RUPRI Center for Rural Health Policy Analysis Fwww.banko Rural Policy Brief Brief No. 2019-3 APRIL 2019 http://www.public-health.uiowa.edu/rupri/ Primary Care Clinician Participation in the CMS Quality Payment Program A. Clinton MacKinney, MD, MS; Fred Ullrich, BA; Keith J. Mueller, PhD Purpose This policy brief examines primary care clinician participation in the Centers for Medicare & Medicaid Services (CMS) Quality Payment Program (QPP), parsed by clinician type (e.g., physician, physician assistant, advanced practice nurse), clinician specialty (e.g., general internal medicine, family medicine, general pediatrics), clinician practice location, and Advanced Alternative Payment Model (A-APM) or Merit-Based Incentive Payment System (MIPS) program participation. Low participation rates could identify opportunities to extend the QPP to more clinicians serving non-metropolitan populations. This analysis will inform policy makers and other key stakeholders working to improve the QPP. Key Findings • Approximately 10 percent of primary care clinicians participate in an A-APM; less than 30 percent of primary care clinicians participate in MIPS. • Nearly 60 percent of primary care clinicians are exempt from MIPS and do not participate in an A-APM. • Metropolitan primary care clinicians are slightly more likely to participate in an A-APM than non-metropolitan primary care clinicians (11.3 and 7.8 percent, respectively); metropolitan primary care clinicians are slightly less likely to participate in MIPS than non-metropolitan primary care clinicians (29.4 and 31.7 percent, respectively). • Metropolitan and non-metropolitan primary care clinicians are exempt from MIPS at approximately the same rate (59.3 and 60.5 percent, respectively). Introduction On March 26, 2015, Congress passed the Medicare and CHIP Reauthorization Act (MACRA) of 2015 (H.R. 2) with considerable bipartisan support. As part of MACRA, the QPP established a new physician (and other clinical professionals) payment system within the Medicare Physician Fee Schedule, incorporating financial incentives and penalties based on performance. The financial incentives and penalties are applied by two primary methods, or “tracks”. MIPS, the first track, evaluates four clinical practice performance categories—Quality, Promoting Interoperability, Improvement Activities, and Cost.1 Alternately, the A-APM track provides a fixed financial bonus for physician participation in certain financial risk-bearing programs. The goal of the QPP is to “improve health outcomes, promote smarter spending, minimize burden of participation, and provide fairness and transparency in operations.”2 RUPRI Center for Rural Health Policy Analysis, This project was supported by the Federal Office of Rural Health Policy University of Iowa College of Public Health, (FORHP), Health Resources and Services Administration (HRSA), U.S. Department of Health Management and Policy, Department of Health and Human Services (HHS) under cooperative 145 Riverside Dr., Iowa City, IA 52242-2007, agreement/grant #U1C RH20419. The information, conclusions and (319) 384-3830 opinions expressed in this document are those of the authors and no http://www.public-health.uiowa.edu/rupri endorsement by FORHP, HRSA, or HHS is intended or should be inferred. E-mail: cph-rupri-inquiries@uiowa.edu Data and Methods A-APM Data CMS reports a total of 418,744 records from 384,761 distinct clinicians/entities for the A-APM program in Performance Year (PY) 20173 (30,651 clinicians, or 8.0 percent, are listed more than once since they participated in more than one A-APM). Since primary care clinicians participating in the Medicare Shared Savings Program Track 1 are not risk-bearing, and thus otherwise ineligible for A-APM participation, they are removed from the A-APM eligibility data (n = 119,302 or 31.0%). 1 The A-APM data contains 44,767 records (3.4% of the combined A-APM and MIPS records) without a matching record in MIPS data. These clinicians are identified by their APM entity as participants in an A-APM, but they had not submitted a Medicare claim (the basis for determining MIPS eligibility). These records are discarded. MIPS Data CMS reports a total of 1,595,427 records from 1,256,843 distinct clinicians/entities for MIPS program in PY 20173 (242,871 clinicians, or 19.3 percent, are listed more than once since they submit claims to Medicare under more than one Taxpayer Identification Number [TIN]). Records from clinicians with multiple TINs are combined to identify MIPS-eligibility and exemption status for all TINs. Clinicians can participate in MIPS as individuals or as a group. To simplify analysis and interpretation, all clinician counts in this report are calculated at the individual level. Provider/Geography Data The CMS Base Provider Enrollment (BPE) files for July 2017 and March 2018 are used to identify clinician types actively approved to bill Medicare.4 BPE data are merged with National Provider Identifier (NPI) data5 from July 2016 through July 2018 to obtain the ZIP code location of all clinicians. Rural-Urban Commuting Area (RUCA)6,7 codes are matched to the BPE/NPI data by ZIP code to identify metropolitan/non-metropolitan locations. ZIP codes classified as RUCA 4.0 or greater are identified as non-metropolitan. Finally, this provider/geography dataset is merged with the A-APM and MIPS datasets. Primary Care Clinician Data This brief focuses on primary care clinician participation in the QPP. Thus, the analysis is limited to primary care physicians (including family practice, general practice, geriatric medicine, internal medicine, and pediatric medicine), physician assistants, and nurse practitioners. Clinicians with multiple specialties are retained if one of the specialties indicated is included on this list. Records are further restricted to clinicians from the 50 states and DC. Final Dataset After removing duplications, missing records, and non-primary care clinicians, the final QPP dataset contains a total of 419,857 primary care clinicians: 45,197 participate in an A-APM, 124,841 participate in MIPS, and 249,819 are exempt from MIPS (and do not participate in an A-APM) (Table 1). Table 1. Primary Care Clinician Participation in the QPP, Performance Year 2017 A-APM 45,197 10.8% MIPS 124,841 29.7% MIPS Exempt 249,819 59.5% TOTAL 419,857 100.0% Findings Slightly more than 10 percent (10.8%, 45,197) of primary care clinicians participate in an A-APM. Primary care clinicians (and other clinicians) not participating in an A-APM may or may not be eligible for MIPS. For PY 2017, clinicians not participating in an A-APM may be exempt from MIPS due to low patient volumes (< 100 beneficiaries) or low allowed charges (< $30,000). 2 Less than one-third 1 MSSP Track 1 was eliminated per the Shared Savings Program Final Rule released December 2018. 2 The thresholds noted here changed per the Shared Savings Program Final Rule released December 2018. 2 (29.7%, 124,841) of primary care clinicians participate in MIPS. Nearly 60 percent (59.5%, 249,819) of primary care clinicians are exempt from MIPS (based on volume and/or charge threshold) in at least one practice TIN. (Table 2). Table 2. Primary Care Clinician Participation in A-APM, MIPS, and MIPS Exempt, Performance Year 2017 Primary Care Total A-APM MIPS MIPS Exempt Clinician Type Count Pct Count Pct Count Pct Family practice 90,535 11,765 13.0% 35,482 39.2% 43,288 47.8% General practice 4,835 277 5.7% 1,700 35.2% 2,858 59.1% Geriatric medicine 1,907 291 15.3% 877 46.0% 739 38.8% Internal medicine 97,651 12,500 12.8% 44,794 45.9% 40,357 41.3% Nurse practitioner 125,989 11,251 8.9% 23,136 18.4% 91,602 72.7% Pediatric medicine 10,586 906 8.6% 305 2.9% 9,375 88.6% Physician assistant 80,582 7,289 9.1% 16,242 20.2% 57,051 70.8% Multiple specialties* 7,772 918 11.8% 2,305 29.7% 4,549 58.5% TOTAL 419,857 45,197 10.8% 124,841 29.7% 249,819 59.5% *Clinicians in “multiple specialties” include those who indicated that at least one of their specialties included those shown elsewhere in this list. We further report primary care clinician participation in the QPP parsed by metropolitan and non- metropolitan practice location. Of the 419,857 primary care clinicians included in the final dataset, 59,483 (14.2%) practice in non-metropolitan areas. This percentage approximates that of the U.S. population, of which 16.3 percent reside in rural areas defined as RUCA 4.0 and greater.8 Metropolitan primary care clinicians are slightly more likely than non-metropolitan primary care clinicians to participate in A-APMs (11.3% v. 7.8%). And metropolitan primary care clinicians are slightly less likely than non-metropolitan primary care clinicians to participate in MIPS (29.4% v. 31.7%). Overall, there is little difference in MIPS-exempt status between metropolitan and non-metropolitan primary care clinicians (59.3% v. 60.5%). Each primary care clinician type has a higher A-APM participation rate in metropolitan areas, and most provider types (all except geriatric medicine) have a higher MIPS participation rate in non-metropolitan areas (Table 3). Table 3: Metropolitan and Non-Metropolitan Primary Care Clinician Participation in A-APM, MIPS, and MIPS Exempt, Performance Year 2017 Metropolitan Primary Care Total A-APM MIPS MIPS Exempt Clinician Type Count Pct Count Pct Count Pct Family practice 73,031 10,172 13.9% 28,393 38.9% 34,466 47.2% General practice 3,974 250 6.3% 1,359 34.2% 2,365 59.5% Geriatric medicine 1,836 284 15.5% 850 46.3% 702 38.2% Internal medicine 89,387 11,762 13.2% 40,805 45.7% 36,820 41.2% Nurse practitioner 105,405 9,812 9.3% 18,656 17.7% 76,937 73.0% Pediatric medicine 9,956 866 8.7% 262 2.6% 8,828 88.7% Physician assistant 70,045 6,609 9.4% 13,685 19.5% 49,751 71.0% Multiple specialties* 6,740 828 12.3% 1,952 29.0% 3,960 58.7% TOTAL 360,374 40,583 11.3% 105,962 29.4% 213,829 59.3% 3 Non-Metropolitan Primary Care Total A-APM MIPS MIPS Exempt Clinician Type Count Pct Count Pct Count Pct Family practice 17,504 1,593 9.1% 7,089 40.5% 8,822 50.4% General practice 861 27 3.1% 341 39.6% 493 57.3% Geriatric medicine 71 7 9.9% 27 38.0% 37 52.1% Internal medicine 8,264 738 8.9% 3,989 48.3% 3,537 42.8% Nurse practitioner 20,584 1,439 7.0% 4,480 21.8% 14,665 71.2% Pediatric medicine 630 40 6.4% 43 6.8% 547 86.8% Physician assistant 10,537 680 6.5% 2,557 24.3% 7,300 69.3% Multiple specialties* 1,032 90 8.7% 353 34.2% 589 57.1% TOTAL 59,483 4,614 7.8% 18,879 31.7% 35,990 60.5% *Clinicians in “multiple specialties” include those who indicated that at least one of their specialties included those shown elsewhere in this list. Participation in one or more financial risk-bearing programs qualifies a clinician to participate in the A-APM. Primary care clinician participation numbers for the Comprehensive ESRD Care Model and Comprehensive Care for Joint Replacement Model are too small to report by geography. Non- metropolitan primary care clinicians are more likely than metropolitan primary care clinicians to participate in the Comprehensive Primary Care Plus Model and less likely to participate in the Next Generation ACO model. The numbers are relatively small, but metropolitan primary care clinicians are more likely to participate in the Oncology Care Model. Metropolitan primary care clinicians are only slightly more likely to participate in the Medicare Shared Savings ACO Program. (Table 4). Table 4. Primary Care Clinician Participation in A-APM Qualifying Programs, Performance Year 2017 A-APM Program Total Metropolitan Non-Metropolitan Count 1 Pct Count Pct Count Pct Comprehensive ESRD Care2 143 0.3% --- --- Comprehensive Care for Joint Repl.2 5 0.0% --- --- Comprehensive Primary Care Plus 6,673 14.8% 5,601 13.8% 1,072 23.2% Medicare Shared Savings Program 21,332 45.0% 18,329 45.2% 2,003 43.4% Next Generation ACO 17,202 38.1% 15,701 38.7% 1,501 32.5% Oncology Care 1,254 2.8% 1,202 3.0% 52 1.1% 1. A clinician may participate in more than one A-APM program. Thus, the total A-APM counts in Table 2 do not equal the total counts in Table 3. 2. Program participation counts are too small to be reported by geography. Among the 374,660 primary care clinicians not participating in an A-APM, 249,819 (66.7%) are exempt from MIPS by virtue of their low-volume status—either low Medicare patient count (< 100 patient-facing encounters) or low allowed charges (< $30,000). Metropolitan primary care clinicians are slightly more likely to be exempt from MIPS. The significant majority of primary care clinicians exempt from MIPS are exempt due to both low patient volume and low allowed charges. Exempt metropolitan primary care clinicians are slightly more likely to be exempt from MIPS due to low patient volumes and slightly less likely exempt due to low allowed charges (Table 5). Table 5. Primary Care Clinician Participation in MIPS and MIPS Exempt, Performance Year 2017 Overall Metropolitan Non-Metropolitan Count Pct Count Pct Count Pct MIPS Exempt 249,819 66.7% 213,829 66.9% 35,990 65.6% Low patient volume 6,611 2.7% 6,132 2.9% 479 1.3% 4 Low allowed charges 51,552 20.6% 42,374 19.8% 9,178 25.5% Both volume and charges 191,656 76.7% 165,323 77.3% 26,333 73.2% For clinicians not exempt from MIPS, CMS analyzes Medicare Part B claims data to determine whether a clinician should be considered as “special status” under the QPP. These special status designations “will affect the number of total measures, activities or entire categories that an individual clinician or group must report.”9 Special status clinician designations include hospital-based, ambulatory surgical center, small practice, and non-patient facing. Note that the special status designations are not mutually exclusive. For example, a single provider could be both hospital-based and in a small practice. CMS designates approximately the same percentage (~25%) of metropolitan and non- metropolitan primary care clinicians as a hospital-based practice. CMS also designates approximately the same percentage (~32%) of metropolitan and non-metropolitan primary care clinicians as in a small practice. Non-metropolitan primary care clinicians are less likely to be designated as non- patient facing, but that may be in part due to a higher likelihood of multiple special status designations. CMS designates nearly two-thirds of MIPS-participating metropolitan primary care clinicians (65.9%) and nearly three-quarters of MIPS-participating non-metropolitan primary care clinicians (70.9%) with at least one special status. Thus, only one third (33.3%) of MIPS-participating primary care clinicians report the original MIPS performance measures (Table 6). Table 6. MIPS Special Status Designation, Performance Year 2017 Special Status Metropolitan Non-Metropolitan Total MIPS Counts 105,962 18,879 Count Pct Count Pct Hospital-based1 No 73,278 69.2% 12,431 65.9% Yes 26,996 25.5% 4,707 24.9% Mixed (mult. locations) 5,688 5.4% 1,741 9.2% ASC2 No 105,931 99.9% 18,877 99.9% Yes 14 0.0% 1 0.0% Mixed (mult. locations) 17 0.0% 1 0.0% Small practice3 No 66,531 62.8% 11,017 58.4% Yes 33,747 31.8% 6,084 32.2% Mixed (mult. locations) 5,684 5.4% 1,778 9.4% Non-patient facing4 No 89,471 84.4% 14,651 77.6% Yes 6,295 5.9% 1,465 7.8% Mixed (mult. locations) 10,196 9.6% 2,763 14.6% Any special status 5 No 36,148 34.1% 5,494 29.1% Yes 69,814 65.9% 13,385 70.9% 1. Hospital based: The clinician furnishes 75 percent or more of his or her covered professional services in inpatient hospital, on-campus outpatient hospital, or emergency room settings (based on place of service codes) during the applicable determination period. 2. Ambulatory surgery center: The clinician furnishes 75 percent or more of his or her covered professional services in sites of service identified by the Place of Service (POS) code 24 used in the HIPAA standard transaction based on claims for a period prior to the performance period as specified by CMS. 3. Small practice: CMS defines a “small practice” as a practice that has 15 or fewer clinicians [National Provider Identifiers (NPIs)] billing under a single Taxpayer Identification Number (TIN). Small practices must attest to the size of the practice. 4. Non-patient facing: The clinician has 100 or fewer Medicare Part B patient-facing encounters (including Medicare telehealth services) during the non-patient-facing determination period. 5 5. Any special status: The clinician qualifies for at least one of the special status factors at one or more of his or her (multiple) practice locations. Discussion The QPP is a major change to the Medicare program that provides both financial bonuses and penalties based on clinician performance. CMS reports that the goal of the QPP is to “improve health outcomes, promote smarter spending, minimize burden of participation, and provide fairness and transparency in operations.”2 However, to achieve these goals, physicians (and other clinicians) must participate. This analysis of QPP Performance Year 2017 finds that while over 10 percent of primary care clinicians participate in an A-APM, nearly 60 percent of primary care clinicians do not participate in either an A-APM or MIPS. Of the 30 percent of primary care clinicians who participate in MIPS, approximately 70 percent are designated with a special status that requires alternative performance reporting. Thus, only 9.9 percent of primary care clinicians report the original MIPS performance measures (41,642/419,857). With the exception of the Comprehensive Primary Care Plus model, metropolitan primary care clinicians are more likely than non-metropolitan primary care clinicians to participate in an A-APM, suggesting that new, risk-bearing primary care models are less available in non-metropolitan areas. In December 2018, CMS finalized 2019 updates to the QPP. CMS estimates that between 165,000 and 220,000 clinicians will become A-APM participants. Despite raising MIPS exempt thresholds from 100 beneficiaries and $30,000 allowed charges to 200 beneficiaries and $90,000 allowed charges, CMS further estimates that approximately 798,000 clinicians will be MIPS-eligible in PY 2019.10 It is unclear how many of the estimated A-APM-participating and MIPS-eligible clinicians will be primary care clinicians. Primary care clinician participation in the QPP is important because primary care clinicians are fundamental to many health reform models, including establishing patient-centered medical homes and attributing enrollees to accountable care organizations. Furthermore, primary care clinicians are a critical source of health care provision for rural Americans; the U.S. primary care clinician geographic distribution is similar to the U.S. population geographic distribution.8 To realize QPP goals, CMS could consider QPP updates that specifically advance primary care clinician participation. Furthermore, CMS could develop A-APM models and programs that specifically include non-metropolitan primary care clinicians. End Notes 1. Centers for Medicare & Medicaid Services. About MIPS Participation. Accessed October 15, 2018. https://qpp.cms.gov/participation-lookup/about?py=2017. 2. Centers for Medicare & Medicaid Services. Strategic Objective for the Quality Payment Program. Accessed November 6, 2018. https://www.cms.gov/Medicare/Quality-Payment-Program/Resource-Library/Quality-Payment-Program-Key- Objectives.pdf. 3. Raw data on Merit-based Incentive Payment System (MIPS) eligibility and Alternative Payment Models (APM) participation for PY 2017 were obtained from CMS through a special request. 4. Centers for Medicare & Medicaid Services. CMS Base Provider Enrollment. Accessed October 15, 2018. https://data.cms.gov/public-provider-enrollment. 5. Centers for Medicare & Medicaid Services. NPI Files. Accessed July 2016, July 2017, July 2018. http://download.cms.gov/nppes/NPI_Files.html. 6. Center for Rural Health. Temporary Zip RUCA 3.10 File Access Page. Accessed July 2017. https://ruralhealth.und.edu/ruca. 7. United States Department of Agriculture. Economic Research Service. Rural-Urban Commuting Area Codes. Accessed November 8, 2018. https://www.ers.usda.gov/data-products/rural-urban-commuting-area-codes.aspx. 8. RUPRI Center analysis of U.S. Census Bureau, 2012-2016 American Community Survey 5-Year Estimates, RUCA (ZIP), and CMS Base Provider Enrollment (July 2017 and March 2018). 9. Centers for Medicare & Medicaid Services. The Quality Payment Program. Accessed November 6, 2018. https://qpp.cms.gov/. 10. Centers for Medicare & Medicaid Services. CY 2019 Updates to the Quality Payment Program – Year 3 Final Rule Executive Summary. Accessed November 7, 2018. https://www.cms.gov/Medicare/Quality-Payment- Program/Resource-Library/Year-3-Final-Rule-executive-summary.pdf. 6