SPECIAL ISSUE BRIEF • MARCH 2018 MINNESOTA’S ACCOUNTABLE COMMUNITIES FOR HEALTH: CONTEXT AND CORE COMPONENTS AUTHORS: INTRODUCTION Minnesota’s Accountable Communities for Health, or ACHs, are community- CARRIE AU-YEUNG, MPH led models of delivering medical and non-medical care and services to Research Fellow, SHADAC improve the health of a target population with substantial health and social needs. ACHs bring together diverse community partners, driven CHRISTINA WORRALL, MPP by the specific needs of the target population and the prevailing health Sr. Research Fellow, SHADAC and social conditions in the community.1 This brief provides the context surrounding the development and implementation of Minnesota’s ACHs and describes key components of ACH models implemented across the state. Subsequent briefs will spotlight ACH activities, outcomes, and Visit us at www.shadac.org sustainability, including findings from data collected from participating providers. Funding: The Federal SIM Initiative ACHs in Minnesota were funded through a $45 million State Innovation Model (SIM) cooperative agreement, awarded to the Minnesota Depart- ment of Human Services in 2013 by the Center for Medicare and Medicaid Innovation (The CMS Innovation Center). Administered by the Minneso- ta Departments of Health and Human Services, the funding was used to implement the Minnesota Accountable Health Model Framework. State-Led Evaluation: SHADAC The Innovation Center required a federal multistate evaluation of the SIM initiative as well as individual state evaluations. The Minnesota Depart- ment of Human Services contracted with SHADAC to design and conduct the state evaluation of Minnesota’s SIM initiative. The evaluation was conducted between 2015 and 2017. The results of this evaluation are not endorsed by the federal government. These findings do not reflect the views of and may differ from the federal government’s evaluation. Evaluation Approach & Data Collection SHADAC’s evaluation of SIM in Minnesota relied on both existing and new data sources and incorporated both quantitative and qualitative methods. The evaluation of Minnesota’s ACHs, in particular, relied on (a) initial and final semi-structured qualitative interviews with state staff and with individuals engaged in each of the ACHs; (b) a survey of health care and other care/service providers; (c) a survey of organizations participat- ing in the SIM initiative in Minnesota; and (d) ongoing systematic review of state, grant, and contract materials. Scope of This Brief Minnesota’s ACHs are described in this brief as implemented during the course of the ACH initiative. ACHs that have sustained beyond SIM funding may have changed in structure, scope, and activities. State Health Access Data Assistance Center SPECIAL ISSUE BRIEF • MARCH 2018 BACKGROUND: ACHS IN CONTEXT Figure 1: State Context Accountable Community for Health (ACH) Grantees Minnesota was one of six states that received a State Innova- tions Model (SIM) award in 2013 to fund the implementation and testing of its particular model for payment and delivery system reform through December 2017.2 As part of its model, Minnesota established Accountable Communities for Health (ACHs) wherein health care providers, community and social services organizations, and other partners work across sectors to improve overall health in a given community by addressing the health and social needs of populations facing barriers to health equity. Minnesota’s ACHs have their state roots in an earlier grant program, called the Community Care Team (CCT) Pilot. The CCT program was administered from 2011 to 2012 by the Minne- sota Department of Health’s patient-centered medical home initiative, called the Health Care Home (HCH) program, which focuses on providing patient-centered primary care for individ- uals with chronic/complex health conditions. The CCT initiative was meant to expand beyond the medical focus of HCHs and to that end provided resources to health care providers to improve National Context: ACHs across the Country existing partnerships between local hospitals, primary care Minnesota was not alone in establishing Accountable Com- clinics, public health, behavioral health, social services, and munities for Health as part of its SIM model, with a number of other community services.3 states, including California, Michigan, Oregon, Vermont, and Washington, also designing and implementing ACH models The state awarded grants to fund the creation of 15 ACHs under as part of their larger health system transformation strate- SIM (Figure 1).4 In all, eight ACHs were anchored in urban areas, gies.6 However, ACH-like models pre-date the SIM initiative. In six were located in rural areas, and one had both an urban and addition to Minnesota’s own CCT pilot, ACH-like arrangements rural presence. Three of the ACHs established in Minnesota that pre-dated SIM include, among others: Community Care of under SIM were outgrowths of CCT pilots.5 Minnesota’s ACH North Carolina; Vermont’s Community Care Teams; Maryland’s vision called for the integration and coordination of social Community-Integrated Medical Home model; and the Commu- services and clinical care for a population identified by the nity Health Partnership in Baltimore.7 community (e.g., people living within a particular geographic area, high utilizers of health care resources, individuals with a Concurrent Efforts: specific health condition or disability, specific underserved or The Accountable Health Communities Model marginalized groups, etc.) across a range of providers with lead- ership from community stakeholders. Eight ACHs established Just as ACH-like arrangements pre-dated the SIM initiative, similar arrangements have continued to emerge and evolve outside of SIM. care coordination models that focused broadly on the medical and social needs of their target populations without regard to a Most notably, the Innovation Center launched its Accountable Health Communities (AHC) Model in late 2016. The AHC Model specific category of health condition; four were focused on aims to test whether systematically identifying and addressing the mental and behavioral health (e.g., depression, substance health-related social needs of Medicare and Medicaid beneficiaries abuse, serious and persistent mental illness, etc.) within their through screening, referral, and community navigation services will target populations; and three focused on specific or comorbid impact health care costs and reduce health care utilization. To this end, the CMS Innovation Center is supporting “bridge organiza- chronic medical conditions within their target populations. tions” to act as hubs in their communities, forming and coordinating (Importantly, while CCT pilots had been driven specifically by consortia that will focus on bridging the gap between clinical and health care providers, ACHs and their care coordination efforts community service providers in one of two ways: (a) Assisting high- risk beneficiaries with community service navigation so they can could be (and were) led by a variety of types of medical and access services to address their health-related social needs; or (b) non-medical service providers. encouraging partner alignment to ensure that community services are available and responsive to the needs of beneficiaries. [1] [1] CMS Innovation Center. “Accountable Health Communities Model.” Retrieved from https://innovation.cms.gov/initiatives/ahcm/ State Health Access Data Assistance Center 2 SPECIAL ISSUE BRIEF • MARCH 2018 ACHS IN MINNESOTA Among ACHs, ACO partners included an IHP fiscal agent, specif- The ACH grant program in Minnesota included several core ic clinics/providers participating in an IHP or ACO arrangement, requirements for individual ACHs: the establishment of a collab- a health plan, and a managed care plan. In nearly all (14) ACHs, orative leadership structure that involved community partners, an ACO representative or provider participant served on the the development and implementation of a community-based ACH leadership team. When the ACH target population and the care coordination system or team, and the implementation of ACO attributed populations overlapped, select ACO partners population-based health prevention plan. Other key elements provided data, data analytics, and connections with providers of Minnesota’s ACH model include a sustainability plan, a mea- and have been a source of patients/referrals for the ACHs. surement plan, and participation in a Learning Collaborative Partnership with Local Public Health and evaluation activities. While not a requirement, ACHs were encouraged to engage ACH Leadership and Partners local public health organizations in their efforts. Two-thirds (10) Minnesota required that each ACH establish a collaborative of ACHs did involve a local public health partner in their work; leadership team with representation from a broad range of the extent and nature of this involvement varied across sites, providers and organizations in the community as well as from with local public health serving as the lead agency for two ACHs individual community members and members of the ACH and serving on the leadership team in seven ACHs. ACHs with a target population. The types of organizations involved in a public health partner reported that public health organizations given ACH as leadership team members and/or operation- brought a health promotion focus, hired key coordination staff, al partners varied widely and included, among others: health conducted population health activities, contributed data and systems, health clinics, hospitals, local public health organiza- evaluation expertise, and supported community engagement tions, behavioral health providers, health plans, human service and relationship building. and social service agencies, schools and/or school districts, housing resources, disability service providers, long-term Community and Target Population Involvement care providers, correctional facilities, law enforcement, faith- The requirement to include community members and members based organizations, legal services, and city governments. of the ACH target population on the ACH leadership body was Many of the ACHs built on existing collaboratives and part- meant to ensure that these individuals would have not just an nerships among providers and organizations, but the majority advisory role but a decision-making role in ACH development of ACHs did include new partnering organizations as well. In and implementation. In all, nine ACHs were successful in in- all, ACHs identified 279 organizational partners, including part- cluding at least one community or target population member ners who held played both leadership and operational roles. on their leadership teams. ACHs also encouraged and facilitat- The number of partners involved in any individual ACH ranged ed community participation through other means, such as the from six to 34. care coordination team, or through focus groups, surveys, inter- views, or other activities. In all except one of the ACHs, members ACHs were also required to identify one partner as the lead of the community or target population were involved in the agency for the project. In several cases, medical systems or ACH in some capacity, even if not at the leadership level. Five clinics were chosen as the lead agencies because of a strong ACHs also relied on existing approaches to and structures for connection between ACH goals and HCH activities; in other community engagement among participating organizations, cases, lead agencies were determined based on agency such as community, patient, or consumer councils, or patient resources/project management capacity, general partner con- representation on clinic boards.9,10 sensus about agency fit, or agency interest.8 Innovation Highlight: Partnership with Accountable Care Organizations Community Consultants Although ACHs were not required to establish payment arrangements for ACH activities, they were required to have at One ACH, Together for Health at Myers-Wilkins, implemented a least one active provider or organization partner engaged in particularly innovative approach to community engagement, contracting with individual community members as paid “com- an Accountable Care Organization (ACO) or a similar account- munity consultants” on the ACH leadership team and further able care model based on performance on measures of cost, supporting their involvement by providing transportation and quality, and experience. This partner could be involved in childcare for meetings and events. a Medicaid Integrated Health Partnership (IHP; Minnesota’s Medicaid ACOs), a Medicare ACO (Shared Savings or Pioneer), a commercial ACO, or another ACO or ACO-like arrangement. State Health Access Data Assistance Center 3 SPECIAL ISSUE BRIEF • MARCH 2018 ACH Community-Based Care Coordination System Minnesota afforded applicants, and ultimately 15 awardees, or Team flexibility in terms of how they could implement model require- The goal of care coordination within the ACHs was to address ments, so individual ACHs in Minnesota varied widely across the challenges that individuals, especially those with complex key ACH elements, including care coordination. As a result of conditions, face in getting the care they need—challenges that this variability, ACHs also varied in terms of care coordination are often rooted in the social determinants of health and there- reach, with the average number of individuals reached per fore extend beyond the capacity of the medical realm. Commu- quarter by a given ACH ranging from fewer than 100 to more nity-based care coordination, as its name suggests, integrates than 300 in 2016.12 the delivery of medical and non-medical services by leveraging Because of the flexibility of the ACH program, as a result of resources available in the community to address non-medical which the ACHs were working with a broad range of target pop- health and social needs. Like medical care coordination, com- ulations across a variety of settings, no single care coordination munity-based care coordination involves the management of model accurately captures the various ACH care coordination referrals and the facilitation of care transitions to reduce care approaches, as the table below shows.13 These models can be fragmentation and avoid the risk of duplicative care coordi- conceptualized by looking at the locus/anchor of care coordi- nation efforts; however, community-based care coordination nation and the intensity of care coordination services provided. extends—and may originate—beyond the medical realm in this work.11 Table 1: Overview of ACH Care Coordination Models ACH Name Care Coordination Model and Target Population ACH for People with Implemented the cloud-based LifePlan tool (a comprehensive care plan) among people with intellectual Disabilities and developmental disabilities who live in the Metro area and conducted assessments, created action plans, and provided services using the tool. CentraCare Health Community health workers provided services to the Latino and East African patient populations in Foundation Stearns County at CentraCare Family Health Center in St. Cloud in order to reduce the incidence of un- managed diabetes among this population. Ely CCT CCT partnering organizations provided collaborative and targeted care coordination using a “no wrong door” approach to people living in poverty or with behavioral health challenges in Ely and surround- ing communities. These services included referrals, warm hand-offs, removing barriers to care, and team coordination of care. They sometimes also involved referrals to a CCT care facilitator and/or being part of a care team with a CCT care facilitator. Greater Fergus Falls ACH Ringdahl Ambulance provided community paramedic services to people on Minnesota Healthcare Plans and uninsured low-income residents in Becker, Clay, and Otter Tail counties based on referrals sent by Lake Region Healthcare in order to coordinate health and social services among this population. HCMC Brooklyn Park ACH Depression screening, treatment, and care coordination were provided by a community health worker, behavioral health specialist, and/or family advocate to patients at HCMC Brooklyn Park Clinic. Hennepin County Vocational, housing, and health care services and referrals were provided by employment consultants and Correctional Clients community health workers for individuals incarcerated at the Hennepin County Adult Correctional Facility (ACF) in order to improve health program enrollment, reduce homelessness, increase employ- ment, and reduce recidivism among this group. Mayo CCT A community-based care coordination team developed action plans across primary care, public health, and community services to address the health and social needs of community-dwelling adults with chronic health conditions in the Rochester area. Morrison County Community The controlled substance care team consisting of a social worker, nurse, physician, and pharmacist Based Care Coordination provided services at St. Gabriel’s Hospital and Family Medical Center to seniors and other individuals in Morrison County in order to mitigate the need for, overuse of, and access to prescription narcotics among this population. New Ulm Medical Center Clinic care coordination services and referrals were provided by nurses and social workers for patients at New Ulm Medical Center in order to decrease emergency department visits and inpatient admissions, and improve health outcomes in New Ulm’s Medical Assistance population. * Bold text indicates target population for care coordination. State Health Access Data Assistance Center 4 SPECIAL ISSUE BRIEF • MARCH 2018 Table 1: Overview of ACH Care Coordination Models (cont.) ACH Name Care Coordination Model and Target Population North Country Community Mental health support services and/or referrals were provided by a care coordinator for students at Wake Health Services of the Woods Elementary and Paul Bunyan Elementary in order to improve the region’s capacity to support at-risk youth in crisis. Northwest Metro Healthy Services and referrals were provided by staff for students at Anoka, Andover, Blaine, Champlin Park, Student Partnership Coon Rapids, and Anoka-Hennepin Regional High Schools to address individual student needs as indicated on the 10th grade health survey. Southern Prairie Community Diabetes risk screening and “I Can Prevent Diabetes” program capacity building were provided in order to Care delay and ultimately prevent Type 2 diabetes among those at risk in a 12-county area in southwestern Minnesota. Together for Health at Services and referrals were provided by a community health worker and a public health nurse to students Myers-Wilkins and family members of Myers-Wilkins Elementary School and the surrounding neighborhood of Duluth in order to address the health and wellness needs of this population. Total Care Collaborative Three models—Rapid Access to Case Management; Care Navigation; and Rising Risk Care Conferences— were employed at North Memorial Health Care, Broadway Family Medicine Clinic, and Vail Place to increase person-centered care for individuals with serious mental illness living with chemical dependency issues and co-occurring chronic diseases in North Minneapolis, Robbinsdale, Brooklyn Center, and Brooklyn Park. UCare/Federally Qualified Enhanced care coordination and outreach services were provided in order to strengthen the processes of Health Center Urban Health care for UCare members enrolled in Special Needs Basic Care at four FUHN clinics in the metro area. Network (FUHN) ACH * Bold text indicates target population for care coordination. Source: SHADAC, "Accountable Communities for Health (ACH) Provider Survey." University of Minnesota School of Public Health, June 2017. Locus of Care Coordination connected to both medical and non-medical services through a Minnesota’s ACH community-based care coordination efforts web of organizational connections in the community.14 were anchored in three different contexts: Nine conducted care coordination from the starting point of a medical facility Intensity of Care Coordination or organization (clinic, hospital, health plan); four initiated care ACH care coordination activities can be broadly understood coordination within a community organization (social service within a framework that scales care coordination activities agency, school, group living community); and two used a com- along a continuum of intensity that increases in tandem with bination of medical and community-based starting points. patient needs. In this framework, lower-intensity coordination includes activities such as assessments and referrals. As patient With medically-anchored ACH care coordination, non-medi- needs expand and care coordination intensifies, coordination cal components of coordination involve the identification of activities expand to include elements such as the development community resources for individuals who need assistance with of individualized care plans, patient and family education and social determinants of health. Individuals were then general- patient engagement, the involvement of a collaborative care ly connected outward to community organizations and social team, and ongoing monitoring and follow-up.15 service agencies that could help directly with these issues. When ACH care coordination was anchored within a community orga- The care coordination efforts of about half of the ACHs fall at nization, social determinants of health were addressed within multiple points along this continuum, with services ranging the non-medical community, either by the anchor organization from the provision of information and resources to a refer- itself or by connecting the individual to other non-medical re- ral and/or handoff to an extensive wraparound. Other ACHs sources, which were typically brought to the individual at the tended to concentrate their work at certain points along the anchor care coordination organization. Individuals who had care coordination continuum, depending largely on the needs medical service needs were then connected outward to medical of their target populations. Four ACHs, for example, targeted organizations (clinics or hospitals). In the two cases where ACH particularly high-need individuals such as those with develop- care coordination involved a combination of medical and com- mental disabilities and mental illness, and therefore focused on munity-based starting points, the ACHs used a “no wrong door” high-intensity coordination work. approach to care coordination entry, such that individuals were State Health Access Data Assistance Center 5 SPECIAL ISSUE BRIEF • MARCH 2018 ACH Population-Heath Based Prevention Plans and population health components.18 Moreover, there was The Minnesota ACH Grant Program required ACHs to develop frequent overlap between care coordination and population and implement a population-based prevention plan that health activities, although some ACHs did implement sepa- aligned with its care coordination target population or focused rate activities for each component. Table 2 below summarizes on diabetes management and prevention, tobacco cessa- population health improvement goals, target populations, and tion, hypertension, obesity, or adverse childhood experiences key population health activities pursued by ACHs. As with care (ACEs).16,17 In developing their population health plans, ACHs coordination, ACHs varied in terms of population health reach. were encouraged to build upon prevention work initiated or Where data was provided/available, ACHs reported reaching underway through other community efforts—for example, the anywhere from 201 to 3000 individuals each through popula- Statewide Health Improvement Program (SHIP; leveraged by six tion health programming and activities. ACHs), Community Transformation Grants (leveraged by three ACHs), or other local public health initiatives (leveraged by nine CONCLUSION The overview provided here is meant to paint a broad picture of ACHs). ACHs also leveraged a variety of resources in developing Minnesota’s 15 unique ACH models across their key characteris- their population goals, with Community Health Needs Assess- tics. Subsequent briefs will dive into the variation across specific ments and input from community members being the most ACH components, profile individual ACHs, and present lessons commonly used. In the end, all ACHs focused in whole or in learned across Minnesota’s ACH initiative. part on the same populations for both their care coordination Table 2: Overview of ACH Population Health Activities ACH Name Population Health Goals, Target Population and Activities ACH for People with Provided education around benefits of physical activity for people with disabilities in order to increase Disabilities physical activity among people with developmental and/or intellectual disabilities in the Metro area. CentraCare Health Foundation Raise diabetes prevention awareness and self-management among east African and Hispanic popula- tions in Stearns County by providing education about diabetes prevention and treatment. Ely CCT Increase walk-ability and bike-ability in the Ely community by increasing bike rack availability; designing kiosks to highlight walking and biking opportunities; and offering a community event to promote educa- tion, safety, and awareness around biking and walking. Greater Fergus Falls ACH Address chronic disease prevalence in the Greater Fergus Falls community by coordinating with the State Health Improvement Program (SHIP) to continue ongoing population health activities around healthy behaviors. HCMC Brooklyn Park ACH Move toward the Triple Aim in the Hennepin County community at-large, with a particular focus on patients attending Brooklyn Park Clinic, by improving clinic-community care coordination delivery model for people with depression in the Brooklyn Park community; developing and implementing strat- egies that promoted a community of health (e.g., culture cohorts); and creating sustainable community relationships. Hennepin County Correctional Increase the dietary health and physical activity of individuals incarcerated at the Hennepin County Clients Adult Correctional Facility (ACF) after release by providing education around healthy eating, physical activity, lifestyle changes, and decreased smoking. Mayo CCT Coordinated with Olmstead County’s Community Health Improvement Program (CHIP) workgroups using a “population health management approach” to target CCT activities for community dwelling adults with multiple chronic conditions in the Rochester area that align with Olmstead County’s Community Health Improvement Program (CHIP) financial stress/homelessness and mental health priorities. Morrison County Community Mitigate the need for prescription drugs through pain management, modify patient access to multiple Based Care Coordination narcotic prescriptions, overcome barriers to patient treatment, and coordinate chemical dependence treatment in the Greater Morrison County community by implementing a care coordination model focused on pain management; raising awareness about long-term effects of opioid use and addiction as well as about treatment options; raising awareness about safe disposal of unused prescriptions; and changing the prescribing practices of providers. New Ulm Medical Center Improve health equity and overall health in core measures among New Ulm residence on medical assistance with chronic conditions by promoting healthy eating, physical activity, and tobacco cessation; improving access to healthy food; and improving bike-ability and walk-ability in New Ulm, focusing on “hot spot” neighborhoods. * Bold text indicates target population for population health activities. State Health Access Data Assistance Center 6 SPECIAL ISSUE BRIEF • MARCH 2018 Table 2: Overview of ACH Population Health Activities (cont.) ACH Name Population Health Goals, Target Population and Activities North Country Community Increase awareness of mental health issues and awareness and adoption of positive mental health strate- Health Services gies (and their benefits) among youth and at-risk youth in Clearwater, Hubbard, Beltrami, and Lake of the Woods counties and among adults who work directly with and support these groups (e.g., teachers, school administrators) by collaborating with SHIP on opportunities to improve the mental health of at risk-children and youth; advocating for ACEs awareness and mental health issues in the region where the target population lives; and coordinating care for youth in crisis within school settings. Northwest Metro Healthy Promotee wellness and a school-wide culture of health among students and teachers who attend/work Student Partnership at high schools in the Anoka-Hennepin School District by providing proactive health education and programming around tobacco use, physical activity, healthy eating, and wellness. Southern Prairie Community Delay and ultimately prevent type 2 diabetes among community members at risk for the disease in the Care 12-county SPCC area by implementing I Can Prevent Diabetes curriculum throughout the community and providing free diabetes screenings and education about physical activity and healthy eating. Together for Health at Provide access for students and family members of Myers-Wilkins Elementary School and the sur- Myers-Wilkins rounding neighborhood of Duluth to a greater number of resources that Myers-Wilkins families have identified as key to creating a healthy community including: economic pathways, mental health pathways, healthy lifestyle supports, and expanded community engagement in improving population health. Total Care Collaborative Reduce overall readmissions and ED utilization among individuals with serious mental illness and serious and persistent mental illness by providing improved coordination and transitions of care using a population management approach. UCare/Federally Qualified Enhance linkages between care coordination entities serving the UCare Special Needs Basic Care Health Center Urban Health enrollees at four FUHN clinics in the metro area using a population management approach to care Network (FUHN) ACH coordination. * Bold text indicates target population for population health activities. Source: ACH self-reported annual reports (2016), quarterly progress reports (2017) and interviews with ACH participants. State Health Access Data Assistance Center 7 SPECIAL ISSUE BRIEF • MARCH 2018 ACKNOWLEDGEMENTS SHADAC would like to acknowledge the many contributions made to the evaluation by staff at the Minnesota Department of Human Services and the Minnesota Department of Health, the state agencies charged with implementation of the Minnesota Accountable Health Model Framework. We would also like to thank the individuals from across the state who shared their time and insights related to their participation in Model programs and activities. Finally, the authors would like to acknowledge Lindsey Lanigan at SHADAC for her assistance with report design, layout, preparation, and exhibit production. ENDNOTES 1 Awsumb, B., Dobbe, C., Ly-Xiong, S., Rodriguez-Hager, R., Rydrych, D., & Terrill, W. (October 2016). “Accountable Communities for Health: Perspec- tives on Grant Projects and Future Considerations.” Retrieved from http://www.dhs.state.mn.us/main/groups/sim/documents/pub/dhs-290682. pdf 2 Centers for Medicare & Medicaid Services. “State Innovation Models Initiative: General Information.” Innovation Center. Retrieved from https://innovation.cms.gov/initiatives/state-innovations/ 3 Minnesota Accountable Health Model. (nd). “Community Care Teams.” Retrieved from http://dhs.state.mn.us/main/idcplg?IdcService=GET_DY- NAMIC_CONVERSION&RevisionSelectionMethod=LatestReleased&dDocName=SIM_Community_Care_Teams 4 Awsumb, B., Dobbe, C., Ly-Xiong, S., Rodriguez-Hager, R., Rydrych, D., & Terrill, W. (October 2016). “Accountable Communities for Health: Perspec- tives on Grant Projects and Future Considerations.” Retrieved from http://www.dhs.state.mn.us/main/groups/sim/documents/pub/dhs-290682. pdf 5 Ibid. 6 Minnesota, Oregon, and Vermont received Round One (2013) SIM Model Test Awards; Michigan and Washington received SIM Round One Mod- el Design Awards and SIM Round Two (2014) Model Test Awards; California received a SIM Round Two Model Design Award but did not receive SIM testing funds. Source: Centers for Medicare & Medicaid Services. “State Innovation Models Initiative: General Information.” Innovation Center. Retrieved from https://innovation.cms.gov/initiatives/state-innovations/ 7 Minnesota Accountable Health Model. (2014). “SIM ACH Resources / Literature Review.” Retrieved from http://www.dhs.state.mn.us/main/id- cplg?IdcService=GET_FILE&RevisionSelectionMethod=LatestReleased&Rendition=Primary&allowInterrupt=1&dDocName=DHS16_189325 8 SHADAC. (May 6, 2016). “Evaluation of the Minnesota Accountable Health Model: First Annual Report – Full Report.” University of Minnesota, School of Public Health. Retrieved from http://www.dhs.state.mn.us/main/idcplg?IdcService=GET_FILE&RevisionSelectionMethod=LatestRe- leased&Rendition=Primary&allowInterrupt=1&noSaveAs=1&dDocName=dhs-287574 9 Ibid. 10 SHADAC. (September 2017). “Evaluation of the Minnesota Accountable Health Model: Final Report – Full Report.” University of Minnesota, School of Public Health. Retrieved from https://www.dhs.state.mn.us/main/idcplg?IdcService=GET_FILE&RevisionSelectionMethod=LatestRe- leased&Rendition=Primary&allowInterrupt=1&dDocName=DHS-297901 11 SHADAC. (May 6, 2016). “Evaluation of the Minnesota Accountable Health Model: First Annual Report – Full Report.” University of Minnesota, School of Public Health. Retrieved from http://www.dhs.state.mn.us/main/idcplg?IdcService=GET_FILE&RevisionSelectionMethod=LatestRe- leased&Rendition=Primary&allowInterrupt=1&noSaveAs=1&dDocName=dhs-287574 It is important to note that quarterly data provided by ACHs may be cumulative or not depending on the ACH and quarterly counts, upon 12 which averages are based, and may include outreach and service provision. 13 SHADAC. (May 6, 2016). “Evaluation of the Minnesota Accountable Health Model: First Annual Report – Full Report.” University of Minnesota, School of Public Health. Retrieved from http://www.dhs.state.mn.us/main/idcplg?IdcService=GET_FILE&RevisionSelectionMethod=LatestRe- leased&Rendition=Primary&allowInterrupt=1&noSaveAs=1&dDocName=dhs-287574 14 Ibid. 15 McDonald, KM, Sundaram, V, Bravata, DM, Lewis, R, Lin, N, Kraft, S., et al. “Closing the Quality Gap: A Critical Analysis of Quality Improvement Strategies (Vol. 7: Care Coordination).” Rockville, MD: Agency for Healthcare Research and Quality (2007). “Request for Proposals: Minnesota Accountable Health Model Accountable Communities for Health Grant Program,” Minnesota Department of 16 Health, September 2, 2014, Retrieved from http://www.dhs.state.mn.us/main/groups/sim/documents/pub/dhs16_184306.pdf. 17 SHADAC. (September 2017). “Evaluation of the Minnesota Accountable Health Model: Final Report – Full Report.” University of Minnesota, School of Public Health. Retrieved from https://www.dhs.state.mn.us/main/idcplg?IdcService=GET_FILE&RevisionSelectionMethod=LatestRe- leased&Rendition=Primary&allowInterrupt=1&dDocName=DHS-297901 18 Ibid. State Health Access Data Assistance Center 8