2 Suuawajdu) Joj saiaje4}s JO/pue [Al|\¥-TWD Ul UoNedioIWed s,uolj}ez|UeZIO ANOA aduUAN|JU! ACW JeY} JdM WO SJUdWA|a AjIQUAP! aSea|q */T steuyled [je 104 “sdauued Aduage A\UNOD Ja4}0 JO $3] YIM 3uNN9e4]U09 0} salJeg/suoseal ay ale JeYM ‘OU J] “GOT “YUM yeI]UOD AeW NOA saljUNOd Jo saizjUZ peasy YyoIyM Ajijuap! aseajd ‘saA 4] “eQT WweJZ01g youeasay Uolen|enq pue so1woU0s] Yea GA OS) yoseasay Aaljod YiJeaH 4OJ JajuaD YIN Moday uoljenjeng jeuly aseD uossag BjoyM | JUawWNsysu| AaAINs JaUed °O xIpuaddy Ea *(quaw||OJUa JO/pue ‘juaWasesua ‘UO!ed4I]UAapI 10} sassaz0Jd ‘saunpasojd pue saoijod suijjeis ed unoA payajje sey ¥231G1NO ET-PIAOD Moy pue JaYyJaYM aqiosap Ayaliq asea|d ‘OZ ‘SadIAJaS 0} SABUeYD “3"3) JdM UI UO!ed!: :6T-PIAOD 0} puodsas 0} Ayjige s,uoljeziuedso JNOA uo }Je3s JO ‘sadiAJas ‘asnyonsysesjul ‘diyssauped JM Jo syeduu! aaiyisod Aue aqiasap Ayalsg asea|d ‘ET IdM UO 6T-CIAOD Jo edu] WweJZo/d youeasay uoNen|eng pue so1woUcs] yea youeasay Aaijog Y3]2aH 404 Ja}UaD VION radey ae Tee otTa) icf) JUaLUNJ}sU| AaAuNs JaULed °O xipuaddy | woday uoenjenq jeuly aseD UOSJad BJOUM IAJAYNS AHL ONILAIdWIOD YO4 NOA ANVHL ‘ajqeoijdde 10u 41 ¥/N aJOUAp asea|q gMOUY 0} SN OJ JUETOMUUI SI YUIY) NOA JEU) payse },UaAeY am BulyyAue aay} S| “EZ éDdM Ul Buljedioiyed Aq pases uoljezIUeZIO ANOA SaSUaTEYD peoig ay} 1am JeYUM ‘ZZ é0dM Ul Bunediniwed Aq pasualiadxe uoljeziUes10 JNOA S}lyauaq Peoig ay} a19M JEUM “TZ aoualadxq dM ||P48A0 UO sUysNoY| BulpnjauoD WeJZ01g youeasay Uolen|enq pue so}woUdsg YyeaH Bate GRRE Yyoseasay Aatiod YIJeaH JOJ JajUaD YIN erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Appendix R: General Glossary Acronym Definition WPC Whole Person Care AHRQ Agency for Healthcare Research and Quality AOD Alcohol and other drugs BAA Business Associate Agreement BHS Behavioral Health Services Cal-AIM California Advancing and Innovating Medi-Cal CBP Controlling Blood Pressure CBP-18-59 Enrollees 18-59 years of age whose BP was <140/90 mm Hg CBP-60-85-D | Enrollees 60-85 years of age with a diagnosis of diabetes whose BP was <140/90 mm Hg Enrollees 60-85 years of age without a diagnosis of diabetes whose BP was <150/90 CBP-60-85-ND |mm Hg ccP Comprehensive Care Plan CCP-A Comprehensive care plan within enrollees’ anniversary of enrollment CCP-E Comprehensive care plan within 30 days of enrollment CDC Comprehensive Diabetes Care CE Coordinated Entry CFR Code of Federal Regulations CHR Community health reccord CHW Community health workers cms Centers for Medicare & Medicaid Services Coc Continuum of Care cs Community Supports DD Difference-in-Difference DHCS California Department of Health Care Services DJI Decrease Jail Incarcerations ECM Enhanced Care Management ED Emergency department EHR Electronic health record Appendix R: General Glossary | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research errr) Health Economics and Evaluation Research Program EMR Electronic medical record FEMA Federal Emergency Management Agency FFS Fee-for-Service FQHC Federally Qualified Health Center HbA1C Hemoglobin A1c HIE Health information exchange HIPAA Health Insurance Portability and Accountability Act HMIS Homeless Management Information System HR At high risk for various reasons HS Housing Services HUD Housing and Urban Development LE Lead Entity MAT Medication-assisted treatment MC/HR Enrollees with multiple chronic conditions or at high risk McP. Medi-Cal managed care plans MDD Major Depressive Disorder MOU Memorandum of Understanding National Quality Forum for Children Who Receive Effective Care Coordination of NQF 0719 Healthcare Services When Needed OBH Overall Beneficiary Health OBH-O Enrollees’ Overall Health OBH-E Enrollees’ Emotional/Mental Health OUD Opioid Use Disorder PDSA Plan, do, study, act PHI Protected health information PMPM Per-member-per-month P40 Pay for outcomes P4R Pay for reporting SCC Small County Collaborative SCWPCC Small County Whole Person Care Collaborative SMI Serious mental illness SMI/SUD/HML Enrollees with serious mental illness (SMI), substance use disorders (SUD), or experiencing homelessness Whole Person Care Final Evaluation Report | Appendix R: General Glossary erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program SUD Substance use disorder TA Technical Assistance VI-SPDAT Vulnerability Index — Service Prioritization Decision Assistance Tool Appendix R: General Glossary | Whole Person Care Final Evaluation Report UCLA Center for Health Policy Research lyr) Health Economics and Evaluation Research Program Appendix S: Enrollee Demographics, Health Status, and Prior Health Care Utilization by Target Population WPC Enrollee Characteristics by Target Population WPC Pilots were required to “receive support to integrate care for a particularly vulnerable group of Medi-Cal beneficiaries who have been identified as high users of multiple systems and continue to have poor health outcomes.” This appendix further examines the following evaluation question, “What were the demographics of pilot enrollees?” by examining characteristics of WPC enrollees by target population. The data sources included Medi-Cal enrollment and claims data between January 2015 and December 2021 and WPC Quarterly Enrollment and Utilization Reports from PY 2 to PY 6 (2017 through 2021). Of the 247,887 total WPC enrollees during program implementation, 228,680 enrollees that had an assigned target population and Medi-Cal enrollment and claims data. The prevalence of chronic conditions was identified using the CMS Chronic Conditions Data Warehouse for WPC enrollees with Medi-Cal claims data, using the primary and secondary diagnosis at each encounter. UCLA calculated standardized rates of utilization to account for variations in length of enrollment in Medi-Cal or size of the population in a given target population and to facilitate comparisons across analytic groups. Utilization was calculated per 1,000 full-scope Medi-Cal member months for six-month intervals in the two years prior to an enrollees’ first WPC enrollment date. Age was time-variant and was identified at the time of WPC enrollment. Time-invariant demographics such as race/ethnicity were identified using the most frequently reported value in enrollment data during the 24 months prior to enrollment into the program. Health status was measured as the presence of a condition at any point within 24 months prior to enrollment. Whole Person Care Final Evaluation Report | Appendix S: Enrollee Demographics, Health Status, and [RE{a} Prior Health Care Utilization by Target Population erry UCLA Center for Health Policy Research Health Economics and Evaluation Research Program Demographics Exhibit 241: Demographics of WPC Enrollees by Target Population, Prior to WPC Enrollment 2 3 8 ie aa 2 ot = 3 5 23 ee .3 3 5 a a 4 o <2 es a S E = Pee ee $s 5 = z 3 #2 |&8 3E /8 Enrollment _| N 126,054 | 119,911 | 50,122 | 45,121 | 22,593 | 50,366 _| 34,580 Age at % 0-17 1% 1% <1% 1% <1% <1% 7% enrollment | 9 18-34 33% 28% 31% 32% 30% 39% 24% % 35-49 27% 30% 30% 31% 30% 32% 26% % 50-64 31% 34% 33% 33% 33% 25% 32% % 65+ 8% 6% 6% 4% 6% 4% 11% Gender % male 52% 64% 61% 64% 60% 69% 56% Race/ % White 25% 28% 28% 31% 30% 23% 21% Ethnicity | 9% Hispanic 28% 25% 31% 34% 36% 34% 20% % Black 24% 28% 26% 21% 18% 32% 23% % Asian 1% 1% 1% 1% 1% <1% 1% % American Indian or Alaskan Native | 4% 2% 2% 2% 3% 1% 10% % Hawaiian or Other Pacific slander 2% 1% 1% 1% 1% 1% 2% % Other 10% 7% 3% 3% 4% 2% 18% % Unknown 6% 7% 8% 6% 7% 7% 6% Primary % English 84% 92% 93% 92% 90% 95% 81% Communi- | % Spanish 11% 6% 5% 6% 1% 4% 10% ation % Other Language 5% 2% 2% 1% 3% 1% 9% Homelessn | Identified as ess homeless by Pilots 41% 67% 66% 65% 58% 69% 41% Source: Medi-Cal enrollment data from January 2015 to December 2021 and Whole Person Care Quarterly Enrollment and Utilization Reports, January 2017-December 2021. Notes: Enrollee population includes 228,680 enrollees who were enrolled during PY 2 through PY 6 and had Medi- Cal enrollment data and at least one target population. All data except for homelessness are reported using Medi- Cal enrollment data during the 24 months prior to WPC enrollment. Homelessness was based on a Pilot-reported indicator collected at enrollment. Enrollees may be reported in more than one target population by Pilots. SMI/SUD is serious mental illness and/or substance use disorder. (0) Appendix S: Enrollee Demographics, Health Status, and Prior Health Care Utilization by Target Population | Whole Person Care Final Evaluation Report Health Status UCLA Center for Health Policy Research Health Economics and Evaluation Research Program December 2022 Exhibit 242: Most Frequent Chronic Conditions Among WPC Enrollees by Target Population, 24 Months Prior to WPC Enrollment 2 g 2 |, s§ | 2 = 3 S 28 |e 1.8 | 2 a a 2 oO <# ge s (5 | |25 |25 | 22 |¢ Ts = a ) uonelpaway ewyysy| $4a}Ua) Sulsaqos| aydsay jerpaw/aie) aaesadnsay| D Sao|Mas Uo}jeSIAeN/UO}ISUes) SUISNOH| sysodaq Suisnoy| Sad|AJag Sujuleysns pue Aoueua| Sulsnop| suoieydepy Ayjiqissaosy jeyuawuodAUy Sarina: Ayuno9 sunedianied IdM (DdM Y8nosy3 pasajyo you) sariAsas syoddns AyunWWO) MAN | IdM Y3NO1Y) pasayjo suoddns AyuNWWOD Ayunoy Aq syioddns AyUNWWOD U! $}O}Id JdM JO UO!ediaiwed, :97Z UW!YXA Ayunoz Aq s8uliajjo yoddns Aylunwwos anisuayaidwos :| xipuaddy Ree RULES WeJZ01g youeasay Uolen|enq pue so}wioUdsg Yea Yyoseasay Aotjod YiJeaH 40J JajUa VION yoday uoenjeag Jeulj asep Uossag BJOYM | AjUNOD Aq sBulsaJJO YWoOddns AyUNWWOD aAisuayasdWwos :| xipuaddy PRyAs “aWOY ay} U! BdUapUadapU! J3}e943 YIM UO!IUNJ 0} jen awoy e 0} suoieydepe jesisAyd ase (suolze: |PU! BY} ajqeua JO ‘jenpl. POA] BWOH “3’a) suoieydepy A: ayy Jo Ayayes pue ‘auesjam ‘yzjeay ayy aunsua 0} AJessazau ale yey? saody JeyUaWUOIIAUA SPINS Adljod WO Dg AWUNWWOD SDHC U! peuljep sy “JdM Japun aaiadas e AjUO Saje21pU! + “IdM JApUN a2!AUas e OS] SaxeIIPU! , ‘IDJ JapUN arIAJaS Sajed!pU! A :S9}0N, “"%Z07 AeW ‘sadiasas asedyyeay jo juawyedag elo, D) 0} pazwgns ‘ue|g aueD paseuew [eD-!pay| Aq sjeayspeaids ui SURLL WIlv-12D :@24un0S OdM ysnowyL ~ - 7 - - - - - | %tZ | %29 | %S9 | %58 | %6z | %OOT paiay¥O OYM SD YBNOwYL QIIAIS BUIIaJJO JUBIIAg TT z € i S € € Z Z st [oe |oe [er |S adIAlas $ BuLLayo 4aquinN J al af af a/® + eunqua,, t tf I 7 LZ tf xf c ewouos x + tf SS sh lah [+ eqseys ty af af «/ af znip eques g Be SS pe ZL DS + af - af x/ + eye] eques t ial os xf xl xf xl oajeW ues i te x/ al af we aS + uinbeor ues ft + al + + + + oasiaues4 ues - t if tt f I al af xf I xf o8aig ues i + af Ly ie oulpseusag ues z ° = = ¢ ° z g 2 z = z 3 s 3 a a 3 3 > g |e £ . 2 = + 9 = = = a eS s s 2 & g, 3 ¢ 2 ® s =< =z 2 S s = g g) 2) 3) | a] §] 2/8] @] 4] e) 3 3 2 8. wv Zz 2 ® 9 a = im 9 = 2 2 2 ¢ g = a 3 2 3 3 $ 3 3 ° o a 3 os 3 s 2 > a. a 2g a = 2 se 3 $ a 2. a s ° ee 2 2 S F/ g/ g/ 3/ =| 2] 3] &] *] go] S| S| #] ® 8) z| S$] #| s| F] gg] & $| $/ 8 8 3 SI 2 Pe cc = 3 3 = 2 a g B) #| ¢| Fi 3) | # z/ 3| £ S Bg | 7) gs] ° =| ¢| & = gs | 3 . g| 8| & a =x g — s o zB 2 z e| 2) 5 iz a ey <= = a 8 & e s e a Ayunod Bunedinnied IdM (dM Y8n01y3 paajyo you) sariAsas syioddng AyunwWwo0} MaN IdM Y8n0143 pasayo syoddns Ayunwiwoz WweJZo/d youeasay uoNen|enq pue so}woU0s] Yea yoseasay Arlo YajeaH 404 JaqUa YIN BET eLTal Ayunop Aq VHA S8Ul2Jjj0 Woddns AyunWWoD anlsuayasdwiod :| xipuaddy | woday uolenjenq jeuly ase) UOSUadg BJOUM |!ef Jo JuaWedap AdUuaSawa 94} 0} payiodsued} aq asimsayjo pjnom pue (s8nip Jayjo JO/pue JoYyos|e 0} anp) payer!xoyU! Ajaijqnd aq 0} punoj aie OYM sjenplAipu! JO} SuO!JeU!IsSap aAI}eUaIe ade SJaqjuad Sulaqos “juawuo4!Aua BuIAl| ajqejsun ue Aq pajyeqiazexa aq pjnom uO!jIpUuOd asoym pue (SUO!TIPUOD YIeaY |e1OIAeYaq Sulpnjaul) ssauj|! 40 Aunfur ue Woy [2dY 0} padu []!3s IN ‘UO!ezI/e}1dsoY aujnbed 4aSuo] OU OYM S|eNpIAIPU! 4OJ a4ed je;JUapISEJ W49}-WOYs S| adsay |e21payi/aseD ealjesadnday ‘pasndas s| Sulsnoy aduo Adueua} ajqejs pue ayes Sululeyulew ainsua sadiAJag Sujuleysns pue Adueua| Sulsnoy ‘pseog pue WOOL a}Nj!}sUOd JOU Op jeu} Pjoyasnoy aIseq 8 ysijqeysa 0} uosiad e ajqeua 0} Alessadau SU0!}e91J|POW PU SAdIAJaS AWI}-9UO BU!pUN{ JO ‘BulINdas ‘Su!}eUIpsOOd ‘BulAy!juap! YIM Ysisse syisodaq Sulsnoy WweJZ01g youeasay Uolen|enq pue so}WoU0sg YyeaH ENE EGRESS) yoseasay Aotjod YiJeaH 404 JajUaD YIN UCLA CENTER FOR HEALTH POLICY RESEARCH HEALTH ECONOMICS AND EVALUATION RESEARCH Appendix U: Pilot Specific Case Studies UCLA CENTER FOR ®*. HEALTH POLICY RESEARCH. e fit Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report UCLA CENTER FOR HEALTH POLICY RESEARCH , «5 as Ait Alameda’s Whole Person Care Pilot PILOT STRUCTURE Target Populations Alameda’s primary populations of focus included high utilizers, defined as those contacting two or more systems (e.g., medical crisis, high acuity care, mental health, substance use treatment, criminal justice) at any point in the last year, and individuals experiencing homelessness. Lead Entity and Partnerships In Alameda, the county Health Care Services Agency (HCSA) served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included 44 partners from diverse sectors, 19 of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. Partnerships between the LE and the local Medi-Cal Managed Care Plans; health centers; hospital systems; city and other county departments; and community- based organizations providing housing, social services, and other services were longstanding and facilitated care coordination services during WPC. Relationships with many other entities were new and developed to facilitate housing and supportive services (e.g., emergency shelter, long-term housing, tenancy support, outreach) for WPC enrollees. All care coordination services were contracted out to community partners. Data Sharing Infrastructure The key mechanism for data sharing with partners was a Community Health Record (CHR) that consolidated client data and was accessible by partners once a data-sharing agreement was established. Initially, only clinical entities covered by the Health Insurance Portability and Accountability Act (HIPAA) were able to participate, but later agencies providing support for social determinants of health were added to the network with additional restrictions as required by law. The CHR was powered by a Social Health Information Exchange (SHIE) platform that integrated data from many sources, including the public hospital and clinic system’s electronic health record (e.g., Epic), behavioral health, the housing system’s Homeless Management Information System (HMIS), county jail incarceration information, and many others. The platform was used in conjunction with PowerBl for data extraction and reporting across partners and to the state. Strengths included the ability to share data with care coordinators and care managers about consumers’ utilization and care team members from outside of their organization to enable a more whole person approach, automating the display of such data as much as possible to limit double data entry. Platform data were also available to many medical providers, community partners, and care team members regardless of their access to Epic. Limitations included ensuring client comfort with- and consent to- sharing information with partners not covered by HIPAA; integrating with non-clinical partners using HMIS; and consistent technical assistance and program guidance required to demonstrate platform value, and support implementation and adoption. Whole Person Care Final Evaluation Report |Appendix U: Pilot Specific Case Studies PILOT IMPLEMENTATION Pilot Enrollment Alameda enrolled 30,722 beneficiaries by the end of December 2021 using an auto-enroll and opt-out process. The average length of enrollment was 20.2 months. Approximately 19% of enrollees ever disenrolled at some point between January 2017 and December 2021. The most common reason for disenrollment was ineligibility for Medi-Cal (19% of total WPC enrollment).+ Enrollee Identification and Engagement Alameda primarily identified and auto-enrolled eligible enrollees through a data-driven approach (e.g., using risk factors and utilization statistics), although referrals and enrollments from Street Health Outreach teams that visited encampments, community partners, and medical providers were also accepted. Prior to enrollment, case managers dedicated time to build trust with individuals, identify basic barriers to services that could be addressed (e.g., transportation), and delineated client goals. Initial outreach and ongoing engagement were more challenging among those without consistent contact information (e.g., phone number or housing location), but case managers tracked hospital and emergency department admissions information to connect with these clients when they interacted with the county system. Outreach and engagement efforts and materials were informed by client input, and many were carried out by peer and community health workers with lived experience. Care Coordination Enrollees in care management were assigned to a care coordination team led by a clinically trained supervisor. Individuals were enrolled in healthcare/medical care management or housing-focused care coordination service bundles depending on their needs and the level of complexity of their medical, behavioral, and social challenges. Care coordination and outreach teams included community health workers or other staff with lived experience, nurses, licensed social workers, and housing navigators. Physicians, nurse practitioners, and clinical psychologists were also engaged in clinical consults and supervision. A single, dedicated care coordinator followed enrollees across all WPC-participating care settings. Most teams had a 1:15 case ratio. Individuals with a history of serious mental illness or experiencing homelessness received a more intensive tier of case management. Care coordination was supported by multidisciplinary team meetings and case conferences to promote team-based care and collaborative care delivery. Comprehensive assessment of all identified social needs for those enrollees in the care management bundle was conducted under the guidance of the Medi-Cal managed care plans who were administering the service bundle. Housing Assistance The Pilot emphasized a “Housing First” model that prioritized finding housing quickly with the expectation that other basic medical and social needs would be addressed more effectively with a stable living environment. WPC funds were used to assist with providing medical respite, conducting street outreach, housing navigation, tenancy support, completing screenings and assessments for the coordinated entry system (CES), housing searches, and to obtain housing funds. Housing funds were used for security deposits, basic home essentials, utilities, necessary housing improvements, legal support, landlord incentives, and ongoing assistance with enrollee-landlord relationships. 1 Beneficiary could no longer remain enrolled in WPC if no longer eligible for Medicaid benefits. Ei:{0) Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report Other Services In addition to care coordination and housing services, Alameda also provided respite services, alternative forms of crisis response, benefits advocacy, employment support, and wellness events. CRITICAL SUCCESS FACTORS ¢ Creation of a customized, consolidated data sharing infrastructure viewable through the community health record to support care managers to collaborate between the multidisciplinary partners involved in enrollee care coordination and case management. e The “Housing First” approach allowed the Pilot to house over 1,100 clients (as of November 2021) and provide the necessary social and medical supports to support the sustainability of those housing placements. e Focusing on trust-building with potential clients before enrollment supported outreach and engagement for enrollees with histories of mistreatment from and mistrust of health and social services systems. WPC enabled payment for this service and made it feasible. e Including peer and community health worker staff improved enrollee engagement and trust through identification with these individuals’ lived experience. ¢ Bringing together a “problem-solving learning community” of multidisciplinary partners from different sectors in person to connect with one another and learn how to access care in different systems, identify opportunities for collaboration, and develop understanding of each other’s point of view. This took a significant investment of time and planning but showed significant benefits. PERCEIVED IMPACT OF WPC Alameda perceived an above average impact on the following aspects of WPC Pilot implementation: ¢ Improved integration of health, behavioral health, and social services ¢ Reduction in inappropriate emergency department visits and hospitalizations e Increased data sharing between LE and WPC partners ® Improved management of care of high risk and high utilizing populations « Identifying enrollees receiving services from more than one system (e.g., medical, behavioral health, social services) e Improved collaborative partnerships for program implementation Alameda perceived an above average improvement on the following aspects of care for enrollees: © Coordination of care ¢ Comprehensiveness of and access to needed services (e.g., health, behavioral health, and/or social services) e Targeted identification, outreach/engagement, and enrollment “Really making sure that people from that community are included. It makes a difference when people are able to identify with someone, whether if they look like them, [or] they've had some shared experience. It's really important that the team actually knows the community, the layouts of where to go. Really build the connections ahead of time, because partners know when you don't know your stuff... it’s really the heart, it's like, you really got to find people that don't mind being in the trenches and don't mind not always seeing results right away. Yeah, being able to be that agent of hope is so important for our Whole Person Care Project.” Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies “Bringing folks from all of these different sectors into one room so that they could meet each other, put a face to the name, hear about each other's work, and be able to ask questions and start to see how their worlds intersect and could develop opportunities and follow-ups to collaborate on the care more efficiently. It takes a lot of time and a lot of regular planning to facilitate that well, when all these providers are really busy... there was a shift towards seeing that this is really a benefit and support of the work.” Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report UCLA CENTER FOR HEALTH POLICY RESEARCH Contra Costa’s Whole Person Care Pilot PILOT STRUCTURE Target Populations Contra Costa focused primarily on serving high utilizers, defined as those with the top 15% of utilization using a predictive risk model that drew on multiple sources of data, including emergency department, inpatient, outpatient, and specialty visit utilization. Lead Entity and Partnerships In Contra Costa, Contra Costa Health Services (CCHS) served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included 12 partners from diverse sectors, eight of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. Partnerships between the LE and primary care health systems were longstanding and facilitated data integration and the centralization of care coordination services during WPC. Relationships with many other entities, such as Lifelong Medical Care, La Clinica de la Raza, and the County Health, Housing, and Homeless Program (H3) were developed to deliver care management services and facilitate enrollee access to housing. Existing relationships with other county divisions, such as the Contra Costa Regional Medical Center (CCRMC) and Emergency Medical Services (EMS) were strengthened to improve emergency department enrollee discharge processes and workflows. Data Sharing Infrastructure All WPC documentation was visible to internal CCHS partners via a shared electronic health record (EHR), Epic. The primary mechanism for data sharing with external partners was a care management platform embedded within the EHR called “Care Everywhere,” which integrated data across county departments and affiliated health system partners. Strengths of Contra Costa’s data sharing infrastructure included integration with the county’s Behavioral Health division, inclusion of robust enrollee contact information, and data visualization features (e.g., case management dashboards to easily track enrollee status across large caseloads). Contra Costa’s information technology department worked closely with program staff to custom build many of the tools embedded into the EHR, tailored specifically for WPC. Additionally, they integrated the Homeless Management Information System (HMIS) system with Epic and BitFocus to exchange care team member information with shelters and integrate shelter information into the EHR. Limitations included difficulty integrating data from substance use programs, anticipating detention release dates, and a lack of expertise for how to translate social services agency data into actionable, public benefits workflows. PILOT IMPLEMENTATION Pilot Enrollment Contra Costa’s Pilot enrolled 57,190 beneficiaries by the end of December 2021. The average length of enrollment was 11.7 months. Approximately 92% of enrollees ever disenrolled at some point between Whole Person Care Final Evaluation Report |Appendix U: Pilot Specific Case Studies January 2017 and December 2021. The most common reasons for disenrollment were lack of engagement (44% of total WPC enrollment)? and beneficiary request (16%).? Enrollee Identification and Engagement Contra Costa’s Pilot organized program enrollment into three tiers with differential mechanisms for identifying enrollees. Tiers 1 and 2 used a predictive risk model to identify eligible Medicaid beneficiaries (defined as the top 15% of utilizers) and auto-enrolled members into the two tiers based on utilization rates. Tier 3 used referrals from partners at point-of-care to enroll individuals not identified through the predictive risk model, who required short-term, high-intensity diversion services. Case managers were assigned enrollees and given 60 days to engage the client in services telephonically. On average, three outreach attempts were made per enrollee, and enrollees not successfully engaged or who opted-out were disenrolled. Early in the Pilot, CCHS improved their risk model to emphasize past utilization over chronic conditions as a driver for ED visits and added the referral-based third tier. Care Coordination There were multi-disciplinary teams of homeless specialists, behavioral health specialists (often LCSWs), community health workers, public health nurses, substance use counselors, social workers, and eligibility specialists. Within the team, a person was assigned to a case manager based on their specific need (e.g., housing) and could be moved to a different case manager within the team if that need changed over time. This was consistent across tiers. Through the shared EHR, all teams had direct access to the multidisciplinary team, with the case manager clearly identified. Enrollees were assigned toa single, dedicated case manager who followed the enrollee across all WPC-participating care settings. Tier 1 was highest acuity, primarily field-based case management with a 1:80 case ratio. Tier 2 was moderate acuity, receiving primarily telephonic support by community health workers with a 1:300 case ratio. Tier 3 was short-term and high-intensity case management focused on ED and inpatient hospital diversion with a 1:25 case ratio. Care coordination was supported by integrated data systems with most partners and health systems, and multidisciplinary team meetings with case presentation to promote team-based care and collaborative care delivery. Comprehensive assessment of all identified social needs was conducted using a homegrown motivational interviewing-based screening tool. Housing Assistance The Pilot emphasized a “Housing First” model. WPC funds were used to assist with housing navigation (e.g., find available temporary or permanent housing stock), tenancy support (e.g., counseling and training individuals to move in or remain in temporary or permanent housing), completing applications for the Coordinated Entry System (CES), and in obtaining housing funds (e.g., housing choice vouchers or rental subsidies). Additionally, the WPC staffing model included homeless services specialist case managers who specialized in supporting clients with housing navigation and housing tenancy services (employed by CCHS). The Pilot co-located social workers, eligibility workers, and In-Home Supportive Services (IHSS) workers. These staff were part of Employment and Human Services (EHSD), but sat within the WPC department at CCHS, allowing them to access both health and human eligibility services systems. 2 Beneficiary refused to participate or did not engage in services. 3 Beneficiary requested disenrollment from WPC. Ey:2) Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report Other Services In addition to care coordination and housing services, Contra Costa’s Pilot also provided transportation, active referrals to legal services/assistance, educational programs, one-on-one job coaching, and connection to public benefits services. CRITICAL SUCCESS FACTORS e Cross-sector integration of the EHR, including bi-directional data sharing agreements with seven county departments and external partner agencies, allowed for quick and real-time access to needed data. e Use of a risk-based algorithm to identify and auto-enroll Medi-Cal beneficiaries enabled large overall client caseloads and data-driven prioritization of staff resources. A dedicated “business intelligence” team facilitated development, maintaining, and automating the integrated data infrastructure both internally and with external partners. e Tiered care coordination model based on need, with all three tiers including a core set of services (e.g., social needs assessment, benefits renewals, and referrals) and more acute enrollees receiving additional in-person coordination services. e Data-driven quality improvement and internal evaluation efforts informed programmatic changes early in the Pilot that enabled efficient use of resources and broader client reach, facilitating internal evaluation of the Pilot’s impact. e Enhanced collaboration with social services, as the Pilot directly employed social workers, eligibility specialists, and IHSS social workers, which created a greater degree of access to these services and broad collaboration between the departments to improve workflows, data flow, and efficiencies. PERCEIVED IMPACT OF WPC Contra Costa perceived an above average impact on the following aspects of WPC Pilot implementation: ¢ Improved integration of health, behavioral health, and social services Improved care quality e Reduced inappropriate emergency department visits and hospitalizations « Increased data sharing between LE and WPC partners ¢ Improved management of care of high risk and high utilizing populations e Identifying clients/patients receiving services from more than one system (e.g., medical, behavioral health, social services) «Improved collaborative partnerships for program implementation Contra Costa perceived an above average improvement on the following aspects of care for enrollees: ¢ Coordination and continuity of care e@ Frequency and quality of communication with enrollee e Extent to which care provided is patient-centered * Overall enrollee well-being Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies “| think some of the key things were definitely the length of our program. So many case management programs are kind of these short three-month case management and having a case management model that was built on a long program. One year was essentially our minimum length of enrollment with people being able to extend that even further. And so, | think this was a real testament that it takes a long time to kind of make impacts on people's lives to kind of change their behavior patterns in terms of where people get care, how can you help them with various social needs? These are not short-term interventions.” “Where we placed our Whole Person Care Pilot made a huge impact, having it based in public health inside the integrated health system at Contra Costa. It’s a unique model for a county-run health system. But it's really like we put this in the heart of the system of the group that is in the community and is also in the health centers and has those existing relationships. But to build it in-house versus contract and sort of patched together, | think it really solidified that network and allowed us to do so many of these other things like the data projects and the evaluation and so forth. But | think having it fully in-house right there at the center and committing to that, not sort of piecing it out through the years really contributed.” Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report UCLA CENTER FOR HEALTH POLICY RESEARCH.» it Kern’s Whole Person Care Pilot PILOT STRUCTURE Target Populations Kern focused on improving transitions of care for recently incarcerated individuals; primary populations of focus included those experiencing homelessness or at-risk of homelessness and those who were high utilizers of care. Lead Entity and Partnerships Kern Medical Center (KMC) served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included 15 partners from diverse sectors, 13 of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. The partnership between the LE and Kern County Sheriff's Office was longstanding and allowed for WPC presence within jails, facilitating identification, engagement, and enrollment. Strategic co-location of the County Department of Human Services staff within KMC clinics facilitated access to needed benefits for enrollees, whereas colocation of WPC staff within county shelters allowed for warm handoff of eligible enrollees. Relationships with many community-based organizations (CBOs) (e.g., for homeless outreach) resulted from WPC. CBOs also recognized shared interest and mutual clients. Care coordination services were provided by Kern and not contracted out. Data Sharing Infrastructure The primary mechanism for data sharing with partners was “Healthy Care,” a care management platform within the existing Cerner electronic medical record (EMR). KMC transitioned to the Cerner EMR during the first years of WPC. The platform was accessible across all county departments, but not for CBOs. It provided a centralized data system for WPC and other similar programs within the county (e.g., Health Homes Program, Transitions of Care). Strengths included the ability to run tailored reports on WPC enrollees and provide direct access to jail and specialty medical records for WPC staff involved in care coordination. Its major limitation was inability for community partners to access data. PILOT IMPLEMENTATION Pilot Enrollment Kern enrolled 2,773 beneficiaries by the end of December 2021. The average length of enrollment was 24.4 months. Approximately 2% of enrollees ever disenrolled at some point between January 2017 and December 2021. The most common reason for disenrollment was beneficiary request (1% of total WPC enrollment).* 4 Beneficiary requested disenrollment from WPC. Whole Person Care Final Evaluation Report |Appendix U: Pilot Specific Case Studies Enrollee Identification and Engagement Enrollees were identified through targeted outreach lists which included names of those recently released from county jails. Direct referrals from CBOs were also utilized but viewed as less useful due to enrollee transience, which complicated follow-up after initial contact. Kern maintained a presence in shelters for continuous outreach and engagement. Co-location and the use of a peer support specialist (i.e., ability to build trust and rapport with people experiencing homelessness based on lived experience) were strategies identified as fundamental to successful engagement. Care Coordination Care coordination services were provided by medical assistants (MAs) functioning as enrollment specialists and care coordinators. MAs were supported by a team including physician champions (working in street medicine, outpatient clinics, and correctional health settings), a social worker, a nurse practitioner, a health educator, and a PharmD. A county human services worker was also co-located in the clinic setting to provide benefits assistance support. Kern aimed to provide a “one-stop” shop for enrollee needs (i.e., largely co-located in a single space). Most teams had a 1:125 case ratio. Care coordination was supported by multidisciplinary team meetings and case conferences to promote team-based care and collaborative care delivery. Comprehensive assessment of all identified social needs was conducted using the Vulnerability Index — Service Prioritization Decision Assistance Tool (VI-SPDAT). Housing Assistance Kern emphasized a “Housing First” model. WPC funds were used to assist with tenancy support (e.g., counseling and training individuals to move in to or remain in temporary or permanent housing); completing applications for the Coordinated Entry System; conducting a housing search (e.g., find available temporary or permanent housing stock); and for providing funds for security deposits, utilities, housing improvements for health, and/or landlord incentives. Other Services In addition to care coordination and housing services, Kern also provided medical respite, health and wellness courses, and sobering centers. CRITICAL SUCCESS FACTORS e Existing, unique integration of KMC and the county jail system, with a dedicated champion. KMC provided medical care for those in jail as a means of WPC enrollment prior to release from jail. Furthermore, the WPC medical director also served as a correctional physician and was present in the jails three days a week, offering continuity of care and serving as a bridge for enrollees as they transitioned from correctional medicine to the outpatient setting. ¢ Co-location in shelters. This facilitated continuous outreach and engagement of enrollees, as well as ability to locate enrollees when needed. Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report County jail and specialty care shared a medical record system (Cerner) with Kern and had access to “Healthy Care,” the care management platform. The ability to see medical records and enrollee history over time supported a more comprehensive understanding of care needs. Placement of a Department of Human Services worker within clinics. This facilitated enrollee access to social services and benefits and helped with system navigation and ensuring linkage to appropriate resources within WPC. Provision of innovative health and wellness courses that promoted self-sufficiency for enrollees. Offered an opportunity to check-in with enrollees on a weekly basis and increase buy- in and interest in WPC by providing simple incentives, such as clothing, shoes, and food. Whole Person Care Final Evaluation Report |Appendix U: Pilot Specific Case Studies PERCEIVED IMPACT OF WPC Kern perceived an above average impact on the following aspects of WPC Pilot implementation: e Improved integration of health, behavioral health, and social services Improved care quality Kern perceived an above average improvement on the following aspects of care for enrollees: ¢ Coordination and continuity of care e Access to needed services (health, behavioral health, and/or social services) e Access to affordable housing * Comprehensiveness and timeliness of available services e Targeted identification, outreach/engagement, and enrollment Frequency and quality of communication with enrollees e Extent to which care provided was patient-centered ¢ Overall enrollee well-being “A lot of [potential WPC enrollees] are very skeptical. They have been in and out of the system...[asking] ‘What are you going to do for us that's any more help than any other entity that I've been referred to in the past that has failed me?’ ...we really do try to make sure that ... they are continuing to experience ... continuity that they never had before.” “| think what makes this program unique is that we're a one-stop shop for our patients. ...! hear from our patients... that is one of the reasons that they love this program is because they can get everything taken care of in one place.” Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report UCLA CENTER FOR HEALTH POLICY RESEARCH , «5 as it i ole Person Care Pilot PILOT STRUCTURE Target Populations Kings’ primary populations of focus included those with chronic physical conditions (diabetes and hypertension) and those with severe mental illness or substance abuse disorders. Kings also served high utilizers (defined as individuals with six or more emergency department visits in a year) and individuals experiencing homelessness or at risk of homelessness. Lead Entity and Partnerships In Kings, the Human Services Agency (HSA) served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2017, the Pilot included nine partners from diverse sectors, all of which were identified as having a high awareness of WPC and actively involved in implementing WPC. Partnership between the LE and Kings County Behavioral Health facilitated close and coordinated entry assessments to connect enrollees quickly and efficiently to services based on severity of need. Existing relationships were leveraged for WPC, and included Probation, Sheriff, and Public Health departments. Contracted services (through Kings View Mental Health Services and Champions Recovery Alternative Program Inc.) streamlined coordination for WPC target populations with acute mental health and substance use disorders. Data Sharing Infrastructure The primary mechanism for data sharing with partners was "Efforts to Outcomes" (ETO), which helped most WPC partners track and coordinate care activities including medical, behavioral health, and social services information. Strengths of ETO included the ability to track referrals, engagement, and key enrollee outcomes to measure performance. Limitations included non-universal partner adoption, limited enrollee-level notifications to partners who did adopt, and significant need for manual data entry. PILOT IMPLEMENTATION Pilot Enrollment Kings enrolled 1,037 beneficiaries by the end of December 2021. The average length of enrollment was 15.8 months. Approximately 50% of enrollees ever disenrolled at some point between January 2017 and December 2021. The most common reason for disenrollment was lack of engagement (38% of total WPC enrollment).> Enrollee Identification and Engagement Kings’ care coordination followed a multi-step process. Eligible enrollees were identified through field- based outreach by peer specialists (e.g., with lived experience similar to enrollees), community referrals, or programs such as Project Roomkey. 5 Beneficiary refused to participate or did not engage in services. Whole Person Care Final Evaluation Report |Appendix U: Pilot Specific Case Studies Care Coordination Enrollees were assigned to a single case manager who followed enrollees across all WPC-participating care settings. Case managers had varied backgrounds and experience (e.g., social work, substance abuse counseling, and on-the-job training). Case managers had access to a broader team that included a housing navigator, employment navigator, community health workers, and a Supplemental Security Income (SSI) advocate. Most case managers had a 1:30 case ratio. Care coordination was supported by multidisciplinary team meetings to promote collaborative care delivery. Comprehensive assessment of social needs was conducted using a standardized intake assessment, and those with mental or behavioral health needs were prioritized for enrollment and services. Housing Assistance Kings emphasized a “Housing First” model, and had a team of seven housing navigators that provided housing navigation, completed applications, and sourced additional housing-related funding. WPC funds were used to assist with security deposits, furnishings, and utilities. Other Services In addition to care coordination and housing services, Kings provided assistance with benefits applications, connection to sobering centers, transportation, and referrals to legal services. CRITICAL SUCCESS FACTORS « Field-based outreach enabled effective and direct communication with enrollees experiencing homelessness. Through a first-hand understanding of barriers to service engagement, Kings was able to provide transportation to screenings, and medical and/or housing appointments when needed. e Formalized workflows to engage individuals upon discharge from the hospital and jail system ensured continuity of care for high-risk populations. ¢ Despite COVID-19 disruptions to field-based outreach, Kings leveraged WPC partnerships to facilitate and streamline a referral process through Project Roomkey. e Data sharing platform, ETO, linked partners and care coordination staff to encourage communication within teams. PERCEIVED IMPACT OF WPC Kings perceived an above average impact on the following aspects of WPC Pilot implementation: ¢ Improved integration of health, behavioral health, and social services ¢ Decreased overall cost of care e Improved enrollee health and well-being « Identified enrollees receiving services from more than one system (e.g., medical, behavioral health, social services) e Improved collaborative partnerships for program implementation Liya) Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report Kings perceived an above average improvement on the following aspects of care for enrollees: © Coordination of care e Access to and comprehensiveness of needed services e Targeted identification, outreach/engagement, and enrollment =~ Frequency and quality of communication with enrollees e Extent to which care provided was patient-centered “My perception is that this program is extremely vital to the community. In the five years it ran, I've personally seen huge success in overcoming long time gaps in services, such as [services for] single homeless adults. ... Gradually through hard work and preservation of not only themselves, but the staff who was willing to go that extra mile and be there when no one else would, some were housed. Some were moved from [encampments] to shelters or room and boards, which | truly feel with this population is a huge, huge success.” “I'm going to sit there with them [enrollees]. I'm going to fill it out with them. I'm going to complete it thoroughly with them. If they're missing documents, I'm going to go to the bank, get their bank statement, whatever they need. Remove those fears, take them to their appointments, do the meet and greet, do the inspections, whatever we need to do to get to that end goal. That | think is what truly defines Whole Person Care. It's such a huge success because we've been there through to the end and we're ready when they are. That's what it's all about.” Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies UCLA CENTER FOR HEALTH POLICY RESEARCH at Los Angeles’ Whole Person Care Pilot PILOT STRUCTURE Target Populations Los Angeles provided services to all WPC target populations, including high utilizers of the medical system, individuals experiencing homelessness and at risk for homelessness, individuals experiencing chronic physical conditions, individuals experiencing severe mental illness/substance use disorders, and individuals involved in the justice system. High utilizers were defined as those with three or more emergency department visits or inpatient stays in the last 12 months. Lead Entity and Partnerships In Los Angeles, the Los Angeles County Department of Health Services (LACDHS) served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included 50 partners from diverse sectors, 45 of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. Partnerships between the LE and other county departments, such as the Department of Mental Health and Housing for Health, were longstanding and essential to supporting enrollee wellbeing during WPC. Relationships with many other entities, such as Medi-Cal Managed Care plan (MCPs), housing and homelessness services, and re-entry providers were new and developed to facilitate connections with enrollees experiencing high utilization patterns and those involved in the justice system. Care coordination services were delivered in-house, and also contracted out to community partners. Data Sharing Infrastructure The primary mechanism for data sharing with partners was “CHAMP,” a consolidated care management platform that tracked eligible enrollees, consent management, enrollee care plan goals and progress, and generated reports. Strengths included customization from the “ground up” with a vendor, ability to account for enrollees’ social determinants of health, and high community partner adoption of the platform allowing universal data access among many key stakeholders. Limitations included the platform’s reliance on case manager input after every enrollee interaction and limited communication with other electronic data management systems and medical records. This was especially challenging for hospital partners and led to communication challenges in some cases. CHAMP did evolve through several iterations as the limits of its’ functionality were tested and readjusted. PILOT IMPLEMENTATION Pilot Enrollment Los Angeles enrolled 76,107 unique beneficiaries by the end of December 2021. The average length of enrollment was 9.2 months. Approximately 79% of enrollees ever disenrolled at some point between January 2017 and December 2021. The most common reason for disenrollment was services no longer needed (45% of total WPC enrollment).° 6 Beneficiary was not appropriate or did not benefit from the services provided. 2) Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report Enrollee Identification and Engagement Identification and enrollment strategy varied by program but generally followed a point of care referral model. For example, in the “Transitions of Care” program, eligible enrollees admitted to the hospital were assigned to onsite case managers with lived experience who would share program information and enroll enrollees during their hospital stay. For other programs, eligible enrollees were referred from local hospitals, community partners, the Sheriff's department, and prison partnerships. Street and facility-based outreach efforts were also used. Community partnerships were key to identification and enrollment. Enrollees graduated from the program once they met or nearly met personal and/or program goals. Care Coordination Los Angeles had 16 programs, each working with a unique target population and offering a slightly different form of care coordination. In most programs, enrollees were assigned to a care coordination team of community health workers (CHWs) led by licensed clinical social workers (LCSWs). CHWs enrolled and engaged with enrollees to support connection to services, while LCSW supervisors supported care coordinators with troubleshooting, escalation of more complex enrollees, and general review of caseloads. Enrollees had different care coordinators based on care setting; care coordinators were responsible for communicating across care settings and coordinating warm hand-offs. The CHW allowed for enrollee-centered care, patient advocacy, social support, culturally appropriate health education, and linguistically and literacy appropriate communication. Most teams had an average 1:25 case ratio. Assessment of all identified social needs was conducted using a comprehensive needs assessment developed internally by a multi-disciplinary team; the development process included iterative plan-do-study-act (PDSA) cycles and focus group testing. This comprehensive needs assessment was a consolidated survey of validated screening tools across multiple domains with an emphasis on social risk factors; it was primarily administered to enrollees by CHWs. Identified needs then drove the development of the comprehensive care plan. Housing Assistance Los Angeles emphasized a “Housing First” model that conducted street outreach and supported enrollee connection to temporary or permanent housing. Lack of available housing presented a considerable barrier to these efforts and enrollees were triaged into various available programs within and outside of WPC depending on specific need. WPC funds were used to support WPC partner organizations assisting with tenancy support (e.g., counseling and training individuals to move in or remain in temporary or permanent housing). Other Services In addition to care coordination and housing services, Los Angeles offered assistance with benefits applications, employment training programs, sobering center services, medical respite, and transportation. Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies CRITICAL SUCCESS FACTORS Centering WPC care coordination with community health workers with lived experience due to their resourcefulness, ability to connect with enrollees, and strong advocacy for enrollee needs. Peer and supervisory support through team huddles and weekly case conferences improved care coordination and team collaboration. The consolidated CHAMP data infrastructure allowed care coordinators, supervisors, and community partners to access eligibility information, enrollee rosters, assessments, care plans, and necessary data reports across most service providers. Having an onsite presence in local hospitals supported partner relationship-building and familiarity with WPC services. Strong partnerships and refined workflows between the LE and its’ partners improved enrollee identification, enrollment, and engagement across multiple providers in the absence of a single care coordinator managing all enrollee services. Development and use of a single, universal consent form across multiple integrated programs improved consent management, allowed for consent-driven data sharing of protected data types including behavioral health data, and ensured protection of patient data while maximizing the potential for effective care coordination and communication. A dedicated capacity building team developed and delivered an extensive curriculum for initial and ongoing staff training programs in core competency areas (e.g., care planning and SMART goals, service linkage, navigating the health system) and professional development, while supporting infrastructure development (e.g., care management platform). The team standardized practices and utilized feedback from frontline workers to ensure meaningful outcomes. Utilized a regional care delivery model with representation across each of the eight service planning areas of Los Angeles, which allowed for locally informed care based on resources and population knowledge. PERCEIVED IMPACT OF WPC Los Angeles perceived a high level of impact on the following aspects of WPC Pilot implementation: Improved integration of health, behavioral health, and social services Collaborative partnerships for program implementation Improved coordination of care for enrollees Improved enrollee health and well-being Reduced inappropriate emergency department visits and hospitalizations Los Angeles perceived an above average improvement on the following aspects of care for enrollees: “I'm afraid that once this progra Coordination of care Access to needed services (health, behavioral health, and/or social services) Extent to which care provided was patient-centered is] no longer available that people are going to just be another case number. | don't think that they're going to get the same care that they need ... | don't think that some of the other programs take the time like we do to make sure and follow through that these things hap, That's my concern.” Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report “Working closely with the community and hospital partners in developing the most integrated approach that we can is where we've seen the most success. We're part of the team or workflows; we're coordinating well with the other care team members. It's that investment of really having that integrated approach and having champions...To identify who those are and work closely with them, to make sure everyone's needs are being met. Really building that relationship is key.” Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies UCLA CENTER FOR HEALTH POLICY RESEARCH Marin’s Whole Person Care Pilot PILOT STRUCTURE Target Populations Marin’s primary populations of focus included high utilizers and individuals at risk of or currently experiencing homelessness. High utilizers were defined as those with two or more emergency department and/or inpatient visits in the past year. Lead Entity and Partnerships In Marin, the Department of Health and Human Services served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included 40 partners from diverse sectors, 24 of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. Partnership between the LE and Marin Community Clinics (which included three Federally Qualified Health Centers, behavioral health agencies, and non-profit housing services organizations) were longstanding and provided additional complex case management capacity during WPC. Other longstanding partnerships such as Marin Housing Authority, Homeward Bound, and the Ritter Center facilitated housing and homelessness services for WPC enrollees. Relationships with many other entities were new, such as a partnership with the Marin County Sheriff's Office developed to facilitate engagement with Marin County’s justice-involved population. Data Sharing Infrastructure The primary mechanism for data sharing with partners was a care coordination technology platform branded locally as “WIZARD,” which helped Marin County communicate with organizations of various types. Strengths included near-universal adoption among partner agencies of the WIZARD platform, and secure messaging and alerting features compliant with the Health Insurance Portability and Accountability Act (HIPAA). Limitations included minimal medical provider interaction with the platform, frontline workers did not consider the platform to be particularly user-friendly, and inability for case managers to access medical records and partners’ systems. Another pillar of data sharing infrastructure was the enrollee’s release of information (ROI) and consent which allowed data sharing between a wide range of participating entities. This, in combination with data sharing agreements between WPC and external partners, provided the legal and policy foundation for cross-sector data collaboration. PILOT IMPLEMENTATION Pilot Enrollment Marin enrolled 1,881 beneficiaries by the end of December 2021. The average length of enrollment was 27.9 months. Enrollment was defined as having signed a valid ROI and a Medi-Cal client identification number (CIN). Approximately 76% of enrollees ever disenrolled at some point between January 2017 and December 2021. The most common reason for disenrollment was WPC services no longer needed (46%).” 7 Beneficiary was not appropriate or did not benefit from the services provided. ci: Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report Enrollee Identification and Engagement Marin utilized a two-prong approach to engage, enroll, and identify clients for the WPC Pilot including the usage of referrals from physicians and proactive outreach from the Homeless Outreach Team (HOT) or other partner entities. If referred through a physician, enrollees met with a complex care navigator supervisor and the medical case manager registered nurse. If enrolled through HOT or other agencies, enrollees were assigned to a case manager from one of the contracted case management partners. Through these case managers, clients were then connected to substance abuse, behavioral health, and housing resources. Care Coordination Enrollees were assigned to a case manager whose training depended on the needs of the enrollee. Teams also included complex care navigator supervisors who assisted case managers in reaching out to specialists from housing, behavioral health, substance abuse, and medical staff. An enrollee might have multiple care coordinators across WPC partners who actively communicated with one another. Most teams had a 1:17 (housing case management) or 1:30 (medical and mild to moderate behavioral health management) case ratio. Care coordination was supported by multidisciplinary team meetings to promote team-based care and collaborative care delivery. Comprehensive assessment of all identified social needs was conducted using the Vulnerability Index — Service Prioritization Decision Assistance Tool (VI-SPDAT) and a homegrown tool to assess socioeconomic needs. Housing Assistance Marin emphasized a “Housing First” model, centered around a housing locator who worked with the case manager, enrollee, and landlord to provide housing. WPC funds were used to assist with funds for security deposits and ongoing assistance with enrollee-landlord relationships. Other Services In addition to care coordination and housing services, Marin also assisted with benefits applications, provided one-on-one training to help secure employment, coordinated transportation, and provided access to needed legal services. CRITICAL SUCCESS FACTORS e Through its field-based and medical referral approach, Marin was able to effectively connect high utilizers of the healthcare system and those who were experiencing homelessness to services that supported clients to make progress on their goals. e Partner adoption of the WIZARD platform facilitated communication across sectors and organizations to coordinate care of vulnerable clients. ® Close coordination and relationships between the housing locator, case manager, and enrollee supported successful relationships with many landlords associated with the Section 8 housing program. e Implementing partnership agreements with over 40 organizations allowed Marin to offer enrollees connection to a wide variety of services. Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies PERCEIVED IMPACT OF WPC Marin perceived an above average impact on the following aspects of WPC Pilot implementation: ¢ Improved integration of health, behavioral health, and social services e Reduction in inappropriate emergency department visits and hospitalizations e Increased data sharing between LE and WPC partners ¢ Identifying enrollees receiving services from more than one system ¢ Improved collaborative partnerships for program implementation e Improved coordination of care for patients/clients Marin perceived an above average improvement on the following aspects of care for enrollees: e Access to affordable housing e Comprehensiveness and timeliness of available services (health, behavioral health, and/or social services) “| would say, our Pilot has been a factor or a crucial component in a lot of people experiencing homelessness becoming housed. And that's been a huge impact. And then one other piece is probably also connecting more providers of services for the most vulnerable people in the community. So just creating some infrastructure for a system that a lot of it was also there, but pretty disconnected.” “I'm just extremely proud of this group for... how person centered [they are]. They're just engaging the patient around what is it. Yes, we have the referral from the provider with what the provider thinks this patient needs, but really working with the patient... they set their goals together and they ask the patient what they want for the buy-in. And then they are accountable. | mean, [with] patients, there's a lot of trust building that happened. They return calls, they follow up on the things that they said they would. So, it is really person centered. And it sounds ridiculous that they don't really give up on patients... just sort of letting people take their time to think about it, to engage.” 609 | Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report UCLA CENTER FOR HEALTH POLICY RESEARCH , «5 as aif Mendocino’s Whole Person Care Pilot PILOT STRUCTURE Target Populations Mendocino’s primary populations of focus included those with severe mental illness, who met two of the following conditions: substance use disorder, high utilization, homelessness, and/or recent law enforcement engagement. Lead Entity and Partnerships In Mendocino, the Health and Human Services Agency (HHSA) served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included eight partners from diverse sectors, half of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. Partnerships between the LE and the Redwood Quality Management Company (RQMC) were longstanding, with RQMC serving as the lead contractor responsible for overseeing and subcontracting with community-based behavioral health services in the county, and later, as the sole entity responsible for employing and supervising the wellness coaches providing care coordination under WPC. Data Sharing Infrastructure The primary mechanism for data sharing with partners was Vertical Change, a cloud-based case management platform that also included information on enrollee demographics and emergency department utilization. Strengths included accessibility of the platform by diverse WPC partners and use of financial incentives to enforce partner use of Vertical Change. Limitations included dual data entry by wellness coaches due to community-based behavioral partners also using a different data management system (Exym). PILOT IMPLEMENTATION Pilot Enrollment Mendocino enrolled 494 beneficiaries by the end of December 2021. The average length of enrollment was 16.2 months. Approximately 65% of enrollees ever disenrolled at some point between January 2017 and December 2021. The most common reasons for disenrollment were lack of engagement (30% of total WPC enrollment)® and WPC services no longer needed (13%).° Enrollee Identification and Engagement Wellness coaches were responsible for enrollee outreach and engagement. Wellness coaches received referrals from WPC partner agencies and other agencies in the community (e.g., Adult Protective Services, medical clinics, hospitals, and law enforcement), and also accepted self-referrals. Wellness coaches enrolled via street-based/shelter outreach, and facility-based outreach at health care facilities. 8 Beneficiary refused to participate or did not engage in services. 9 Beneficiary was not appropriate or did not benefit from the services provided. Whole Person Care Final Evaluation Report |Appendix U: Pilot Specific Case Studies Care Coordination Enrollees were assigned to wellness coaches, who were responsible for providing all care coordination services and other traditionally “non-billable” services that enrollees needed, and for following enrollees across participating care settings. Wellness coaches were supported by teams comprised of licensed social workers, mental health counselors, and physicians. Most wellness coaches had a 1:19 case ratio. Care coordination was supported by complex care conferences to promote team-based care and collaborative care delivery. Comprehensive assessment of all identified social needs was conducted using the Vulnerability Index-Service Prioritization Decision Assistance Tool (VI-SPDAT) and a homegrown assessment tool to probe WPC qualifying factors and social needs such as housing assistance, utilities, and transportation. Housing Assistance Mendocino emphasized a “Housing First” model in working with enrollees experiencing homelessness, and leveraged other agencies to access housing navigators. WPC funds were used to assist with tenancy support, housing search, and obtaining housing funds. Other Services During the pandemic, Mendocino hosted community vaccine clinics and when needed, wellness coaches shopped for groceries and did grocery door drops on behalf of enrollees. Other offerings included assistance with benefits applications, sobering center services, medical respite, and active referrals to legal services. CRITICAL SUCCESS FACTORS e¢ Widespread data-sharing platform that allowed easy coordination between providers and wellness coaches. e A“Wellness Coach”- centered model that prioritized peer support and a community-based, person-centered approach to delivering services. ¢ Small caseloads, use of multidisciplinary complex care conferences, and frequent team meetings to ensure accountability for care. e Diverse teams with medical staff, social workers, housing support, and substance specialists to offer beneficiaries access to a wide variety of care and services. PERCEIVED IMPACT OF WPC Mendocino perceived an above average impact on the following aspects of WPC Pilot implementation: «Improved integration of health, behavioral health, and social services ¢ Improved care quality © Decreased overall cost of care e Increased data sharing between LE and WPC partners e Identifying enrollees receiving services from more than one system ¢ Improved coordination of care for enrollees (Jey) Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report Mendocino perceived an above average improvement on the following aspects of care for enrollees: * Comprehensiveness of available services ¢ Timeliness of services provided (health, behavioral health, and/or social services) e Targeted identification, outreach/engagement, and enrollment =~ Frequency and quality of communication with enrollees e Extent to which care provided is patient-centered “Our coordination with our behavioral health agencies and the medical teams [is innovative]. We have a Release of Information (ROI) that allows us to share information on our platform and it helps everyone to have all this information... We’re able to call meetings if we have a client we’re worried about and bring their whole team together... The medical component is really big. We receive updates from the ED directly when our clients go in... Our ROI covers a lot of different agencies, mostly medical, but also criminal justice or any other supports they have in the community.” “One of the unspoken qualifiers [for our program] is... severe mental illness and we really wanted people who could show up and actively engage in a care plan, because the resource of our wellness coaches was precious... there’s a whole chasm of stuff needed to support our communities that [wasn’t previously] available. Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies UCLA CENTER FOR HEALTH POLICY RESEARCH it Monterey’s Whole Person Care Pilot PILOT STRUCTURE Target Populations Monterey’s primary population of focus was individuals experiencing homelessness, particularly those with physical and/or mental health comorbidities and/or a history of high utilization of the medical system. Lead Entity and Partnerships In Monterey, the Monterey County Health Department (MCHD) served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included 17 partners from diverse sectors, all of which were identified as having high awareness of and active involvement with implementation of WPC. Longstanding partnerships within the LE (e.g., Public Health and Behavioral Health bureaus) and community partners (e.g., Sun Street Center and Franciscan Workers of Junipero Serra), facilitated enrollee identification and access to services during WPC. Partnerships with other entities, such as the Housing Authority of Monterey County, were new and developed to facilitate enrollee access to housing services. Relationships with the City of Salinas Police Department facilitated enrollee outreach opportunities and diversion from incarceration. Care coordination services were provided both in-house, as well as contracted out to behavioral health, housing services, social services, legal services, and homeless service providers. Data Sharing Infrastructure For care coordination, Monterey utilized an existing county electronic health record (EHR) through OCHIN Epic. Excel spreadsheets sent via encrypted emails were used for data sharing with partners, as allowed through standing agreements and patient consent forms. Strengths included LE access to needs assessment, comprehensive care plan, and referrals within the same system. Limitations included lack of real-time access to medical, behavioral health, or social service data; lack of external partner access to care plans and other enrollee data available in the EHR; and inability to use case-note data for tracking and analysis. PILOT IMPLEMENTATION Pilot Enrollment Monterey enrolled 836 beneficiaries by the end of December 2021. The average length of enrollment was 20.3 months. Approximately 26% of enrollees ever disenrolled at some point between January 2017 and December 2021. The most common reasons for disenrollment were WPC services no longer needed (9% of total WPC enrollment)?° and lack of engagement (8%).1" Enrollee Identification and Engagement Monterey primarily identified eligible individuals through outreach at shelters, encampments, and healthcare facilities, as well as through referrals from partner organizations. Initially, Monterey used eligibility lists provided by the county’s managed care plan but ceased early in the Pilot due to lack of 10 Beneficiary was not appropriate or did not benefit from the services provided. 11 Beneficiary refused to participate or did not engage in services. Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report contact information for individuals experiencing homelessness. Eligible beneficiaries were engaged and enrolled through warm-handoffs at co-located medical and behavioral healthcare partners and community-based locations including shelters, homelessness services, and housing providers. Care Coordination Enrollees were assigned to a care coordination team led by an MCHD public health nurse (PHN). Teams included case managers employed by partner organizations including an alcohol or drug counselor, mental health counselor, housing navigator, benefits support, and clinical psychologist. WPC enrollees had different care coordinators in different care settings in which they were involved, and WPC staff were responsible for communicating with non-WPC coordinators about respective accountability and coordinating hand-offs. Most teams had a 1:43 case ratio. Monterey provided one-on-one case management and linked enrollees to partner services based on need. Care coordination was supported by multidisciplinary team meetings, case conferences, and semi-annual convenings to promote team-based care and collaborative care delivery. Co-located WPC staff at medical and mental health partners facilitated access to services and resources. Comprehensive assessment of all identified social needs was conducted using the Vulnerability Index — Service Prioritization Decision Assistance Tool (VI-SPDAT), the Patient Health Questionnaire-9 (PHQ-9), and a homegrown comprehensive needs assessment. Housing Assistance Monterey participated in streamlining processes or programs that affected delivery and financing of housing services and promoted policy to increase housing availability. WPC funds were used to assist with tenancy support (e.g., counseling and training individuals to move in or remain in temporary or permanent housing), complete applications for the Coordinated Entry system, housing search (e.g., find available temporary or permanent housing stock), obtain housing funds (e.g., housing choice vouchers or rental subsidies), and direct housing funds (e.g., security deposit, furnishings, utilities, legal support, motel vouchers, short-term shelter housing, permanent long-term housing). Other Services In addition to care coordination and housing services, Monterey also provided employment assistance, access to sobering centers, assistance with benefits applications, and transportation to services and appointments. CRITICAL SUCCESS FACTORS ¢ Contracted with a diverse array of cross-sector service providers to provide fee for service (FFS) case management and enrollee prioritization for services, in addition to per member per month (PMPM) WPC care coordination by PHNs. ¢ Population of prioritized people experiencing homelessness with high rate of comorbidities enabled targeted outreach to those with immense need. « Memorandums of Understandings (MOUs) with housing developers to secure dedicated housing placement for WPC enrollees in exchange for delivery of supportive services to residents. e Effective outreach strategy included coordination with WPC-affiliated case managers within existing behavioral and social service providers. Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies PERCEIVED IMPACT OF WPC Monterey perceived an above average impact on the following aspects of WPC Pilot implementation: ¢ Improved integration of health, behavioral health, and social services e Improved care quality e Reduction in inappropriate emergency department visits and hospitalizations ¢ Decreased overall cost of care ® Improved management of care of high risk and high utilizing populations ¢ Improved collaborative partnerships for program implementation Improved coordination of care for enrollees Monterey perceived an above average improvement on the following aspects of care for enrollees: ¢ Coordination and continuity of care e Access to needed services (health, behavioral health, and/or social services) e Access to affordable housing Comprehensiveness of available services e Targeted identification, outreach/engagement, and enrollment Frequency and quality of communication with enrollees e Extent to which care provided is patient-centered “Once you get the housing and you're there to work with these people on a daily basis, everything else will fall into place. So yeah, | don't know how many counties are doing this type of program, but we definitely need to have everybody on board... | think it would've been very, very difficult to launch this without the State financial support, and having that dollar-for-dollar match made everything so much more possible.” “As far as legacy things, you could look at the bigger pictures, like the Project Homekey that is growing now, the new shelter that was built. These are things that we weren’t responsible for entirely, [but] we had some role in... on the more human scale, we housed 53% of all people who were in our program. Again, | guess that you can't say that we did that single handedly, we didn't. But nevertheless, we were part of helping people restore the lives that they wanted to leave because of the assistance that they got because of WPC here.” Ea Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report UCLA CENTER FOR HEALTH POLICY RESEARCH , «5 as ag ~Napa’s Whole Person Care Pilot PILOT STRUCTURE Target Populations Napa’s primary populations of focus included individuals experiencing homelessness, particularly those who were high utilizers. The latter were defined as Medi-Cal beneficiaries within the top 15% of medical system utilization. Lead Entity and Partnerships In Napa, the Health and Human Services Agency (HHSA) served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included 13 partners from diverse sectors, nine of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. Partnerships between the LE and Queen of the Valley Hospital were longstanding and facilitated close coordination of medical services during WPC. Relationships with other entities, such as the HHSA — Public Health were new, and developed to facilitate enrollee connection to resources. Care coordination services were contracted to the CARE (Case Management; Advocacy; Resource and Referral; and Education) Network for the highest acuity patients. Data Sharing Infrastructure The primary mechanism for data sharing with partners was “Bifocus,” which helped staff communicate with partner agencies. Strengths included high rates of participation and adoption by partners, ability to easily generate case reports, and customizability of enrollee Homeless Management Information System (HMIS) data. Limitations included requiring data entry in multiple systems, and reliance on case conferencing due to lack of necessary data (e.g., not all data would make it onto the platform). PILOT IMPLEMENTATION Pilot Enrollment Napa enrolled 771 beneficiaries by the end of December 2021. The average length of enrollment was 13 months. Approximately 62% of enrollees ever disenrolled at some point between January 2017 and December 2021. The most common reasons for disenrollment were WPC services no longer needed (22% of total WPC enrollment)? and lack of engagement (20%).? Enrollee Identification and Engagement Enrollees were identified through referrals from various organizations and partners, including healthcare clinics, police and fire departments, and shelter systems. Outreach was conducted in shelters (e.g., South Napa Shelter helped locate individuals eligible for — but unconnected to — WPC services) and through street-engagement (e.g., monthly visits to nine locations throughout Napa County) by a multi- disciplinary team. Outreach workers enrolled individuals at point of contact and had an enrollment success rate of 90%. 1? Beneficiary was not appropriate or did not benefit from the services provided. 13 Beneficiary refused to participate or did not engage in services. Whole Person Care Final Evaluation Report |Appendix U: Pilot Specific Case Studies Care Coordination Many enrollees were assigned to a care coordination team led by case outreach workers. Teams included registered nurses for medical evaluations, housing navigators for housing-related support, and alcohol and other drug (AOD) specialists for substance abuse services. Most teams had a 1:40 case ratio. Enrollees identified with higher needs were engaged more frequently (e.g., biweekly care coordination meetings). Care coordination was supported by case management meetings with healthcare partners and mental health services at least twice a month to promote team- based care and collaborative care delivery. Comprehensive assessment of all identified social needs was conducted using the Vulnerability Index — Service Prioritization Decision Assistance Tool (VI-SPDAT). Housing Assistance Napa emphasized a “Housing First” model and employed housing navigators to help enrollees become housed. WPC funds were used to assist enrollees to become “document ready” for housing (e.g., identification cards, verification of disability and homelessness). Other Services In addition to care coordination and housing services, Napa also provided respite care, vaccination campaigns, and transportation to appointments. CRITICAL SUCCESS FACTORS ¢ Framing housing as a medical intervention to improve long-term medical outcomes; Napa provided a variety of housing related services including tenancy support, becoming document ready, and housing search. e Athree-pronged referral and outreach system (referrals from partners, street-based outreach, and shelter-based outreach) worked efficiently with healthcare and shelter partners to create pathways to enrollment in WPC. * Frequent interdisciplinary staff meetings enabled collaboration and communication to discuss enrollees requiring extra services. PERCEIVED IMPACT OF WPC Napa perceived a significant impact on the following aspects of WPC Pilot implementation: e Improved integration of health, behavioral health, and social services ® Improved care quality © Decreased overall cost of care e Increased data sharing between LE and WPC partners e Improved management of care of high risk and high utilizing populations e Identifying clients/patients receiving services from more than one system (e.g., medical, behavioral health, social services) e Improved collaborative partnerships for program implementation Improved coordination of care for patients/clients (Je}:3) Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report “Whole Person Care is a great program for our staff, for our clients, the amount of flexibility in the funding, our ability to reduce our emergency capacity by providing the support of Whole Person Care... the coalition that we have here along with our Continuum of Care has really made a difference in our population and our ability to utilize [WPC] money to hire staff so that we can continue to reduce the population of homelessness, at least here in Napa, has been extraordinary.” “We're fortunate that Napa is a pretty small community and there aren't a lot of players, which can be good or bad. We truly are able to get everyone at the table on a biweekly meeting to coordinate, and then we all have access to the same homeless management information system. So, we can see notes that other agencies have put and kind of connect the dots on, "This person says this and you're looking for this. Let's make it happen," type of thing. It's been really beneficial to have those biweekly meetings.” Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies UCLA CENTER FOR HEALTH POLICY RESEARCH ge’s Whole Person Care Pilot PILOT STRUCTURE Target Populations Orange’s primary populations of focus included individuals experiencing homelessness with mild to moderate and/or severe mental illness/substance use disorders (SMI/SUD). Oftentimes, these individuals were high utilizers of care. The latter were identified using administrative data from CalOptima, the county’s Medicaid managed care plan (MCP) and Orange County Health Care Agency’s Behavioral Health Division. Lead Entity and Partnerships In Orange, the Health Care Agency (HCA) served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included 35 partners from diverse sectors, all of whom were identified as having a high awareness of WPC and as actively involved in implementing WPC. Partnerships between the LE, the MCP, and multiple community partners (e.g., St. Joseph Hospital) existed prior to WPC and facilitated data sharing and care coordination efforts during WPC. New relationships were also developed with medical centers, housing agencies, and other community-based organizations (CBOs) to further facilitate data sharing, case management, and care coordination in WPC. Care coordination services were provided by the WPC team in conjunction with county Behavioral Health Services (BHS), Public Health Services, county contracted providers, hospitals, and community clinics. Data Sharing Infrastructure The primary mechanism for data sharing with partners was an electronic, web-based care coordination platform, “WPC Connect,” developed by Safety Net Connect. The platform allowed appropriate communication and coordination of all services for WPC enrollees between providers for continuum of care including but not limited to medical, behavioral, social supportive services, and housing-related needs. It was a comprehensive view of service referrals and services being provided to WPC clients and store client information. Strengths included daily data feeds from the MCP to facilitate identification of eligible enrollees; automatic feeds from 10 local hospitals and feeds from clinics; ability to identify high utilizers; collection of behavioral health service data, including outreach and engagement information; ease of using the system for closed loop referrals; and the ability to automatically export data into state- required reporting templates. Limitations included lack of interoperability with electronic health record (EHR) systems being used by many WPC partners, staff resistance to entering data in WPC Connect in addition to internal EHR, and limited availability of behavioral health data (e.g., only showed whether enrollee received any behavioral health services, and did not include a diagnosis). PILOT IMPLEMENTATION Pilot Enrollment Orange enrolled 13,861 beneficiaries by the end of December 2021. The average length of enrollment was 9.6 months. Approximately 94% of enrollees ever disenrolled at some point between January 2017 610 | Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report and December 2021. The most common reasons for disenrollment were lack of engagement (72% of total WPC enrollment)** and ineligibility for Medi-Cal (20%).*> Enrollee Identification and Engagement To generate referrals, Orange allowed referrals from hospitals and community clinics, and also used street and shelter-based outreach to identify individuals not engaged in traditional healthcare settings. Care Coordination Enrollees were assigned to a care coordination team comprised of community health workers, nurses, and housing navigators. Team members were employed by different WPC partners (e.g., county behavioral health, hospitals, and community clinics) and communicated regularly to coordinate care across settings. A single, more centralized care coordinator role was considered in PY 4 (2019) but ultimately not implemented. In addition to conducting a needs assessment and developing a care plan, care coordinators also helped arrange transportation to and from appointments, assisted with medication management and adherence, and ensured warm hand-offs to other providers when referrals were needed. Most teams had a 1:35 case ratio. Bi-weekly to monthly meetings were used to promote team-based care and collaborative care delivery. Comprehensive assessment of all identified social needs was conducted using the Vulnerability Index — Service Prioritization Decision Assistance (VI-SPDAT) tool. Housing Assistance Orange emphasized use of a “Housing First” model. WPC funds were used to assist with obtaining housing and support short-term housing in a shelter and permanent long-term housing. Other Services In addition to case management, care coordination and housing services, Orange also provided assistance with various social supportive services, benefits applications, transportation to appointments/services, and recuperative care. CRITICAL SUCCESS FACTORS ¢ Collaboration with community partners was perceived as innovative and a key driver for identifying individuals for the WPC program. e Care coordination platform that enabled communication between organizations was viewed as a critical tool that fostered collaboration among partners, improved care management, and streamlined reporting. ¢ Developing a model of care with shared goals that addressed each WPC partner’s priorities was critical to successful collaboration between WPC partners and required significant upfront investment. e Engaging frontline staff and partners to solicit input in the design, implementation, and evaluation of the pilots was perceived as effective in garnering buy-in and securing match funding for aspects of the WPC program. 14 Beneficiary refused to participate or did not engage in services. 15 Beneficiary could no longer remain enrolled in WPC if no longer eligible for Medicaid benefits. Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies e Expanding outreach to streets and shelters enabled the team to enroll individuals who would benefit from WPC services but were difficult to reach without phone access and engagement with traditional healthcare settings. PERCEIVED IMPACT OF WPC Orange perceived an above average impact on the following aspects of WPC Pilot implementation: e Improved integration of health, behavioral health, and social services Improved care quality e Reducing inappropriate emergency department visits and hospitalizations * Decreased overall cost of care ¢ Improved enrollee health and well-being e Increased data sharing between LE and WPC partners Orange perceived an above average improvement on the following aspects of care for enrollees: e Access to needed services (health, behavioral health, and/or social services) «Access to affordable housing Frequency and quality of communication with enrollee “Some of [our staff] routinely actually go out [to streets or shelter] and are identifying the clients there and saying, "Hey, we could help you. We could support you. Would you be interested?" There has to be that willingness to be out in the community to leave your building and go forth and interact with these clients. Otherwise, there's no chance of it being successful at all.” “And [WPC] also built trust among the providers with us. So, with housing, it's building trust with the recuperative care providers or vice versa. So that they know that this is a good provider, that they can refer a client to the more likely to get housed, et cetera... [And] for the housing collaborative meetings, it's been really useful to have everybody at the table, including the housing authorities, the office of care coordination, and the providers.” Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report UCLA CENTER FOR HEALTH POLICY RESEARCH , «5 as Ait Placer’s Whole Person Care Pilot PILOT STRUCTURE Target Populations Placer included all WPC target populations, including high utilizers, individuals experiencing homelessness or at risk of homelessness, chronic physical conditions (two or more), severe mental illness/substance use disorder, and justice involved. Enrollees could fall into more than one population of focus. High utilizers were defined as those with three or more emergency department visits in the last year. Lead Entity and Partnerships In Placer, Health and Human Services (HHS) served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included 24 partners from diverse sectors, 18 of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. Partnerships between the LE and subdivisions of HHS were longstanding (including public health, human services, and adult systems of care) and facilitated data sharing and access to needed services for enrollees. Existing relationships with Advocates for Mentally Ill (AMI) Housing and The Gathering Inn, two permanent supportive housing providers, were expanded as part of the Pilot and were considered critical to the Pilot’s success. New partnerships were developed with several medical providers (e.g., Chapa-De Indian Health, WellSpace Health, Kaiser Permanente) to facilitate enrollee access to healthcare services. Placer also worked closely with Sutter Health to conduct emergency department follow-up visits and receive real time alerts on enrollees. All care coordination services were provided directly by HHS, rather than through contracts with external service providers. Contracts were used for supportive housing services. Data Sharing Infrastructure The primary mechanism for data sharing with partners was “AVATAR,” an electronic health record (EHR), which tracked care coordination activities and contained health, behavioral health, and social service data available across various subdivisions of HHS. An electronic system called “Pre-Manage” complemented the EHR, and provided real-time notifications when enrollees received hospital or emergency department services. Some partners directly accessed information in Pre-Manage, while others contacted care coordinators for relevant information as needed. Strengths of Placer’s data sharing included all care coordinators were provided cell phones and laptops to access data in the field; real time notifications; and access to a wide variety of data streams. Limitations included need to access two separate systems, and partners outside of HHS had limited access. PILOT IMPLEMENTATION Pilot Enrollment Placer enrolled 501 beneficiaries by the end of December 2021. The average length of enrollment was 12.2 months. Approximately 87% of enrollees ever disenrolled at some point between January 2017 and Whole Person Care Final Evaluation Report |Appendix U: Pilot Specific Case Studies December 2021. The most common reasons for disenrollment were graduation (22% of total WPC enrollment)*¢ and lack of engagement (19%).1” Enrollee Identification and Engagement Placer used the Continuum of Care’s “by-name” list, which assigned a vulnerability score to each individual who called 2-1-1 (a service that connected individuals to needed services, including housing). As openings occurred, Placer would reach out and try to engage individuals on the list with the highest scores. Community based referrals were also used for identification of potential enrollees, but priority was given to those who were on the “by-name” list. Care Coordination Enrollees were assigned to a primary care coordinator, who followed enrollees across all WPC settings. This care coordinator could be an individual with lived or family experience with homelessness, mental health issues, or substance use problems or an individual with a master’s level expertise in an area of identified need. Staff were responsible for providing not only care coordination but also case management. Care coordinators were supported by nurses, clinicians, and housing specialists. In some cases, care coordination services were available outside of typical business hours (e.g., evenings or weekend). Most teams had a 1:20 case ratio. Supervisors met weekly with care coordinators to provide support around crisis management and case consultation. A comprehensive assessment of all identified social, health, mental health, and substance use needs was conducted at the start of services and periodically updated as needed. Housing Assistance Placer emphasized a “Housing First” model, and all supportive housing services (e.g., tenancy support, housing search, landlord incentives, funds for security deposits) were provided by LE using WPC funds except medical respite, short-term shelter, and permanent long-term housing (which were provided by WPC partners). Other Services In addition to care coordination and housing services, Placer also connected enrollees to public benefits, provided one-on-one coaching/education programs to assist with employment, medical respite, coordination of transportation, and referrals to legal services. CRITICAL SUCCESS FACTORS e Placer focused on interpersonal relationships and promoted active engagement of partners at every level of the organization (e.g., leadership, management, and frontline staff). e Contracts with partners clearly delineated expectations, particularly around a systemic approach to improving permanent supportive housing principles. «The organizational structure of the LE facilitated data sharing and collaboration; subdivisions within HHS had pre-existing relationships and were all on the same electronic health record. 16 Beneficiary achieved desired goals. 1 Beneficiary refused to participate or did not engage in services. (39) Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report e Despite limited housing stock, Placer identified funding opportunities through partner organizations for direct housing funds. For example, Sutter Health made a significant contribution to purchase 20 units of dedicated housing for WPC enrollees. e Emphasis on hiring staff with a personal fit for the program who exemplified dedication and compassion; Placer had low turnover rates amongst care coordinators in early years of the program. The majority of care coordinators had lived or family experience which facilitated trust and rapport building with clients; emphasis on developing relationship with client through bonding activities (e.g., cooking a meal, or taking them to their new home). This was possible by relatively small caseloads. e Aperson-centered, strength-based clinical approach provided the working paradigm that allowed for enrollees to thrive. PERCEIVED IMPACT OF WPC Placer perceived an above average impact on the following aspects of WPC Pilot implementation: ¢ Improved integration of health, behavioral health, and social services e Improved care quality © Decreased overall cost of care e Improved enrollee health and well-being e Identifying clients/patients receiving services from more than one system (e.g., medical, behavioral health, social services) e Improved collaborative partnerships for program implementation e Improved coordination of care for enrollees Placer perceived an above average improvement on the following aspects of care for enrollees: ¢ Continuity of care e Access to affordable housing e Comprehensiveness of services provided @ Frequency and quality of communication with enrollee e Extent to which care provided is patient-centered “A nice thing about Placer County is we're not too big of a county [such] that each separate system is so siloed and so separate and so much of a different culture. And we're not so small that it's like everybody in town knows each other. Placer County is a good size where you do have separate organizations doing separate things, but it's small enough where people can connect and talk and work on whatever things are going on.” “.,.[the program manager's] leadership was a big strength... and | think coupled to that is the collaboration that we've been able to have. Our program manager just picked a team of people that were go-getters from the start. And so even when some of us hadn't even done this, we just went out, we figured it out, we made calls, we found out what the resources were.” Whole Person Care Final Evaluation Report |Appendix U: Pilot Specific Case Studies UCLA CENTER FOR HEALTH POLICY RESEARCH at Riverside’s Whole Person Care Pilot PILOT STRUCTURE Target Populations Riverside’s sole population of focus was the justice involved population, defined as individuals exiting incarceration, on probation/parole for at least 12 months, and either affected by physical and mental health conditions or at risk of homelessness. Lead Entity and Partnerships In Riverside, the Riverside University Health System (RUHS) served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included 15 partners from diverse sectors, nine of whom were identified as having a high awareness of WPC and as actively involved in implementing WPC. Partnerships between the LE and most partners were longstanding and facilitated provision of medical, behavioral health, housing, and other social services during WPC. Partnerships with the Cal State San Bernardino Reentry Initiative and the Riverside County Probation Department were new, and developed to facilitate services for the justice-involved population. Care coordination services were not contracted out and were provided in-house by RUHS. Data Sharing Infrastructure The primary mechanism for data sharing with partners was “SAS Viya,” a data management platform that helped integrate detention health, behavioral health, and other data from the county’s public hospital, behavioral health system, county jail, and other systems in a single location. Strengths included a relatively inexpensive cost with ability to use SAS Visual Analytics features to more readily create reports and provide real time information to partners. Limitations included need for training on how to use SAS in addition to existing data management systems already in use (e.g., Epic, TechCare, etc.). PILOT IMPLEMENTATION Pilot Enrollment Riverside enrolled 13,531 beneficiaries by the end of December 2021. The average length of enrollment was 23.7 months. Approximately 58% of enrollees ever disenrolled at some point between January 2017 and December 2021. The most common reason for disenrollment was lack of engagement (55% of total WPC enrollment).1® Enrollee Identification and Engagement Eligible enrollees for Riverside were identified by registered nurses (RNs) who were located onsite at probation offices. Probationers were screened to evaluate their health, behavioral health, substance use, housing, and social needs. Once needs were determined, RNs connected these individuals to community and county resources, and in some cases, to care managers. Additional eligible individuals were identified through outreach in the community at targeted events such as probation resource fairs. In the event that eligible individuals did not follow through with a referral, the RN made an effort to contact the individual up to four times. When appropriate, the RN worked with probation officers to 18 Beneficiary refused to participate or did not engage in services. (SIM) Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report determine the ideal mode of communication with eligible individuals, this included reaching them through friends and family. Care Coordination Probationers were assisted with care coordination by RNs. Referrals were made to teams including specialists in mental health, alcohol and drug dependence, housing, and benefits eligibility. Additionally, peer support specialists with lived experience similar to the enrolled population were available to encourage enrollee engagement. Care coordination was delivered through a single, dedicated RN care coordinator who followed enrollees across all WPC-participating care settings. Most RN care managers had a 1:50 case ratio. Care coordination was supported by regular “huddles” and monthly multidisciplinary team meetings to promote team-based care and collaborative care delivery. Comprehensive assessment of all identified social needs was conducted using a homeless screening tool, a substance use disorder questionnaire, a behavioral health questionnaire, and a homegrown WPC-specific assessment to assess use of prescription medications, medical conditions, health insurance coverage, food stamps, and other needs. Housing Assistance Riverside emphasized a “Housing First” model and provided services to coordinate housing for enrollees transitioning out of incarceration. All housing-related services were provided by partners either using WPC funds or alternative funds. WPC funds were used to assist with landlord incentives and ongoing assistance with enrollee-landlord relationships even after enrollees were housed. Other Services In addition to care coordination and housing services, Riverside also provided assistance with benefits applications, sobering center services, transportation to services and appointments, and linkages to legal services. CRITICAL SUCCESS FACTORS e Strong cross-sector collaboration between RUHS and the probation office, parole sites, and medical, behavioral health, and social service providers allowed for more efficient care management. ¢ Onsite outreach at probation offices allowed for successful recruitment of eligible individuals during a vulnerable transition. This was evidenced by the 94% acceptance rate following the initial screening. « Use of RNs for initial screening and care management fostered trust between eligible enrollees and the RNs. The RNs were actively involved in helping enrollees making medical, behavioral health, and social services appointments as necessary. e Standardized data sharing system allowed all individuals to view and share information regarding enrollees across multiple read-only platforms, which was crucial to coordinating care and services for enrollees. ¢ Monthly multidisciplinary team meetings held individuals responsible for aspects of enrollee success within WPC accountable and facilitated communication between service providers. e Flexibility and incorporating RN feedback into the Pilot improved program success and allowed for adjustments based on enrollee experience. Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies PERCEIVED IMPACT OF WPC Riverside perceived an above average impact on the following aspects of WPC Pilot implementation: Improved integration of health, behavioral health, and social services Improved care quality Reduced inappropriate emergency department visits and hospitalizations Decreased overall cost of care Improved enrollee health and well-being Improved management of care of high risk and high utilizing populations Identifying enrollees received services from more than one system (e.g., medical, behavioral health, social services) Improved collaborative partnerships for program implementation Riverside perceived an above average improvement on the following aspects of care for enrollees: Coordination and continuity of care Access and comprehensiveness of needed services (health, behavioral health, and/or social services) Access to affordable housing Timeliness of services provided Targeted identification, outreach/engagement, and enrollment Frequency and quality of communication with enrollees Extent to which care provided is patient-centered Reduction of re-incarceration The nurse of course would say, our goal is to prevent re-incarceration and also to get you substance use, behavioral health, and physical health care in the best setting so that you don't have to go to the emergency department for care. And our acceptance rate was 94%. So it's pretty high. We've now offered over 15,000 individuals screening.” “.. These people who really needed somebody to care about them, they got people to care about them. To me, that was the biggest resource that helped these people be successful, who have been. Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report UCLA CENTER FOR HEALTH POLICY RESEARCH at Sacramento’s Whole Person Care Pilot PILOT STRUCTURE Target Populations Sacramento’s primary populations of focus included high utilizers, defined as those meeting crisis system utilization criteria (e.g., more than one inpatient hospital stay; more than four emergency department (ED) visits; more than four crisis interventions), and individuals experiencing homelessness, based on provider- or self-report. Lead Entity and Partnerships In Sacramento, the City of Sacramento served as the lead entity (LE) responsible for program implementation and reporting to the state; though management of the program was contracted out to the consulting firm Transform Health. As of January 2020, the Pilot included 32 partners from diverse sectors, 17 of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. Partnerships between the LE and partners such as Sacramento Covered, Lutheran Social Services, Sacramento Self Help Housing, and Sacramento County Department of Human Assistance facilitated referrals and care coordination during WPC as well as collaboration between City and County services. Partnerships with federally qualified health centers (FQHCs) such as One Community Health, WellSpace Health, Sacramento Native American Health Center, and Elica Health Centers helped ensure enrollee connection to needed medical care. All care coordination services were contracted out to community partners. Data Sharing Infrastructure The primary mechanism for data sharing with partners was an existing Salesforce care management platform called “Shared Care Plan” which helped share enrollee medical, behavioral health, and other information between designated staff at service partner organizations. Strengths included real-time care coordinator access to medical, behavioral health, and social services data, access to needs assessment, care plan, and referrals in one platform. Limitations included read-only access for many partners, and lack of integration with all partners causing dual data entry on multiple platforms. Clinical service partners (e.g., hospitals, FQHCs) maintained independent electronic health records (EHRs) but used the Shared Care Plan to document care management and service coordination information. PILOT IMPLEMENTATION Pilot Enrollment Sacramento enrolled 2,345 beneficiaries by the end of December 2021. The average length of enrollment was 15.4 months. Approximately 72% of enrollees ever disenrolled at some point between January 2017 and December 2021. The most common reasons for disenrollment were lack of engagement (36% of total WPC enrollment)*9 and graduation (21%).?° 19 Beneficiary refused to participate or did not engage in services. 20 Beneficiary achieved desired goals. Whole Person Care Final Evaluation Report |Appendix U: Pilot Specific Case Studies Enrollee Identification and Engagement Sacramento primarily used referrals from specified referring entities (e.g., hospital systems, managed care plans (MCPs), police and fire departments) using a standardized referral form to identify eligible enrollees, though warm handoffs from Pathways community health workers (CHWs) were also accepted. Enrollee outreach was multi-faceted and included street- or shelter-based outreach and outreach staff placed at health care facilities. Outreach CHWs provided ongoing connection to social services and typically had lived experience similar to the enrollee population. Eligible beneficiaries were enrolled at healthcare facilities and warm handoffs at co-located organizations. Sacramento aimed to have a CHW engage enrollees within two hours of accepting a referral. Eligibility lists from MCPs were also used but less effective than warm handoffs from community partners. Care Coordination Sacramento organized providers into four categories based on service provided: eligibility and enrollment, outreach and referrals, housing, and health care. Enrollees were supported by a team of multiple care coordinators across multiple WPC partners that included an Outreach CHW, clinical hub provider, and housing provider. Clinical hub teams were led by a program manager, often licensed clinical social workers, though nurse practitioners, physicians, and psychologists were available for more intensive case management or consult as needed. Much of the care management work in the clinical hub was performed by CHWs or case managers. Teams had varying case ratios based on program or “hub”: health care providers had an average case ratio of 1:50 (range 25-75); housing providers had a caseload of 1:55 (range 35-75); and outreach had a 1:75 case ratio. Care coordination was supported by multidisciplinary team meetings and case conferences to promote team-based care and collaborative care delivery. Comprehensive assessment of all identified social needs was conducted using Vulnerability Index — Service Prioritization Decision Assistance Tool (VI-SPDAT) and a tool developed by Sacramento in collaboration with all partners. Housing Assistance Sacramento emphasized a “Housing First” model and promoted streamlining processes or programs that affected the financing or delivery of housing services. WPC funds were used by partner organizations to assist with housing application fees, security deposits, furniture and appliances, utilities, legal support for tenancy-related issues, and ongoing assistance with enrollee-landlord relationships. Other Services In addition to care coordination and housing services, Sacramento also assisted with public benefits, employment support, health education, and referral to legal services. Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report CRITICAL SUCCESS FACTORS * Responsibility for care coordination was distributed among multi-disciplinary teams that included a clinical provider, housing navigator, and community health worker. e Street presence and encampment-based outreach helped maintain connection with enrollees experiencing homelessness. © WPCstaff at health care facility “hubs,” combined with hospital alert platform used by some hubs (not program wide), enabled communication with enrollees and follow through with enrollee goals. * Coordination between partners streamlined workflows which clarified pathways for enrollees and care coordinators to meet enrollees’ needs. e Data sharing via the Shared Care Plan allowed all partners to communicate and document enrollee contact information. ¢ Community Health Workers championed enrollees and provided a myriad of supports based on what the enrollee wanted to prioritize in their care plan. PERCEIVED IMPACT OF WPC Sacramento perceived an above average impact on the following aspects of WPC Pilot implementation: ¢ Reducing inappropriate emergency department visits and hospitalizations © Decreased overall cost of care e Improved enrollee health and well-being e Increased data sharing between WPC partners e Identifying and connecting clients/patients receiving services from more than one system (e.g., medical, behavioral health, social services) e Improved coordination of care for enrollees * Coordinating housing resources to house enrollees Sacramento perceived an above average improvement on the following aspects of care for enrollees: Targeted identification, outreach/engagement, and enrollment Frequency and quality of communication with enrollee Extent to which care provided is patient-centered “Prior to [WPC], housing providers and federally qualified health centers, weren't necessarily communicating with each other and were much more siloed. Whereas [after WPC], they work together as acare team, and now we're able to cultivate relationships with each other that we are hopeful will exist outside of [WPC]. And it's...one of the best things to have come out of the program.” “Our program is based on a collective impact model. So we're bringing together multiple stakeholders, not just primary care providers, behavioral health providers, housing providers, hospital systems, health plans, to work towards a common goal of serving this population that's need and vulnerable. For us, that's really the crux of the program and why it's so important and how we primarily serve this population.” Whole Person Care Final Evaluation Report |Appendix U: Pilot Specific Case Studies [yan UCLA CENTER FOR HEALTH POLICY RESEARCH um =6san Bernardino’s Whole Person Care Pilot PILOT STRUCTURE Target Populations San Bernardino’s primary population of focus included high utilizers of county facilities with two or more chronic conditions; prospective enrollees were identified and prioritized by a scoring algorithm applied to administrative data from multiple partners. High utilizers were defined as those with six or more emergency department visits, or three or more inpatient hospital stays, in the prior six months. Lead Entity and Partnerships In San Bernardino, the Arrowhead Regional Medical Center (ARMC) served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included nine partners from diverse sectors, six of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. Partnerships between the LE and managed care plans (MCPs) were longstanding and facilitated data sharing during WPC. Other meaningful partnerships included those with the Sheriff's Department, Behavioral Health, and Human Services; having a “champion” for WPC in these organizations was seen as a facilitator to identifying enrollees for WPC and ensuring appropriate receipt of services. Care coordination services were provided directly by San Bernardino (i.e., not contracted out to partner organizations). Data Sharing Infrastructure The primary data sharing infrastructure was a population health management platform developed by Forward Health, specifically for WPC. San Bernardino consciously chose not to utilize the electronic medical record (EMR) utilized by the broader ARMC because they served the entire county. Strengths included quick agreement amongst county partners on the memorandum of understanding (MOU) that guided the data sharing approach for WPC; use of incentives to encourage data sharing amongst partners based on volume of data shared; and using data to inform strategic identification and enrollment of prospective enrollees. Limitations included reporting errors from partners. PILOT IMPLEMENTATION Pilot Enrollment San Bernardino enrolled 1,552 beneficiaries by the end of December 2021. The average length of enrollment was 16.1 months. Approximately 70% of enrollees ever disenrolled at some point between January 2017 and December 2021. The most common reasons for disenrollment were lack of engagement (25% of total WPC enrollment)?+ and WPC services no longer needed (15%).?2 Enrollee Identification and Engagement San Bernardino developed an algorithm that was applied to shared administrative data from multiple partners (e.g., including public health and behavioral health) and lists provided by MCPs — the targeted eligibility lists were reviewed during WPC meetings and then shared with care coordination teams. 21 Beneficiary refused to participate or did not engage in services. 22 Beneficiary was not appropriate or did not benefit from the services provided. (yh) Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report PERCEIVED IMPACT OF WPC San Bernardino perceived an above average impact on the following aspects of WPC Pilot implementation: Improved integration of health, behavioral health, and social services Improved care quality Reduction in inappropriate emergency department visits and hospitalizations Increased data sharing between LE and WPC partners Identifying clients/patients receiving services from more than one system (e.g., medical, behavioral health, social services) Improved collaborative partnerships for program implementation San Bernardino perceived an above average improvement on the following aspects of care for enrollees: Coordination and continuity of care Comprehensiveness and timeliness of available services (health, behavioral health, and/or social services) Targeted identification, outreach/engagement, and enrollment Frequency and quality of communication with enrollee Overall enrollee well-being “| think what is most important and what works for me personally is building that rapport with the client from day one. From that first encounter that you have with that client, again, just going to their level and allowing for them to understand that you're there necessarily just to help them, just to provide guidance, education, support. Although the program is designed to focus on the client's overall health, essentially, if clients’ basic needs are not met, they're not going to care if they went to a doctor's appointment or not. They're more concerned about a roof over their head, food for their stomach, or clothing for their children, or whatever the case may be.” “Nothing is set in cement. It's a moving target. We have to evolve and change with our clients. All of our clients are different. We can't just make a mold and say, ‘Here, sit in this mold somewhere.’ It's always changing...lf programs aren't willing to change, | don't think they'll succeed.” Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report Outreach was street-based and at point of care by WPC outreach teams, supported by two navigators who were well versed in homelessness and available support services. Care Coordination San Bernardino consisted of ten mobile teams comprised of patient navigators with shared lived experience who were supported by three specialists: RN care manager, social worker, and alcohol and drug counselor. There was also a utilization technician and office assistant who facilitated administrative activities. While the patient navigator provided the “primary touch,” other team members could follow enrollees depending on the WPC-participating care setting. Each mobile team met twice a month to review enrollee needs and promote communication across roles within the team. Most teams had a 1:55 case ratio. For the first four years of WPC, the Patient Activation Measure survey was used to stratify enrollees into tiers based on acuity level, which helped strategically understand differences in enrollee need. Care coordination was supported by multidisciplinary team meetings and “WPC Accountability Review” conferences (a monthly review of every enrollee with each team by the Program Manager) to promote team-based care and collaborative care delivery. Housing Assistance San Bernardino emphasized a “Housing First” model. WPC partners used WPC funds to provide short- term and long-term housing, ongoing assistance with enrollee-landlord relationships after enrollees were housed, and to provide motel vouchers or equivalent to help cover short-term stays. Other Services In addition to care coordination and housing services, San Bernardino also provided linkages to sobering center stays, medical respite, access to educational activities, and connection to public benefits. CRITICAL SUCCESS FACTORS e Pre-existing integration facilitated strong partnerships, complemented by initiative at the beginning of the program to get all partners in strategic alignment (e.g., MOUs, workflows, vision/goals of program). e Expectedly, there have been transitions of staff, but the program manager was dedicated to facilitating transitions and hand-offs in a way to best support WPC, which required consistent monitoring and oversight. ¢ Patient navigators with lived experience provided trust and rapport building with enrollees. e A largely field based case management approach allowed for flexibility in working with enrollees and the ability to engage effectively with unhoused enrollees. e Each enrollee was strategically discussed at monthly meetings between the program manager and responsible team, allowing for strategic assessment and comprehensive planning for each enrollee. ¢ Data drove identification of potential enrollees, which was facilitated by incentives that encouraged comprehensive data sharing from partners. Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies UCLA CENTER FOR HEALTH POLICY RESEARCH A fit PILOT STRUCTURE Target Populations San Diego’s primary populations of focus included those experiencing homelessness or at-risk of homelessness and high utilizers. The latter were defined as those with three or more emergency department (ED) visits in the year prior to enrollment. Lead Entity and Partnerships In San Diego, the county Health and Human Services (HHS) Agency served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included 20 partners from diverse sectors, 13 of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. Partnerships between the LE and most Medi-Cal managed care plans (MCPs) were longstanding and facilitated data sharing and collaboration during WPC. Relationships with several other entities, such as law enforcement, were new and developed to facilitate data sharing and ensure appropriate referrals to or from WPC. Care coordination services were contracted to two community partners via an extensive request for proposal process. Data Sharing Infrastructure The primary mechanism for data sharing with partners was “ConnectWellSD,” which helped integrate data across county HHS departments and served as the primary vehicle for contracted agencies to share program-relevant data with the LE. Strengths included real-time alerts (e.g., ED visits); updates on when Medi-Cal coverage was set to expire; and Sheriff’s Department notifications of current incarcerations. Limitations included difficulty in navigating menus and in extracting data needed to fulfill WPC reporting requirements. Consequently, contracted agencies continued to maintain separate, internal data management systems, resulting in duplicative data entry requirements for frontline staff. PILOT IMPLEMENTATION Pilot Enrollment San Diego’s Pilot enrolled 958 beneficiaries by the end of December 2021. The average length of enrollment was 15.0 months. Approximately 76% of enrollees ever disenrolled at some point between January 2017 and December 2021. The most common reasons for disenrollment were lack of engagement (30% of total WPC enrollment)?? and graduation from the program (20%).24 Enrollee Identification and Engagement 23 Beneficiary refused to participate or did not engage in services. 24 Beneficiary achieved desired goals. Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies I] San Diego’s Pilot initially used MCP-generated lists to identify eligible enrollees. However, difficulty locating eligible enrollees in a timely fashion resulted in a shift to requesting referrals from community partners (e.g., hospitals, homeless outreach teams, law enforcement, and community health centers). To facilitate appropriate referrals, San Diego developed a one-page referral sheet outlining basic eligibility criteria and referral processes. Following referral, WPC teams had 30-60 days to locate and engage prospective enrollees; enrollees not successfully engaged during this time-period were disenrolled but could be re-referred and re-enrolled at a later date. Care Coordination Enrollees were assigned to a care coordination team led by a supervising clinical case manager. Teams included community health workers or peer staff, mental health counselors, and housing navigators to support outreach and care coordination. Nurses and licensed and unlicensed social workers were available for clinical consult, as needed. Care coordinators within a team followed enrollees across care settings based on availability/schedule and/or expertise. Most teams had a 1:25 case ratio. Enrollees requiring more intensive case management (e.g., at least 3-5 hours/week) were assigned to specialized teams with a smaller case ratio of 1:10 to allow for more intensive support. Care coordination was supported by multidisciplinary team meetings and case conferences to promote team-based care and collaborative care delivery. Comprehensive assessment of all identified social needs was conducted using the Vulnerability Index — Service Prioritization Decision Assistance tool (VI-SPDAT) and a homegrown, provider- developed biopsychosocial assessment tool. Housing Assistance The Pilot emphasized a “Housing First” model, and co-located housing services with other social service providers. All housing-related services were provided by partners either using WPC funds or alternative funds (e.g., Housing and Disability Advocacy Program funds). WPC funds were used to assist with completing applications for Coordinated Entry, housing searches, obtaining housing funds (e.g., vouchers or subsidies), paying for utilities and necessary items, and ongoing management of enrollee-landlord relationships following housing placement. Other Services In addition to care coordination and housing services, San Diego’s Pilot also provided benefits assistance, employment services, respite or recuperative care, transportation, health education, legal services, and services specifically designed to address life post-incarceration. CRITICAL SUCCESS FACTORS e Strong cross-sector collaboration resulted from existing relationships between the county HHS and local Medicaid MCPs. In addition, regular meetings with all partners was perceived by San Diego to improve community awareness of gaps in existing systems of care for enrollees with medical or mental health need who were experiencing homelessness or were at-risk of homelessness. These meeting also strengthened relationships between WPC partners in ways that would benefit future efforts to integrate care. In particular, the LE Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report reported improved relationships between county Health and Human Services Agency and local behavioral health, law enforcement, and housing providers. Intensive field-based outreach by staff with lived experience and training in client-centered, trauma-informed approaches were perceived as critical for identifying and engaging enrollees, particularly given absence of a street medicine program within the county. Frontline staff emphasized the importance of in-person outreach to build trust and rapport, to the point where enrollees would eventually proactively reach out to them rather than wait to be contacted or found. Low and stratified caseloads were perceived as important for ensuring that enrollees received appropriately intensive case management support and mitigating care coordinator burnout. Burn out was a common challenge amongst care coordinators due to the emotionally demanding nature of their work (i.e., time intensive involvement with a full case load enrollees who might be dealing with multiple and complex personal and health issues). Clearly defined scope of work for contracted service providers reduced ambiguity in WPC program goals and allowed for careful selection of community partners with appropriate geographic scope and networks to effectively implement WPC. This process was challenging because it delayed initial program implementation and increased administrative burden. Blending housing funds in contracts with community providers allowed care coordinators and housing navigators to provide enrollees with critical supports non-billable through WPC. One example was blending of WPC One-Time Housing Funds and Housing and Disability Advocacy Program funds. Developing appropriate data sharing agreements was viewed as critical to successful development of a comprehensive data sharing platform, but also extremely time-consuming. Coordinating with seven MCPs for the data necessary for metric reporting posed challenges in early years of the program, but existing relationships with the MCPs was viewed as a facilitator to generating eventual buy-in. Pilot utilized partnerships to pay directly for housing (e.g., Housing Development Assistance Programs Funds)— 634 individuals were housed and 514 were permanently housed through WPC from inception to the end of 2020. PERCEIVED IMPACT OF WPC San Diego perceived an above average impact on the following aspects of WPC Pilot implementation: @ Decreased overall cost of care e Increased data sharing between LE and WPC partners e Improved management of care of high risk and high utilizing populations e Improved collaborative partnerships for program implementation e@ Improved coordination of care for enrollees San Diego perceived an above average improvement on the following aspects of care for enrollees: e Delivery of care coordination services to enrollees e@ Continuity of care Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies Access to needed health, behavioral health, and/or social services Access to affordable housing Comprehensiveness and timeliness of available services Frequency and quality of communication with enrollees Provided care is patient-centered Overall enrollee well-being “There's such a huge difference between what HUD pays for and then what is provided by Medi- Cal and knowing really that this [Whole Person Care] ... this really filled a gap” “... It [Whole Person Care] is truly changing the way social services are delivered, flipping the expectation that individuals need to come into the office for services... Persistence and highly skilled ... [and] trained staff are a requirement for making the kind of inroads that ...[were] made with respect to ... improving quality of life [for enrollees].” Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report UCLA CENTER FOR HEALTH POLICY RESEARCH at San Francisco’s Whole Person Care Pilot PILOT STRUCTURE Target Populations San Francisco’s primary population of focus was individuals experiencing homelessness, measured at any point in a rolling 12-month period. Many also had chronic physical health conditions, social determinants of health, along with medical and behavioral health needs. Lead Entity and Partnerships In San Francisco, the Department of Public Health (SFDPH) served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included nine partners from diverse sectors, five of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. Partnerships between the LE and many entities were longstanding, such as SFDPH medical and behavioral divisions, the Department of Homelessness and Supportive Housing (HSH), Human Services Agency (HSA; providing Medi-Cal enrollment), and San Francisco Fire Department (providing community paramedics). These partner agencies shared clients and had similar needs for care coordination and data sharing. Relationships with several community-based entities, such as HealthRIGHT 360 and Positive Directions, were new and developed to support enrollee access to behavioral healthcare. Housing care coordination services were contracted to multiple county and nonprofit organizations. Operationally, San Francisco worked horizontally on data sharing, care coordination support, and innovative inter-agency coordination projects. They worked vertically to establish a high-level city-wide system of care for the target population, including legal authorizations to share data at mid-level to support clinicians in their daily work and with front-line services to improve access for clients in need. Data Sharing Infrastructure Prior to WPC, San Francisco had established a Coordinated Care Management System (CCMS) database that integrated essential health, behavioral health, and social determinants of health on persons aged 18 and older, and individuals experiencing homelessness. CCMS received data from internal and external databases and local spreadsheets, integrated the data, and presented client profiles to authorized users via a WPC interface. As part of WPC, consultants evaluated current and desired functions, and a request for proposal (RFP) was issued to replace the CCMS vendor. After a lengthy process, an Epic data-sharing and care coordination module called “Compass Rose” was embedded within San Francisco’s electronic health record. This module enabled San Francisco and several partners. to collaborate on enrollee care plans via CareLink. Strengths of San Francisco’s data sharing infrastructure included field-based access to care coordination data, partner ability to add notes to case files, and integration of Behavioral Health Services data. Limitations included a restricted database and smaller user group due to relative novelty of the Compass Rose platform among partners; furthermore, county departments were structured as separate organizational entities, which increased the challenge of developing appropriate data sharing agreements and meant that multiple partners continued to maintain separate, internal records (rather than only using the same platform as the LE). Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies (] PILOT IMPLEMENTATION Pilot Enrollment San Francisco’s Pilot enrolled 22,749 beneficiaries between January 2017 and the end of December 2021. The average length of enrollment was 18.1 months. Approximately 71% of enrollees ever disenrolled at some point. The most common reasons for disenrollment were lack of engagement (30% of total WPC enrollment)?> and graduation (20%).?° Enrollee Identification and Engagement San Francisco identified and auto-enrolled beneficiaries using a data-driven approach within their CCMS records. Individuals experiencing homelessness had a flag placed on their service record whenever an HSH homeless specific service like shelters, navigation centers, housing assessment, or housing case management was utilized. In addition, a flag was placed if the individual reported to a health provider that they were homeless. New enrollments and engagement occurred when staff of the county’s Homeless Outreach Team (SFHOT) or Street Medicine and Shelter Health programs met with and enrolled previously unidentified individuals experiencing homelessness. Care Coordination Following engagement, care coordination was the focus, especially the coordination needed to assist individuals experiencing homelessness presenting at several city locations with multiple health and social needs. All HSH programs became WPC participants for engagement and care coordination. HSA provided Medi- Cal enrollment. In SFDPH only programs like the sobering center, medical respite, psychiatric respite, and Street Medicine/Shelter Health, they were WPC affiliated and within them the focus was on their engagement and care coordination services, not direct treatment. Three high-intensity care teams were established to target high need enrollees. Their purpose was to demonstrate the value of interagency care coordination of shared integrated health and social determinants of health information, and to test city-wide care plans. The Shared Priority team helped individuals with behavioral health needs that previously prevented a housing placement; the Fire Department’s Emergency Medical Services EMS6 team used community paramedics to help those with medically complex long-term chronic diseases who frequently used 911 and emergency department (ED) crisis services; and after 2020, the Inter-agency Care Coordination (ICC) team helped enrollees in COVID- 19 shelter-in-place locations move to permanent supportive housing (PSH). On average, care coordination teams had a 1:176 case ratio. Care coordination was supported by weekly multidisciplinary team meetings and case conferences with various city agencies to promote ED diversion, housing placement, and collaborative care delivery. The “shared priority” intervention successfully placed 87% of enrollees in housing. The other teams were ongoing. Housing Assistance San Francisco emphasized a “Housing First” or “housing is a health service” model. The sequence in the redesigned system of care and measurement by metrics was to identify individuals experiencing homelessness, conduct coordinated entry assessment to identify health and social needs using the federal homeless management information system (HMIS) coordinated entry tool, refer to and 25 Beneficiary refused to participate or did not engage in services. 26 Beneficiary achieved desired goals. fe) Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report participate in housing navigation assistance, move into permanent supportive housing, and remain stable longer than six months. Other Services For the duration of WPC, San Francisco paid for engagement and care coordination activities of participating housing and health programs and for building data sharing infrastructure. Potential other services were excluded, such that funds did not pay directly for housing searches, applications to coordinated entry system, tenancy support, security deposits, utilities, motel vouchers, or short-term shelter housing, nor did they directly fund the treatment within medical respite, sobering center, assistance with benefits applications, transportation, and access to legal services. CRITICAL SUCCESS FACTORS Employed a population health perspective to help redesign systems of care for individuals experiencing homelessness, which enabled policy advocacy within San Francisco’s strategic approach to show that housing is a health service and sharing social determinants of health along with health and behavioral data is crucial to success. Increased collaboration and partnership between previously siloed city agencies (e.g., DPH, HSH, HSA, SFFD) through participation in WPC. Selected vendor Epic and its care coordination modules as a platform for sharing integrated data on health and social determinants of health among city-wide WPC partners. Developed engagement, housing prioritization, and housing navigation centers in collaboration with shelters, navigation centers, community-based organizations, and street medicine programs. Demonstrated the value of sharing data on social determinants of health alongside health data when serving enrollees with high health and social needs. There were three targeted care management pilot studies: “Shared Priority” initiative to move high behavioral need clients into permanent supportive housing, “EMS6” program to prevent inappropriate emergency and inpatient utilization by enrollees with high volume of 911 calls, and ICC to move complex need clients from temporary COVID housing into PSH or other housing. Prepared for CalAIM, especially readying enhanced care management, medical respite, sobering center, and housing navigation programs for future services. PERCEIVED IMPACT OF WPC San Francisco perceived an above average impact on the following aspects of WPC Pilot implementation: Improved integration of health, behavioral health, and social services Improved care quality Increased data sharing between LE and WPC partners Improved management of care of high risk and high utilizing populations Identifying enrollees receiving services from more than one system (e.g., medical, behavioral health, social services) Improved collaborative partnerships for program implementation Improved coordination of care for enrollees San Francisco perceived an above average improvement on the following aspects of care for enrollees: Targeted identification, outreach/engagement, and enrollment Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies “If anything, its mended partnerships. Its shown the overlap in our work, and its shown the need and the value of sharing data and also brought up the challenges in sharing data... It's one of those things that we've seen that it's not going to just happen in a vacuum... So just that oversight and that force and that push for integrated data and its’ usage, and that [at] the population level is something that's very much needed... not just working on Whole Person Care, but working across the department to make sure that services were tailored to the individual and that services weren't duplicated and that people were able to access things and make informed decisions as best as possible.” “(As a result of WPC, when the pandemic hit] We had systems in place to quickly gather information that would [previously] have taken a long time. We had partnerships that already existed that really made it easy to [set up multidisciplinary care] in a coordinated way... You had people who were planning it, who'd already been working together, planning based on the Whole Person Care and the shared priority pilot ina way that didn't exist before...Regularly [after the pandemic hit], | felt | was very aware that the relationships that we built largely through Whole Person Care and other related work then meant that if | was talking to somebody and | had to plan something, it was usually not somebody who I'd just met... [There was] Immediate trust there that wouldn't have existed unless we'd been doing that work. So that's the biggest thing.” Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report UCLA CENTER FOR HEALTH POLICY RESEARCH at San Joaquin’s Whole Person Care Pilot PILOT STRUCTURE Target Populations San Joaquin’s primary populations of focus included high utilizers, individuals experiencing homelessness or at risk for homelessness, individuals with severe mental illness, and/or individuals with substance use disorders. High utilizers were defined as those with five or more emergency department visits in the last year. Lead Entity and Partnerships In San Joaquin, the San Joaquin County Health Care Services Agency (HCSA) served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included 25 partners from diverse sectors, 10 of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. Partnerships between the LE and most partners were longstanding; exceptions included the LE’s relationship with Medi-Cal managed care plans (i.e., Health Net and Health Plan of San Joaquin) and the San Joaquin Community Health Information Exchange (SCHIE), which were new as a result of WPC. Relationships with county Behavioral Health Services facilitated timely information sharing and support for WPC enrollees, while relationships with Correctional Health Services helped facilitate referral to WPC as individuals transitioned back into the community. Most care coordination services were contracted out to WPC partners rather than provided directly by San Joaquin. Data Sharing Infrastructure The primary mechanism for data sharing with partners was “Activate Care,” a cloud-based care coordination platform implemented in collaboration with the San Joaquin SCHIE. Strengths included the ability to collaborate and co-manage enrollees across all partner organizations, the customizability of the platform, and inclusion of real-time emergency department and inpatient alerts. Limitations included the need to double-document information across Activate Care and other required platforms, like Cerner, and the inability to automatically integrate data from the justice system (e.g., bookings, release dates). PILOT IMPLEMENTATION Pilot Enrollment San Joaquin enrolled 3,201 beneficiaries by the end of December 2021. The average length of enrollment was 19.7 months. Approximately 72% of enrollees ever disenrolled at some point between January 2017 and December 2021. The most common reason for disenrollment was ineligibility for Medi-Cal (13% of total enrollment).?” Enrollee Identification and Engagement Enrollees were identified through street- or shelter-based outreach, health care facility outreach, referrals, and administrative data. Once a potential enrollee was identified, case managers spent time 27 Beneficiary could no longer remain enrolled in WPC if no longer eligible for Medicaid benefits. Note: “other” disenrollment reason (42% of total enrollment) was not defined by the Pilot and therefore not detailed above. Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies building rapport and addressing basic needs such as food, housing, and clothing. Enrollee engagement varied across enrollees depending on enrollee needs and trust, but enrollees were engaged at a minimum of once per month. Best practices around enrollee engagement included working through a known entity, like a partner agency, to facilitate warm handoffs to potential enrollees. Care Coordination Enrollees were assigned to multidisciplinary care coordination teams that worked in partnership to provide them with needed services. Specific types of staff varied based on the organization(s) with which the enrollee was involved. For example, initial outreach was performed by community health workers, nurses, social workers, mental health counselors, or substance abuse counselors. These individuals, along with medical assistants and housing navigators, also provided care coordination services. Nurses, licensed social workers, mental health and substance abuse counselors, physicians, and nurse practitioners were also available for clinical consult. Most teams had a 1:75 case ratio. Clearly identified points of contact within partner agencies and integration of data via the Activate Care platform were critical for ensuring provision of high-quality coordinated care. San Joaquin required partners to work closely with enrollees in developing a care plan, but allowed partner agencies to use their own approach for assessing enrollee social needs. Housing Assistance San Joaquin emphasized a “Housing First” model that prioritized finding enrollees stable housing, whether through a shelter, temporary shelter, Section 8 housing, shared housing, or recuperative care. WPC funds were used to assist with providing medical respite to individuals experiencing homelessness and providing short-term housing in a shelter. Other Services In addition to care coordination and housing services, San Joaquin also provided medical respite, sobering center services, and transportation to services/appointments. CRITICAL SUCCESS FACTORS ¢ Clearly identified points of contact within each partner agency supported accountability for care coordination, data integration, and service provision. «Time spent establishing rapport with enrollees during initial outreach facilitated subsequent engagement for some enrollees. e Integration of multiple data sources and systems via the Activate Care platform was critical for improved inter-organizational collaboration and care coordination. ¢ On-site, pre-release connections with incarcerated individuals helped reduce enrollee recidivism in the justice system and were only possible due to strong relationships between San Joaquin and Correctional Health Services. PERCEIVED IMPACT OF WPC San Joaquin perceived an above average impact on the following aspects of WPC Pilot implementation: e Reducing inappropriate emergency department visits and hospitalizations © Decreased overall cost of care e Increased data sharing between LE and WPC partners «Improved collaborative partnerships for program implementation Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report San Joaquin perceived an above average improvement on the following aspects of care for enrollees: e Extent to which care provided is patient-centered ¢ Overall enrollee well-being “We coordinate with the enrollee, first and foremost... | may very well have all these different resources for the enrollee, but at the end of the day, the enrollee will direct me on what they feel is their need. | may feel that they need housing or | might feel that they need medical intervention or psychiatric. But if the enrollee is not open to those services, | have to meet the enrollee where they're at [and work on] what our enrollee feels is in the best interest of themselves.” “Whole Person Care has really helped a lot of people get to their next step in life. And it just wasn't one case manager assigned to them. It took a Whole Person Care village. It takes BHS Program Manager's team, my team, Public Health team, somebody from HSA's team, it takes different organizations, the Housing Authority's team, to all work together to help that enrollee, that individual, get their needs met to write a new chapter in their life. So that's what Whole Person Care did, it built those relationships to improve people's lives.” Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies UCLA CENTER FOR HEALTH POLICY RESEARCH at San Mateo’s Whole Person Care Pilot PILOT STRUCTURE Target Populations San Mateo’s primary population of focus was high utilizers, which they defined as individuals with four or more emergency department (ED) visits in the last 12 months and experiencing homelessness or affected by mental health challenges and/or substance use disorder. Lead Entity and Partnerships In San Mateo, the San Mateo County Health System served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included eight partners, five of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. A key focus of San Mateo was to address internal silos within county systems; partnerships among health system partners were strengthened as a result of WPC and facilitated enrollee connection to needed services. Relationships with community entities such as Brilliant Corners were new and facilitated outreach and access to housing support for enrollees experiencing homelessness. Care coordination services were primarily provided in-house with a smaller portion served by contracted partners. Data Sharing Infrastructure The primary mechanism for data sharing was a local health information exchange (HIE) that integrated electronic health record data from five divisions within the county (i.e., San Mateo Medical Center, Behavioral Health and Recovery Services (BHRS), Correctional Health Services (CHS), Adult and Aging Services, and Family Health). Strategic partners such as the Health Plan and local area hospitals, as well as community-based organizations (CBOs), were also integrated into the HIE. Strengths of the HIE included care team access to data on enrollee medical health, behavioral health, and social determinants of health (e.g., housing status, incarceration history) and real-time notifications when enrollees utilized the ED. Limitations included lack of field-based access to the HIE by all members of the care team and care coordinators’ inability to input data directly into the HIE. PILOT IMPLEMENTATION Pilot Enrollment San Mateo enrolled 4,163 beneficiaries by the end of December 2021. The average length of enrollment was 26.1 months. Approximately 66% of enrollees ever disenrolled at some point between January 2017 and December 2021. As an “opt-out” county that auto-enrolled, initially based upon data and thereafter at point of referral, the most common reasons for disenrollment were ineligibility for Medi-Cal (45% of total WPC enrollment)?* and WPC services no longer needed (15%).?° Enrollee Identification and Engagement Eligible enrollees were identified using administrative data (e.g., on ED visits) as well as through affiliation with existing programs serving the target population and auto enrolled at the beginning of the 28 Beneficiary could no longer remain enrolled in WPC if no longer eligible for Medicaid benefits. 23 Beneficiary was not appropriate or did not benefit from the services provided. (esq) Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report pilot. Thereafter, program affiliation and referrals from health system partners, field-based outreach teams, and community partners were also used to engage clients into WPC services. Care Coordination San Mateo supported/implemented multiple care coordination programs, including Bridges to Wellness (BTW) which provided intensive case management and linkage of individuals experiencing homelessness and co-occurring mental health and/or substance use disorders to primary care and behavioral health homes. The Integrated Medication Assisted Treatment (IMAT) team connected enrollees to needed substance use treatment services, as well as mental health services. Enrollees were assigned to teams by a triage nurse following consideration of enrollee needs and acuity or were enrolled into the team based upon outreach efforts. Teams included community health workers, nurses, social workers, mental health and alcohol or drug counselors. Nurse practitioners, psychiatrists, social workers, and alcohol or drug counselors were available for clinical consult. Enrollees were supported by multiple care coordinators across WPC partners who communicated with each other, as needed. Average case ratios varied across teams. For example, BTW teams served highest-risk utilizers and provided intensive care management with an average 1:10 case ratio while BHRS IMAT teams provided alcohol and drug-related care coordination, with an average 1:30 case ratio. Care coordination was supported by weekly multidisciplinary care team meetings to promote team-based care and collaborative care delivery. The Bridges to Wellness team piloted the use of a comprehensive assessment to identify needs in multiple domains including medical, mental health, housing, substance use, and social service needs. That team also piloted the use of the Patient Activation Measure (PAM) and Coaching for Activation to support chronic disease management. Housing Assistance San Mateo enhanced by the ability to provide housing location services and direct housing subsidies for unhoused persons through the uses of local dollars. They emphasized a “Housing First” model and worked with partners to identify the most vulnerable members through the use of a modified Vulnerability Index Service Prioritization Decision Assistance Tool (VI -SPDAT). San Mateo was able to leverage the housing location services to successfully house individuals awarded mainstream and permanent supportive housing vouchers. Other Services In addition to care coordination and housing services, San Mateo also provided recuperative care and provided additional staffing at the sobering center. CRITICAL SUCCESS FACTORS e Multiple teams of internal and external care coordinators provided diverse identification, outreach, and engagement opportunities for eligible enrollees. e Real-time notifications when enrollees utilized the ED enabled swift interception of enrollees and navigation to appropriate services. « Dedicated care coordinator to facilitate continuity of services for WPC enrollees during re- entry into community from jail. © Continued field-based outreach fostered trust between clients and care coordinators. ¢ Stratified care coordination programs based on client acuity and specialized need enabled delivery of drug and alcohol treatment when needed. Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies PERCEIVED IMPACT OF WPC San Mateo perceived an above average impact on the following aspects of WPC Pilot implementation: ¢ Improved enrollee health and well-being e Increased data sharing between LE and WPC partners « Improved management of care of high risk and high utilizing populations e Improved collaborative partnerships for program implementation e Improved coordination of care for enrollees San Mateo perceived an above average improvement on the following aspects of care for enrollees: e Access to affordable housing e¢ Comprehensiveness of available services (health, behavioral health, and/or social services) e Targeted identification, outreach/engagement, and enrollment “We are seeing increased communication between teams and programs... We've made it known to partners what resources are available for what services, so people know who to reach out to... And Whole Person Care has created forums where people can communicate and coordinate services for clients such as the complex case conferences or the operating committee meetings where different programs can present the work that they are doing so that others can learn of their existence and know what services they offer. And this has created a lot of linkages between programs that has been really important.” “The best practices are really around meeting the client where they're at and doing field-based work, accepting the client, accepting what it is they want out of services. So obviously, you don't walk up to people and say, ‘Hey, you've got a methamphetamine problem, let's deal with that.’ You first walk up to people and say, ‘Hey, | see that your life might not be going so well. What is it that you'd like to work on?” It really is best practices around accepting people, treating them with respect and asking them what it is they want. And surprisingly, when you do these very simple things, people will often engage and say, ‘You know what | really need is? | need some food.’ So, if you're able to say, ‘Hey, here's a $25 gift card for Safeway, get yourself some food. And hey, can! call you tomorrow?’ So, it really is those known best practices around meeting people where they're at, respecting people and offering them the services that they're asking for.” Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report UCLA CENTER FOR HEALTH POLICY RESEARCH at Santa Clara’s Whole Person Care Pilot PILOT STRUCTURE Target Populations Santa Clara’s primary population of focus was high utilizers, defined as those engaged in two or more systems of care and in the top 5% of utilizers for emergency, inpatient, and urgent care over the past year. Lead Entity and Partnerships In Santa Clara, Santa Clara Valley Health and Hospital System (SCVHHS) served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included 43 partners from diverse sectors, 30 of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. Partnerships between the LE and its county partner, Behavioral Health Services Department (BHSD), was longstanding and facilitated mental health triage, drug recovery, and behavioral health support during WPC. Relationships with many other entities, such as Children Family Services, O’Connor Hospital, Saint Louise Hospital, De Paul Health Center, and Custody were new and developed to facilitate social services, medical and physical health, and justice. Most care coordination services were provided by Santa Clara and one-third of services were contracted to external community clinics and other external partners. Data Sharing Infrastructure Santa Clara adopted a strategic approach for data integration; Santa Clara utilized a single health record and associated database where possible (e.g., SCVHHS, associated medical and behavioral health care facilities). Supplemental integration alternatives were offered for other partners where appropriate. Data sharing infrastructure included a WPC database (developed to track demographics and service utilization of enrollees); this was connected to the county health system’s electronic health record (EHR), homeless management information system (HMIS), and partner EHR systems. Advanced workflows for care coordination utilized “Epic’s HealthLink,” which notified care coordinators of enrollee admissions to emergency department, hospital, and psychiatric services; other features included identification of eligible enrollees, development of care plans, and tracking of interventions. Strengths of Santa Clara’s data sharing infrastructure included the ability to incorporate data from the HMIS database, comprehensive access to enrollees’ health records, and compatibility with Tableau for data analysis and visualization for real-time data dashboards. Limitations included maintenance of multiple data systems for WPC partners as not all community health center partners utilized Epic, and entries for social determinants were not standardized. PILOT IMPLEMENTATION Pilot Enrollment Santa Clara enrolled 7,431 beneficiaries by the end of December 2021. The average length of enrollment was 21.4 months. Approximately 80% of enrollees ever disenrolled at some point between Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies January 2017 and December 2021. The most common reasons for disenrollment were loss of Medi-Cal coverage (31% of total WPC enrollment)?° and graduation (22%).3? Enrollee Identification and Engagement Santa Clara initially utilized an opt-in enrollment process and identified eligible individuals by referral through lists provided by the county-operated Valley Health Plan (VHP), a delegate of Santa Clara Family Health Plan for Medi-Cal managed care. However, difficulty locating eligible enrollees resulted in an additional strategy of identifying and engaging individuals at community health centers, emergency departments, and emergency psychiatric departments. In these settings, care coordinators met with individuals in-person to enroll them into WPC. A standardized assessment for WPC eligibility was built into Epic HealthLink to determine an individual’s eligibility. On average, care coordinators interacted with enrollees three times within the first month of enrollment and tapered to once per month during the rest of enrollment. Care Coordination Enrollees were assigned to diverse care coordination teams led by various roles dependent on client need. Teams included community health workers with lived experience, registered nurses, complex care nurses who often assumed the role of care coordinators, and pharmacists who aided in medication adjustment. Care coordination at most sites was provided by a single, dedicated care coordinator who followed an enrollee across care settings and worked with social support agencies. Where appropriate, care coordination was provided by a team with relevant and specialized expertise, especially those working with homeless persons at Valley Health Homeless Program (VHHP). Moreover, a deliberate effort was made to contract with community clinics that were trusted providers serving ethnically diverse communities. Most teams had a 1:30 case ratio, while those providing more intensive care management services had caseloads between 10-20 enrollees. Care coordination was supported by multidisciplinary team meetings and functionality within the data-sharing platform to promote team-based care and collaborative care delivery. Comprehensive assessment of all identified social needs was conducted using several instruments, including Screening, Brief Intervention and Referral to Treatment (SBIRT); Patient Health Questionnaire-9 (PHQ-9); and for homeless patients, the Vulnerability Index — Service Prioritization Decision Assistance Tool (VI-SPDAT). Housing Assistance Santa Clara emphasized a “Housing First” model and partnered with housing specialists within the community. Under contract with the Institute on Aging (IOA), WPC patients received managed care plan (MCP) authorized funds to assist with minor housing improvements. Under contract with a supportive housing subcontractor, this supported those seeking housing and those in housing who needed ongoing assistance with enrollee-landlord relationships and other services. Medical respite provided housing and care for those who were homeless and needed a safe place to recover before returning to the streets. During the pandemic, Federal Emergency Management Agency (FEMA)-funded hotel vouchers were provided to patients experiencing homelessness. Furthermore, the Aunt Bertha platform was funded to locate needed resources including housing, furniture, food, and other social needs for the WPC patients. 39 Loss of Medi-Cal coverage was an “other” reason defined by the Pilot as: “enrollees who have lost Medi-Cal eligibility when in custody”. 31 Beneficiary achieved desired goals. Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report Other Services In addition to care coordination and housing services, Santa Clara also piloted a self-referral mental health respite, resources for drug recovery, transportation, and phones (loaded with the MyHealthOnline app and care coordinator and primary care provider (PCP) office numbers). CRITICAL SUCCESS FACTORS e Developing appropriate data sharing infrastructure and agreements was viewed as critical to care coordination and program management. Santa Clara developed a “Trust Community” between WPC partners to facilitate data sharing and as a result, was able to execute data use agreements with all key partners. « Clearly defined roles for community health workers within the care teams at hospitals, Community Clinics, and Ambulatory Care Clinics was seen as a necessary step to integrate a key role into the enrollee’s care team. In some settings, community health workers were welcomed with little resistance, however, the integration of community health workers into hospital teams was more challenging. e Reduction of Emergency Department admissions/readmissions was accomplished by utilizing the community clinic navigators, peer respite, and sobering center programs. Patients in peer respite had zero hospital admissions and reduced emergency department visits for the six months after self-referral to Peer Respite. The sobering center (located across from the main jail) was an alternative for intoxicated patients brought in by local police departments. © Strong partnerships and cross-sector collaboration with county agencies, medical centers, and community partners expanded the resources available to enrollees and increased referrals to wPec. PERCEIVED IMPACT OF WPC Santa Clara perceived an above average impact on the following aspects of WPC Pilot implementation: e Improved integration of health, behavioral health, and social services e Identifying enrollees receiving services from more than one system Santa Clara perceived an above average improvement on the following aspects of care for enrollees: « Comprehensiveness of available services e Targeted identification, outreach/engagement, and enrollment © Overall enrollee well-being “It's just more about [being] patient-centered, focusing on what the patient needs. | think one of the ways that I've seen [a] more powerful reach is for those providers that have flexible service models. So not just on the phone, not just in the clinic, but they're also going into the community or to home visits, not for a hundred percent of the population, but for the ones that are really challenging to reach, that they're willing to go to those lengths. And then there's more success with being able to engage. “| think the care coordination needs to meet them where they're at physically, emotionally, educationally, and in their stability level, based on either behavioral, substance, physical health issues across the board. That's the adaptability that we need to do, [it’s] what it takes to get them to that next level.” Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies UCLA CENTER FOR HEALTH POLICY RESEARCH at Santa Cruz’s Whole Person Care Pilot PILOT STRUCTURE Target Populations Santa Cruz’s primary populations of focus included with severe mental illness/substance use disorders (SMI/SUD) and at least two other specific criteria related to chronic physical conditions and/or homelessness. Identification of SMI/SUD was based on diagnosis, receipt of behavioral health services, or client self-report. Lead Entity and Partnerships In Santa Cruz, the county Health Service Agency (HSA) served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included 18 partners from diverse sectors, eight of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. All partnerships between the LE and partner entities existed prior to WPC. Partnerships with other county divisions (e.g., Behavioral Health, Public Health, Human Services, Probation) facilitated data sharing and enrollee service access. Partnerships with community-based partners, such as Front Street, facilitated enrollee access to behavioral health services. All care coordination services were provided directly by Santa Cruz through multidisciplinary teams within the county’s Integrated Behavioral Health program. Data Sharing Infrastructure The primary mechanism for sharing care coordination data with partners was “Together We Care,” an electronic case management platform procured specifically for WPC and embedded within the county’s Health Information Exchange (HIE). Together We Care was comprised of two components: a care coordination platform provided via a contract with Activate Care, and a closed loop social service referral system provided by United Us. Prior to implementation of Together We Care, Santa Cruz used separate electronic health records (EHRs), Epic and Avatar, to share medical and behavioral health data with county partners, and Excel and Access databases to share data with community-based partners. Strengths of Santa Cruz’s data sharing infrastructure included field-based access to enrollee medical information, real-time notifications when enrollees entered emergency departments, and prior existence of an HIE, which facilitated development of new data sharing agreements needed to implement Together We Care. Limitations included lack of field-based access to behavioral health data, need for dual data entry (e.g., in Together We Care and system-specific EHRs), and need for intensive training and socialization to promote uptake of Together We Care. 32 These criteria included: Two or more chronic health conditions (e.g., diabetes, hypertension, COPD); prescribed five or more medications for chronic health conditions; homeless or at risk for homelessness; four or more psychiatric hospitalizations in a 12-month period; two or more medical hospitalizations in a 6-month period; institutional living in the last 12-months or currently living in an IMD or jail. Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report PILOT IMPLEMENTATION Pilot Enrollment Santa Cruz enrolled 603 beneficiaries by the end of December 2021. The average length of enrollment was 33.0 months. Approximately 69% of enrollees ever disenrolled at some point between January 2017 and December 2021. The most common reasons for disenrollment were lack of engagement (50% of total enrollment)?3 and deceased (7%).** Enrollee Identification and Engagement Santa Cruz’s primary strategy for identifying eligible enrollees was through referrals from partner organizations, enrollee self-referral, referrals from primary care providers, or other care coordinators within the Health Services Agency. Eligible beneficiaries were enrolled at point of care (e.g., healthcare facilities, community sites, home, or encampment visits) or via telephone. Care Coordination Enrollees were assigned to a multidisciplinary care coordination team. Teams were led by licensed social workers, but also included a community health worker to provide peer support and coaching, and a housing navigator. Care coordination was provided by a single, dedicated care coordinator (non- licensed case manager) who followed enrollees across all participating WPC care settings. Most teams had a 1:30 case ratio. Care coordination was supported by weekly one-on-one supervision and multidisciplinary team meetings, and monthly meetings with emergency department staff to promote team-based care and collaborative care delivery. Comprehensive assessment of all identified social needs was conducted using the Vulnerability Index — Service Prioritization Decision Assistance Tool (VI-SPDAT) and the Protocol for Responding to and Assessing Patients’ Assets, Risks, and Experiences (PRAPARE). Housing Assistance Santa Cruz emphasized a “Housing First” model. For enrollees with housing assistance needs, Santa Cruz assisted with completing applications to the Coordinated Entry System, obtaining housing funds (e.g., housing choice vouchers or rental subsidies) and ongoing assistance with enrollee-landlord relationships once housed. WPC funds were used at partner organizations to provide tenancy support, housing search, fund security deposits and furniture needs, and medical respite. Other Services In addition to care coordination and housing services, Santa Cruz also provided benefits enrollment assistance, medical respite, transportation assistance, and health education. CRITICAL SUCCESS FACTORS e Early establishment of data sharing agreements and releases of information with all partners enabled WPC care coordinators to build relationships with other case management providers and more quickly coordinate care for enrollees. e Robust referral pathways between partners and WPC staff in healthcare settings mitigated need for extensive outreach. 33 Beneficiary refused to participate or did not engage in services. 34 Beneficiary died. Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies e Contracting with partners with established relationships with other county divisions allowed Santa Cruz to efficiently coordinate care. «Partnerships strengthened through WPC helped improve the culture of process improvement in the county. « Case managers developed strong rapport with and acted as communication liaisons between external partners which supported data sharing in the absence of a single integrated data system. PERCEIVED IMPACT OF WPC Santa Cruz perceived an above average impact on the following aspects of WPC Pilot implementation: « Increased data sharing between LE and WPC partners «Improved management of care of high risk and high utilizing populations e Improved collaborative partnerships for program implementation e Improved coordination of care for enrollees Santa Cruz perceived an above average improvement on the following aspects of care for enrollees: * Coordination of care e Access to affordable housing “We have people that are happy, that are healthier, that are living much more purposeful lives now than they were before when they were just surviving or when they weren't prioritizing their health, not accessing healthcare... These are things that they now can do with our support and with our coaching.” “Whole Person Care has given us an opportunity to work more robustly in [our county Health Services Agency] over behavioral health, public health, our county federally qualified health centers, and environmental health... Even though we're all one agency, there's still silos... Whole Person Care has helped us bridge some of those.” Eg Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report UCLA CENTER FOR HEALTH POLICY RESEARCH at Shasta’s Whole Person Care Pilot PILOT STRUCTURE Target Populations Shasta’s primary population of focus was high utilizers, who were homeless or at-risk of homelessness. High utilizers were defined as those with two or more emergency department (ED) visits, or one inpatient stay in the previous three months. Lead Entity and Partnerships In Shasta, the County Health and Human Service Agency served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included 15 partners from diverse sectors, nine of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. Partnerships between the LE and community clinics were longstanding and facilitated effective coordination of care during WPC. Additionally, Shasta had the support of Health Alliance of Northern California, a local network of community clinics and health centers, which facilitated early buy-in and collaboration from these community partners for WPC. Relationships with several other entities, such as Aegis Treatment Center and Dunamis Wellness Center were new, and developed to facilitate substance use treatment for enrollees during WPC. Care coordination services were contracted to two primary care providers within the community, Hill Country Health and Wellness Center and Shasta Community Health Center. Data Sharing Infrastructure The primary mechanism for data sharing with partners was the web-based platform “SharePoint,” which served as a central database for sharing critical care coordination documents (e.g., care plan, referrals) and enrollee profiles (e.g., contact information, medical history). Prior to implementation of SharePoint, staff primarily used spreadsheets, encrypted emails, and paper documents to share information. Strengths of Shasta’s data sharing infrastructure included use of a central storage location for all enrollee information and cloud-based application, which allowed real-time updates and access (e.g., in field from phone or laptop). Limitations included competing demands with partners’ time, which created challenges to prioritize the data sharing needed for reporting purposes. PILOT IMPLEMENTATION Pilot Enrollment Shasta enrolled 581 beneficiaries by the end of December 2021. The average length of enrollment was 5.8 months. Approximately 94% of enrollees ever disenrolled at some point between January 2017 and December 2021. The most common reasons for disenrollment were “other” (29% of total WPC enrollment)*> and lack of engagement (20%).°° Enrollee Identification and Engagement Primary methods for identifying enrollees were street/shelter-based outreach and referrals from WPC partner agencies, which were effective for facilitating handoffs and ensuring appropriate fit (i.e., 35 Shasta defined “other” as “inability to find or contact beneficiaries that were administratively enrolled”. 36 Beneficiary refused to participate or did not engage in services. Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies partners had awareness of enrollment criteria and program goals). Shasta utilized a three-part enrollment process: referrals were screened for basic eligibility; then if eligible, were reviewed by the WPC care team. If deemed an appropriate fit for the program, the enrollee was administratively enrolled. Staff were then provided with a 30-day window to contact and engage enrollees and complete the comprehensive care plan. Care Coordination Care coordination services are provided through multidisciplinary “teamlets,” which included case managers (some which were master’s level), nurses located in partner Federally Qualified Health Centers (FQHCs), and a housing case manager who provided social work and benefits support. There were multiple care coordinators within a care coordination team based on availability or expertise who followed enrollee across all WPC-participating care settings. Some care coordination services were contracted out through the FQHCs. Most teams had a 1:23 case ratio. Care coordination staff used multidisciplinary team meetings and case review conferences to communicate and collaborate on enrollee care. Comprehensive assessment of all identified social needs was conducted using the Vulnerability Index — Service Prioritization Decision Assistance Tool (VI-SPDAT). Housing Assistance Shasta emphasized a “Housing First” model and utilized some staff who had lived experience with homelessness to provide housing/supportive services to enrollees. All housing support items (e.g., tenancy support, completed Coordinated Entry Applications, obtaining housing funds, and housing search) were provided by WPC partner organization(s) using WPC funds. No direct housing resources and services were provided by the LE (e.g., funds for utilities, landlord incentives, short-term housing in a shelter), but instead by a combination of partner organizations with and without WPC funds. Other Services In addition to care coordination and housing services, Shasta also provided access to benefits, employment assistance, transportation, and health education. Despite significant investment in planning and development of a sobering center, the Pilot was only able to provide this service to a handful of enrollees due to partners’ inability to adhere to contract agreements. CRITICAL SUCCESS FACTORS * Design of WPC was iterative, collaborative, and involved frontline workers’ input, allowing for lessons learned to be integrated into implementation and for partners to work across sectors in developing WPC. Through this process, partners were able to recognize shared values and goals. Shasta emphasized human centered service design. « Strong collaborative relationships between frontline staff helped facilitate informal data sharing despite difficulty developing a more robust, formal data sharing platform. Furthermore, Shasta had a very hands-on approach with participating partners, training them in appropriate documentation and importance of reporting quality. @ WPC served as an opportunity to highlight siloed approaches to care and gaps in existing services, while providing an opportunity for partners to understand the interconnectedness of housing and medical needs through structured conversations and initiatives. Ea Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report PERCEIVED IMPACT OF WPC Shasta did not report an above average impact of WPC at the Pilot level (e.g., decreased cost of care, improved management of high-risk populations, improved data sharing between LE and WPC partners), but Shasta perceived an above average improvement on the following aspects of care for enrollees: *® Coordination and continuity of care e Access to affordable housing Comprehensiveness of available services (e.g., health, behavioral health, and/or social services) e Targeted identification, outreach/engagement, and enrollment Frequency and quality of communication with enrollees e Extent to which care provided was patient centered “| was new to this pilot two and a half years ago when | came on and | was able to see a lot of things | hadn't seen before and it still resonates with me that this pilot was designed from the ground up by people who were going to be doing the direct services. Like it wasn't a top-down kind of design, and the fact that everybody was weighing in on how the services would be provided at every level, and it was super collaborative from the very start, | think it's really novel and it's part of what made it so effective.” “| would say that the other way that Whole Person Care has changed our services is that we have expanded our language. When we first put everybody at the table, we might've been saying the same words, but we were not talking about the same thing. So we really had to learn medical language. What are you saying? What do you mean? housing language, what are you saying? What are you meaning?... So, just really trying to create a common ground where we could meet people and then account for the enrollee voice and choice.” Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies rue Small County Collaborative’s Whole Person HEALTH POLICY RESEARCH um Care Pilot PILOT STRUCTURE Target Populations Primary populations of focus included high utilizers and individuals with severe mental illness and substance use disorders (Mariposa), and high utilizers and individuals at risk for or experiencing homelessness (San Benito). In both Mariposa and San Benito, high utilizers were defined as individuals with three or more emergency department visits and/or one or more hospitalizations. Lead Entity and Partnerships Small County Whole Person Care Collaborative (SCWPCC) served as the lead entity (LE) responsible for program implementation and reporting to the state. The SCWPCC originally included three county entities (Mariposa Human Services Agency, Plumas Behavioral Health, and San Benito Health and Human Services Agency), but Plumas chose to discontinue participation prior to implementation due to significant leadership turnover, concern over administrative burden, and lack of partner support in the broader community. Mariposa and San Benito discontinued WPC participation in PY 6 (2021). As of January 2020, the Pilot included 11 partners from Mariposa and 10 partners from San Benito. Partners came from diverse sectors. In Mariposa, nine partners were identified as having high awareness of and active involvement in implementing WPC, while in San Benito, seven partners were identified as having high awareness of and active involvement in implementing WPC. Both counties identified partnerships that were critical for facilitating referrals and outreach (in Mariposa, the Alliance for Community Transformations and John C. Fremont Healthcare District; in San Benito, the County Department of Behavioral Health and Hazel Hopkins Hospital). In both counties, the partnership with Anthem, one of the Medicaid managed care plans (MCPs), was also described as helpful, due to use of MCP case management staff to help link enrollees to additional services not provided by WPC and assist with hand-offs following enrollee graduation from WPC. In San Benito, a new partnership with California State University - Monterey enabled hiring of masters-level social work students to help staff the program. Data Sharing Infrastructure The primary mechanism for data sharing with partners was an e-Client management system which helped consolidate enrollee data into a single tool and facilitated data exportation and reporting. Strengths included the ability to customize the tool from its inception and ease staff use. Limitations included challenges with the e-Client management system vendor resulting in limited functionality of the tool and time-intensive double-checks until the system was finalized. Lessons learned included the importance of selecting an experienced vendor when creating a data management system, and the benefits of working with a highly skilled evaluator to assist in the design for increased efficiency in the data entered and extracted from the system. PILOT IMPLEMENTATION Pilot Enrollment SCWPCC enrolled 143 beneficiaries by the end of December 2020. The average length of enrollment was 10.9 months. Approximately 99% of enrollees ever disenrolled at some point between January 2017 Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report and December 2020. The most common reasons for disenrollment were graduation (37% of total WPC enrollment)?” and lack of engagement (17%).2 SCWPCC chose to discontinue WPC at the end of 2020 and did not enroll in 2021. Enrollee Identification and Engagement Outreach to potential enrollees was based on partner referral and review of Pilot databases. Lack of pre- screening by partners prior to referral was identified as a challenge. Prospective enrollees were not automatically enrolled; instead, case managers would spend several sessions getting to know enrollees to build trust and begin to work toward enrollee-identified goals, sometimes even prior to official enrollment. Time spent on building rapport prior to enrollment was perceived as critical for assessing prospective enrollees’ motivation to participate, and in facilitating engagement following enrollment. Care Coordination In Mariposa, care coordinators were co-located within county behavioral health, social services, and public health, and enrollees were assigned to specific care coordinators based on primary type and complexity of need. Specific types of staff responsible for care coordination included medical assistants, nurses, substance use counselors, or mental health counselors. In San Benito, enrollees were assigned to a single, dedicated care coordinator that followed them across participating care settings. Staff providing care coordination services in San Benito included unlicensed social workers or social work students, housing navigators, benefits support staff, and office and vocational assistants. Most care coordinators in Mariposa had a 1:10 case ratio, while those in San Benito had a 1:13 case ratio. Enrollees were tiered based on complexity of need, with more experienced care coordinators assigned higher complexity enrollees. Care coordination was supported by regular care coordination team meetings, led by the care coordinator, to promote team-based care and encourage collaborative care delivery amongst relevant partners. Comprehensive assessment of all identified social needs was conducted using a social needs screening tool adapted from another WPC pilot program as well as the Vulnerability Index — Service Prioritization Decision Assistance Tool (VI-SPDAT) to assess housing needs. Housing Assistance Mariposa emphasized a “Housing First” approach and worked to streamline processes or programs affecting financing and delivery of housing services, promote policy and legislation to increase housing availability, and co-locate housing services with other service programs. In Mariposa, WPC funds were used to assist with funds for furniture, appliances, home items, utilities, housing improvements specific to health needs, landlord incentives, and ongoing assistance with enrollee-landlord relationships even after enrollees were housed. San Benito emphasized streamlining processes around delivery of housing services and workforce training in housing navigation. In San Benito, WPC funds were used to assist with legal support for issues related to housing/tenancy issues, ongoing assistance with enrollee-landlord relationships even after enrollees were housed, providing motel vouchers or their equivalent to cover a few days stay, and providing short-term housing in a shelter. CRITICAL SUCCESS FACTORS 3” Beneficiary achieved desired goals. 38 Beneficiary refused to participate or did not engage in services. Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies al e Sharing resources around professional development, evaluation, and implementing processes for problem solving in real time facilitated collaboration within SCWPCC. e Assigning one person to manage data reporting and quality management was key to consolidating multiple streams of information and ensuring reporting deadlines were met. e Partnerships with local service providers and Medi-Cal managed care plans facilitated continuity of care, e.g., by allowing for warm hand-offs as enrollees transitioned out of WPC. e Investing time in multiple visits with individuals before they enrolled in the Pilot fostered positive rapport and trust with care coordinators, which in turn supported ongoing enrollee engagement. « Policy incentives requiring local hospitals to discharge clients experiencing homelessness into supportive environments improved partnerships with local health systems more than financial incentives. PERCEIVED IMPACT OF WPC Both counties in SCWPCC perceived an above average impact on the following aspects of WPC Pilot implementation: e Improved integration of health, behavioral health, and social services «Improved care quality *® Reducing inappropriate emergency department visits and hospitalizations © Decreased overall costs of care e Identifying enrollees receiving services from more than one system Mariposa perceived an above average improvement on the following aspects of care for enrollees: ¢ Coordination and continuity of care © Access to needed services «Access to affordable housing e Targeted identification, outreach/engagement, and enrollment Frequency and quality of communication with enrollees e Extent to which care provided is patient-centered San Benito perceived a high level of improvement on the following aspects of care for enrollees: ¢ Coordination and continuity of care e Targeted identification, outreach/engagement, and enrollment e Frequency and quality of communication with enrollee e Extent to which care provided is patient-centered “The thing | want to underscore is just how profound the changes were for people who've struggled in their lives... An untrained observer would think there's no hope for that person, and the compassionate skilled staff in these counties helped that transformation happen. It's just remarkable... People not going to the ER anymore, not being hospitalized, or getting care for things that they needed to get care for that they hadn't before...That transformation, | think, is really the big success story of Whole Person Care.” “It's not monetary incentives that are the most effective. In San Benito, they started getting referrals like crazy after the passage of [new legislation, SB-1152] ... where basically, the hospital couldn't discharge a client if they were homeless without ensuring that that client was being discharged somewhere where Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report they could have shelter. Once that law came into place... the hospital came to the table in a big way. So those regulatory incentives, | think, are far more effective than the cash incentives.” Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies UCLA CENTER FOR HEALTH POLICY RESEARCH , «5 as it Solano’s Whole Person Care Pilot PILOT STRUCTURE Target Populations Solano’s primary populations of focus included high utilizers and individuals with severe mental illness/substance use disorders. Lead Entity and Partnerships In Solano, Solano County Health and Social Services (SCHSS) served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included 12 partners from diverse sectors, eight of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. Partnerships between the LE and other county departments (e.g., Behavioral Health and Family Health Services) and medical centers (e.g., La Clinica de la Raza and North Bay Medical Center) facilitated referrals and enrollee access to services during WPC. Care coordination services were initially contracted to Solano Coalition for Better Health and later to Bay Area Community Services (BACS). Solano discontinued participation in Whole Person Care in PY 6 (2021). Data Sharing Infrastructure The primary mechanisms for data sharing with partners were “ETO,” a case management platform and manual queries from Avatar (an electronic health record for behavioral health), NextGen (an electronic health record for primary care), and the Homeless Management Information System (HMIS). Strengths included a shared system for electronically documenting and sharing case notes, a behavioral health component within the system, and ability to document client progress. Limitations included that not all contracted partners had access to all systems and difficulty managing multiple systems. PILOT IMPLEMENTATION Pilot Enrollment Solano enrolled 247 beneficiaries by the end of December 2021. The average length of enrollment was 14.2 months. Approximately 93% of enrollees disenrolled at some point between January 2017 and December 2021. The most common reasons for disenrollment were lack of engagement (36% of total WPC enrollment)?° and graduated (23%).*° Enrollee Identification and Engagement Solano initially used managed care plan (MCP)-generated lists to identify eligible enrollees. However, difficulty connecting with eligible enrollees via telephone resulted in a shift to requesting referrals from medical centers. Solano’s team developed successful relationships with discharge planners at medical centers who alerted WPC staff about eligible enrollees. WPC staff met with individuals in person prior to discharge to share WPC program details. 39 Beneficiary refused to participate or did not engage in services. 40 Beneficiary achieved desired goals. (syd) Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report Care Coordination Enrollees were assigned to a single, dedicated care coordinator who followed them across all WPC- participating care settings. Care coordination teams were led by a master’s level clinician and included mental health clinicians, housing specialists, peer outreach workers, substance abuse specialists, employment specialists, and public health nurses. Most teams had a 1:35 case ratio. Care coordination was supported by monthly multidisciplinary team decision-making meetings with clients and key stakeholders in their care, as well as one-on-one meetings with partners to promote team-based care and collaborative care delivery. Comprehensive assessment of all identified social needs was conducted using a combination of Patient Health Questionnaire (PHQ)-9, and Acuity Assessment, and if the client score was nine or higher, a Suicide Risk Assessment. Housing Assistance Solano emphasized a “Housing First” model and prioritized temporary housing and stabilization to link enrollees to long-term housing. WPC funds were used to assist with ongoing assistance with enrollee- landlord relationships even after enrollees were housed. Other Services In addition to care coordination and housing services, Solano also provided mental health, substance use abuse, and employment services. CRITICAL SUCCESS FACTORS e Strong relationships with hospital systems provided data on visits to the emergency department and inpatient admissions, which aided with case management. e Low caseloads were perceived as important for enrollees to receive appropriately intensive case management and allowed for WPC teams to focus on core issues for higher need enrollees. © Monthly operations planning meetings with partners to discuss and resolve program issues, such as referral challenges, contributed to successful program management. ¢ One-on-one meetings with partners were viewed as critical to partner success as they generated a sense of ownership and accountability amongst partners. PERCEIVED IMPACT OF WPC Solano did not perceive an above average impact on aspects of WPC Pilot implementation but did perceive a high level of impact on improved enrollee health and improved well-being and collaborative partnerships for program implementation. Solano did not perceive an average improvement on aspects of care for enrollees but did perceive a high level of impact on extent to which care provided is patient-centered and overall enrollee well-being. “A lot of the people that we were meeting and finding weren't trustful of government, of a program. And it took a couple of months before they would trust you and want to engage into the program. And that's where Whole Person Care really helped, because [before] we didn't have the funding to provide... Let's say they had a broken windshield, and we have their windshield fixed, and so now they can drive to work. Something little like that, it was like, ‘Oh, | can trust this program, Okay.’ That helps with engagement.” Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies “(The housing liaison’s] job was to go out in the community and make relationships with landlords, hotels, board, and cares. They went out and made those relationships and when we found a client that needed a hotel stay, [landlords, hotels, board, and cares] would house that person because they knew they were connected to Whole Person Care. Or if somebody needed a room, they already knew a landlord that would take [the client] because they knew that that client had this whole team behind them to help support in whatever they needed... a housing specialist was definitely very important.” Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report UCLA CENTER FOR HEALTH POLICY RESEARCH it Sonoma’s Whole Person Care Pilot PILOT STRUCTURE Target Populations Sonoma’s primary populations of focus included individuals experiencing homelessness or at risk of homelessness, and/or experiencing severe mental illness and/or substance use disorder (SMI/SUD). Sonoma also served high utilizers, individuals with chronic conditions, and justice involvement. High utilizers were defined as individuals who had been to the emergency department or crisis stabilization unit three or more times or had two or more inpatient stays in the last 12 months. Lead Entity and Partnerships In Sonoma, the Department of Behavioral Health within the County Department of Health Services (DHS-DBH) served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included 16 partners from diverse sectors, 11 of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. Partnerships between the LE and other county departments (e.g., health, human services, housing) were longstanding but deepened as a result of WPC. Relationships with other entities, such as West. County Community Services, Reach for Home, and Committee on the Shelterless (COTS) were new, and developed to facilitate homelessness and other social services delivery. Care coordination services were largely contracted to community clinics; however, the contracting process was viewed as administratively cumbersome, and the billing structure often resulted in delayed payments, taking time away from Sonoma to meaningfully work and engage partners. A key lesson learned was that contracts could have been better structured to facilitate key relationships and specify partner involvement, which required additional staffing for management. Data Sharing Infrastructure The primary mechanism for data sharing with partners was IBM’s “Watson Care Manager” case management platform, which was originally envisioned as part of a larger data hub that would integrate enrollee information from multiple systems into a single record. However, the platform was still being developed in the last year of WPC. A strength of the platform was that it allowed for receipt of alerts from county human services and probation; however, it did not allow for data sharing with community partners and was therefore only used for internal (i.e., within-LE) tracking. As a result, Sonoma used several workarounds, relying primarily on word of mouth and relationships with partners and providers to get the necessary information to effectively coordinate care. Specifically, use of a discharge planner and close relationships with jails facilitated informal data sharing. An additional limitation Sonoma faced was the dominant cultural perception that did not support data sharing, as it was thought to be an infringement of patient privacy. PILOT IMPLEMENTATION Pilot Enrollment Sonoma enrolled 4,181 beneficiaries by the end of December 2021. The average length of enrollment was 12.8 months. Approximately 46% of enrollees ever disenrolled at some point between January 2017 Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies and December 2021. The most common reasons for disenrollment were lack of engagement (27% of total WPC enrollment)* and “other” (13%)*. Enrollee Identification and Engagement Eligible enrollees were identified using referrals, primarily from community clinics and health centers, but also from other community-based organizations, county agencies, and the county jail. Outreach occurred largely at shelters where co-located WPC peer outreach workers engaged and enrolled individuals in WPC. When referrals were received, they were reviewed by a clinical health program manager and assigned to a single case manager. Length of enrollment depended on the individual’s progress in achieving agreed upon goals. Care Coordination Once assigned an enrollee, care coordination teams were led by WPC care managers who worked with Federally Qualified Health Center (FQHC) nurses to coordinate care. Care was provided by behavioral health clinicians, eligibility/social service workers, substance use counselors, a nurse practitioner, clinical psychologist, and housing navigator. A single, dedicated case manager was assigned to each enrollee; case managers were matched based on their strengths and the enrollee profile. Most teams had a 1:20 case ratio. Care coordination was supported by multidisciplinary team meetings and case conferences to promote team-based care and collaborative care delivery. Comprehensive assessment of all identified care needs was conducted using Adult Needs and Strengths Assessment (ANSA) comprehensive assessment. Housing Assistance Sonoma emphasized a “Housing First” model and used peer staff as housing navigators (in partner organizations) to provide supportive housing services to enrollees. WPC funds were used to assist with housing deposits, application fees, and moving expenses. Through a community-based partnership, Sonoma was able to provide enrollees with 30 days in a motel following transition from the hospital, or if they qualified due to medical vulnerability, or for short-term placement while awaiting permanent housing. Other Services In addition to care coordination and housing services, Sonoma also provided linkage to public benefits, educational programs to assist with finding employment, sobering center services, transportation, and referrals to legal services. CRITICAL SUCCESS FACTORS e Strategic division of roles involved in “outreach and engagement” and “intensive case management” allowed specialization of staff roles and increased ability to serve clients. « Anemphasis on pairing enrollees with a care coordinator based on background, experience, and demographics facilitated trust and rapport building. e Incentive payments to FQHCs enabled Sonoma to hire and retain nursing staff for outreach, engagement, and case management activities. 41 Beneficiary refused to participate or did not engage in services. * Sonoma defined “other” as “did not meet WPC criteria” (Sein) Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report e Despite lack of data sharing infrastructure, Sonoma relied heavily on discharge planners and informal communication mechanisms to obtain information needed for effective care coordination. PERCEIVED IMPACT OF WPC Sonoma perceived an above average impact on the following aspects of WPC Pilot implementation: ¢ Improved integration of health, behavioral health, and social services e Improved care quality e Improved enrollee health and well-being * Improved collaborative partnerships for program implementation ¢ Improved coordination of care for enrollees Sonoma perceived a high level of impact on the following elements at the enrollee level: © Continuity of care Access to needed services (health, behavioral health, and/or social services) « Comprehensiveness and timeliness of available services e Targeted identification, outreach/engagement, and enrollment “My experience has been everyone has the same goals, but everyone [has been] doing it on their own. And so people are more than happy to collaborate and work together. | think that that's been the biggest thing... | think a lot of clients who fall through the cracks in other ways are able to come into our program and have a little more support.” “| believe that, as a team, we do an amazing job. What | would like to see is after people are with us, that there is a follow through, of more services and people going out and doing more of what we do. But that's where the failure part is, when | think of our team, getting people really stable and just being with them that whole time has been really... It shows a lot. And they're actually really grateful.” Whole Person Care Final Evaluation Report |Appendix U: Pilot Specific Case Studies UCLA CENTER FOR HEALTH POLICY RESEARCH , «5 as it Ventura’s Whole Person Care Pilot PILOT STRUCTURE Target Populations Ventura’s primary populations of focus included high utilizers defined as those with four or more emergency department visits or two or more in-patient visits in the prior 12 months. Lead Entity and Partnerships In Ventura, Ventura County Health Care Agency served as the lead entity (LE) responsible for program implementation and reporting to the state. As of January 2020, the Pilot included 46 partners from diverse sectors, 22 of which were identified as having a high awareness of WPC and as actively involved in implementing WPC. Partnerships between the LE and other county agencies (e.g., Behavioral Health Department, Continuum of Care, Human Service Agency) facilitated pathways to service provision and access to enrollee data during WPC. Relationships with many community-based entities were new and developed to facilitate enrollee referral (e.g., through shelters and recuperative care facilities) and engagement (e.g., at co-located coordinated service events) during WPC. All care coordination services were provided by Ventura, apart from contracting recuperative care services to a community partner. Data Sharing Infrastructure Ventura care coordinators utilized multiple data systems, but the primary mechanism for data sharing with partners was an existing Cerner electronic health record (EHR) which provided county partners point-of-care access to enrollee medical, mental health, and substance use treatment encounter data, and other data related to justice-involvement and housing status. Strengths included field-based care coordinator access to client data, and real-time notifications of emergency room and hospital admissions and discharges at Ventura County Medical Center and Santa Paula Hospital. Limitations included read-only access for Public Health and Behavioral Health partners, and lack of integration with Behavioral Health data system or Homeless Management Information System (HMIS). PILOT IMPLEMENTATION Pilot Enrollment Ventura’s Pilot enrolled 1,520 beneficiaries by the end of December 2021. The average length of enrollment was 21.8 months. Approximately 58% of enrollees ever disenrolled at some point between January 2017 and December 2021. The most common reason for disenrollment was lack of engagement (27% of total WPC enrollment)*?. Enrollee Identification and Engagement Initially, Ventura primarily used data from managed care plans to identify eligible beneficiaries with high rates of utilization, though referrals from WPC-partners resulted in highest enrollee engagement. On average, one to four outreach attempts were made before an individual enrolled in Ventura’s WPC pilot. Enrollment occurred at health care facilities, shelters, and in field-based outreach. Additional outreach and enrollee engagement was facilitated through backpack medicine providers and “One Stop” events providing homelessness services. 43 Beneficiary refused to participate or did not engage in services. (3) Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report Care Coordination Enrollees were assigned to a care coordination team led by a community health worker. Teams included community health workers providing care coordination, nurses, substance counselors, and social workers available for clinical consult, a health educator, and staff to support enrollment in public benefits. Enrollees were supported by multiple care coordinators within the care coordination team based on availability and expertise who followed the enrollee across all WPC-participating care settings. All enrollees received field-based outreach and care coordination with an average case ratio of 1:100. Those identified with highest utilization rates had a 1:50 case ratio. Care coordination was supported by clinical case reviews and daily multidisciplinary team meetings to promote team-based care and collaborative care delivery. Comprehensive assessment of all identified social needs was conducted using the Vulnerability Index — Service Prioritization Decision Assistance Tool (VI-SPDAT) and a Ventura- developed “WPC Social Needs Survey” adapted from other validated tools. Housing Assistance Ventura emphasized a “Housing First” model and provided medical respite to enrollees experiencing homelessness. WPC funds were used to assist with tenancy support, housing navigation, completing applications for the Coordinated Entry System, and obtaining housing funds (e.g., housing choice vouchers or rental subsidies). Other Services In addition to care coordination and housing services, Ventura also provided benefits assistance, employment assistance, medical respite, health education, transportation, and referrals to legal assistance. CRITICAL SUCCESS FACTORS « Providing all WPC services in house enabled strengthened partnerships and alignment with county safety-net providers to approach delivery of care from a population health perspective. e Supported outreach to and engagement with unhoused enrollees through “One Stop” events that gathered multiple providers (e.g., public benefits assistance, child and family services, and housing navigators) in mobile locations to deliver services. ¢ Universal consent form facilitated data sharing across WPC partner organizations. ¢ Referrals from community-based partners and through field-based outreach allowed patient engagement closer to the point of care and at a time of established need, resulting in a higher referral completion rate. PERCEIVED IMPACT OF WPC Ventura perceived an above average impact on the following aspects of WPC Pilot implementation: ¢ Improved integration of health, behavioral health, and social services ¢ Improved care quality © Decreased overall cost of care e Identifying enrollees receiving services from more than one system (e.g., medical, behavioral health, social services) Whole Person Care Final Evaluation Report | Appendix U: Pilot Specific Case Studies Ventura perceived an above average improvement on the following aspects of care for enrollees: ¢ Access to affordable housing ¢ Comprehensiveness and timeliness of available services e Targeted identification, outreach/engagement, and enrollment Frequency and quality of communication with enrollee e Extent to which care provided is patient-centered “You don't think of a health care agency as being necessarily on the front lines for encampments. That's not a traditional role that the health system has been. We just recently applied and were successful for an encampment outreach ... That's something that four years ago, | don't think our agency ever would have even applied to. | don't know that we could have even made the case to play that role, but because of Whole Person Care, | think we've really been able to shift and take on something like that and be part of the solution. And so having a field-based team who can engage with them where they are and over a period, kind of live with them along those stages of change and get them connected with the services that they need and support them as they engage in those services, is a real value that Whole Person Care has provided.” “| would say some of our community partners. We have partners that helped provided locations and have provided support for our care pods. Those have been really integral to our success. We've been working a lot with law enforcement in a lot of different communities to identify persons of concern and serve them through Whole Person Care... Those are the main ones that have really worked.” 660 | Appendix U: Pilot Specific Case Studies | Whole Person Care Final Evaluation Report UCLA CENTER FOR : HEALTH POLICY RESEARCH fit The views expressed in this report are those of the authors and do not necessarily represent the UCLA Center for Health Policy Research, the Regents of the University of California, or collaborating organizations or funders. Copyright © 2022 by the Regents of the University of California. All Rights Reserved. The UCLA Center for Health Policy Research is affiliated with the UCLA Fielding School of Public Health and the UCLA School of Public Affairs. Phone: 310-794-0909 Fax: 310-794-2686 Email: chpr@ucla.edu www. healthpolicy.ucla.edu UCLA Center for Health Policy Research | 10960 Wilshire Blvd. | Suite 1550 | Los Angeles, CA 90024 | t:310.794.0909 | f: 310.794.2686 | healthpolicy@ucla.edu Rare uss emtre eee