1、Invited CommentaryOpportunities to Address Health Disparities in Performance-Based Accountability and Payment ProgramsCheryl L. Damberg, PhD; Marc N. Elliott, PhDDisparities in health care and health outcomes are a pressing policy problem in the United States. Recent attention has focused on the rol

2、e of social risk factors in disparities in health care quality1 and the ramifications of not accounting for social risk factors in value-based payment and accountability initiatives.2 In this issue of JAMA Health Forum, Meyers and colleagues3 report findings from a study in which they constructed si

3、mulated star ratings for Medicare Advantage health plans (ie, contracts). These simulated star ratings were based on 22 Medicare Advantage quality measures estimated for subgroups of enrollees in each contract, specifically those who were Black, Hispanic, or White individuals, and those with low or

4、high socioeconomic status (SES). The authors found lower simulated star ratings associated with low SES among Black and Hispanic enrollees compared with high SES and White enrollees. As in prior studies, they found variation in performance and disparities by the sociodemographic composition of contr

5、act enrollees. The authors also report estimated contract-level associations in performance across subgroups that likely understate true associations, being attenuated by the limited reliability of some measurements.4The study by Meyers and colleagues3 adds to the growing body of studies describing

6、quality-performance disparities for individuals with greater burdens of social risk factors.2 The authors found evidence of disparities in 1 of the largest quality-incentive programs in the United States, consistent with prior findings reported by Joynt and colleagues5 in 9 federal performance-based

7、 payment systems. The evidence prompts consideration of potential actions to address disparities in the context of performance accountability and value-based payment programs. We propose a 4-part approach to aligning measurement, reporting, and incentive payments with equity goals in such programs.

8、Each component addresses a different facet of this pressing problem, and their complementary effects may offset the limitations that each component might have in isolation.+ Related articleAuthor affiliations and article information are listed at the end of this article.Measure Performance Accuratel

9、y to Reduce Provider Incentives to Avoid Disadvantaged PatientsPerformance measurement and incentive programs can have unintended consequences if better health care providers (ie, physicians, physician groups, hospitals, health systems, skilled nursing facilities) and health plans avoid caring for i

10、ndividuals at greater social risk because the scoring approach does not accurately measure the value a clinician or health care institution or a health plan adds. When measuring performance, it is important to remove bias in provider or health plan scores that does not reflect provider performance b

11、ut is instead associated with patient or beneficiary characteristics outside of the control of providers or health plans. Disparities are composed of within- provider disparities (eg, the extent to which patients of low SES receive worse care than patients of high SES from the same clinician or heal

12、th care institution) and between-provider differences (the extent to which all patients of a clinician or health care institution receive worse care).Improving measurement accuracy requires adjusting for the mean within-provider differences associated with factors appropriate for adjustment, but not

13、 adjusting for between-provider differences. Case-mix adjustment may be implemented in regression models, with fixed effects for reporting units (providers, hospitals, or health plans), producing the scores that providers would receive if they all served the same patients.6 As Nerenz and colleagues7

14、 note, adjusting for within-provider performance Open Access. This is an open access article distributed under the terms of the CC-BY License.differences related to social risk factors gives providers credit for caring for higher-risk patients without masking between-provider differences in performa

15、nce. The Medicare Advantage Star Ratings Program adjusts for within-contract differences via the Categorical Adjustment Index, indirectly approximating the effect of case-mix adjustment for low income (ie, dual eligibility for Medicare and Medicaid or receipt of the low-income subsidy) and disabilit

16、y as the reason for Medicare entitlement.2 The adjustment factor assigned is based on the contracts percentage of beneficiaries with low income and/or disability and is added to or subtracted from a contracts overall star rating. The Categorical Adjustment Index is intended to improve care indirectl

17、y for beneficiaries with low income and/or disability by removing a disincentive to enroll them that high-performing contracts may have.Make Disparities Visible Through Public Reporting of Stratified PerformanceIncreasing the visibility of performance in patient subgroups promotes awareness of where

18、 disparities exist and may motivate providers and health plans to address disparities through reputational effects. One such example has been developed by the Centers for Medicare & Medicaid Services Office of Minority Health, which has stratified reporting of clinical and patient experience perform

19、ance measures by race/ethnicity and by gender for Medicare Advantage contracts.8 To accurately differentiate subgroup performance, these stratified performance reports pool data over 2 years, require a minimum of 100 cases per subgroup, and enforce minimum reliability standards of 0.6 to report resu

20、lts.Specifically Incentivize Providers and Health Plans to Improve Care for Disadvantaged PatientsAs recommended by Meyers and colleagues,3 financially incentivizing providers and health plans to improve care for enrollees with a greater burden of social risk factor through an equity measure in valu

21、e-based payment schemes could provide the impetus for providers and health plans to invest quality improvement resources in enrollees who may be more costly to treat. Such measures could include measures of performance for groups at greater social risk. Agniel and colleagues9 describe the developmen

22、t of a Health Equity Summary Score, including a proof-of concept application to Medicare Advantage plans that could be extended to other settings, social risk factors, and measures. The Health Equity Summary Score focuses on stratified measurement of each plans performance for beneficiaries at great

23、er social risk to highlight and incentivize good performance for social risk factor groups. The summary score directly measures and incentivizes high performance for beneficiaries who are dually eligible for Medicaid, Black, Hispanic, and Asian or Pacific Islander. This score incorporates cross- sec

24、tional performance to recognize good care provided to those with social risk burden, somewhat like the stratified star ratings recommended by Meyers and colleagues,3 and it accounts for within-plan and overall improvement to incentivize low-performing but improving plans.Modify Performance-Based Pay

25、ment Systems to Avoid Redistributing Resources Away From Providers Who Care for Disadvantaged PatientsMost performance-based incentive schemes are structured as zero-sum games, where financial rewards for high performers come at the expense of low performers. As such, they can create unintended effe

26、cts because providers who care for disproportionate numbers of disadvantaged patients tend to perform less well than other providers, even when quality is measured accurately, reflecting persistent between-provider and between-plan disparities. Under some performance- based payment systems, these be

27、tween-provider disparities may redistribute resources away from the providers who most need them to improve care. To keep resources from being transferred awayfrom providers who care for disadvantaged patients, it is necessary to act on the distribution of the incentive payments that are themselves

28、based on unbiased measures of performance, as described in our first recommendation.In a prior study,10 we tested an alternative incentive payment approach that started with a standard incentive payment allocation but then “postadjusted” provider payments using predefined patient or provider charact

29、eristics. The approach held the mean incentive payout constant across subgroups while providing larger incentives for better performers within all subgroups. In 1 application, this approach nearly doubled payments to providers caring for disadvantaged patients and reduced payment differentials acros

30、s providers according to patients income, race/ethnicity, and region.Health equity is a high priority for policy makers and is crucial to the nations health. The study by Meyers and colleagues3 contributes to a body of evidence-based tools to advance this aim. A piecemeal approach might address with

31、in-provider or within-plan differences, but not between- provider or between-plan differences, or such an approach might not address cost obstacles to improving equity. Only by advancing a coordinated, complementary set of strategies, such as those described herein, can we ensure accurate measuremen

32、t that spotlights high performance and areas requiring improvement, along with payment systems that most effectively promote better care for all patients, particularly for those who are disadvantaged.ARTICLE INFORMATIONPublished: June 11, 2021. doi:10.1001/jamahealthforum.2021.1143Open Access: This

33、is an open access article distributed under the terms of the HYPERLINK /pages/cc-by-license-permissions/?utm_campaign=articlePDF%26utm_medium%3DarticlePDFlink%26utm_source%3DarticlePDF%26utm_content%3Djamahealthforum.2021.1143 CC-BY License. 2021 Damberg CL et al. JAMA Health Forum.Corresponding Aut

34、hor: Cheryl L. Damberg, PhD, RAND Corporation, 1776 Main St, Santa Monica, CA 90407 ( HYPERLINK mailto:damberg damberg).Author Affiliations: RAND Corporation, Santa Monica, California.Conflict of Interest Disclosures: Drs Damberg and Elliott reported support from the Centers for Medicare & Medicaid

35、Services and grants from the Agency for Healthcare Research and Quality outside of the submitted work.REFERENCESThe National Academies of Sciences, Engineering, and Medicine. Accounting for Social Risk Factors in Medicare Payment: Criteria, Factors, and Methods. The National Academies Press; 2016.So

36、rbero ME, Paddock SM, Damberg CL, et al. Adjusting Medicare Advantage star ratings for socioeconomic status and disability. HYPERLINK /pubmed/30222924 Am J Manag Care. 2018;24(9):e285-e291.Meyers DJ, Rahman M, Mor V, Wilson IB, Trivedi AN. Association of Medicare Advantage star ratings with racial,

37、ethnic, and socioeconomic disparities in quality of care. JAMA Health Forum. 2021;2(6):e210793. doi:10.1001/ jamahealthforum.2021.0793Bird CE, Elliott MN, Adams JL, et al. How do gender differences in quality of care vary across Medicare Advantage plans? J Gen Intern Med. 2018;33(10):1752-1759. doi:

38、 HYPERLINK /10.1007/s11606-018-4605-5 10.1007/s11606-018-4605-5Joynt KE, De Lew N, Sheingold SH, Conway PH, Goodrich K, Epstein AM. Should Medicare value-based purchasing take social risk into account? N Engl J Med. 2017;376(6):510-513. doi: HYPERLINK /10.1056/NEJMp1616278 10.1056/NEJMp1616278Cefalu M, Elliott MN, Hays RD. Adjustment of pati