Hispanic-White Differences in Double Bonuses for Quality of Care in Medicare Advantage

The growth of Medicare Advantage (MA), in which beneficiaries enroll in private health plans, has been remarkably rapid for Hispanic enrollees, increasing from 33% to 48% from 2009 through 2018.¹ However, evidence suggests that Hispanic enrollees may receive lower quality of care than their non-Hispanic White counterparts.² The structural drivers of these disparities are myriad, including immigration policy, health literacy, and the COVID-19 pandemic.³ One underexplored driver may be disparities in health care financing. Since 2012, high-quality MA plans have received financial bonuses under the Quality Bonus Program. In “double-bonus” counties with high MA enrollment and low fee-for-service spending, high-quality plans receive bonuses twice as large as those in non–double-bonus counties. Double bonuses totaled $10.2 billion from 2012 through 2018, yet prior research has shown that they do not improve quality of care and are offered less frequently to Black enrollees than to White enrollees.⁴ We examined whether Hispanic enrollees were less likely than White enrollees to reside in counties eligible for double bonuses and, if so, whether this reduced payments to plans caring for Hispanic enrollees.

First, using 2012-2019 Medicare Beneficiary Summary Files and MA ratebook data, we estimated the likelihood of residing in a double-bonus county among Hispanic and White enrollees via a serial cross-sectional logistic regression model estimated at the enrollee-year level, controlling for RTI race and ethnicity code (91% sensitive and 99% specific for Hispanic ethnicity⁵), age, sex, reason for Medicare entitlement, Medicaid dual eligibility, and year fixed effects (eMethods and eFigure in the Supplement). Given the paucity of data on Hispanic payment disparities in Medicare, we focused on Hispanic and White MA enrollees but used analogous methods from prior work.⁴ Second, we estimated mean payments to plans in double-bonus and non–double-bonus counties via a linear regression model estimated at the plan-year level, weighted by enrollment, and adjusted for plan star rating, county-specific benchmark payments, and year fixed effects. Finally, we calculated Hispanic-White payment disparities by multiplying the Hispanic-White difference in residing in a double-bonus county (step 1) by the difference in payments to plans in double-bonus vs non–double-bonus counties (step 2). The University of Michigan institutional review board exempted our study from review because it used deidentified administrative data. This cross-sectional study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline. Our analyses were performed in June 2021 and used Stata, version 16 (StataCorp LLC).

MA enrollees contributed 113 876 310 enrollee-years, comprising 12.8% Hispanic and 87.2% White enrollees (Table 1). Hispanic enrollees were younger and more likely to be dual eligible for Medicaid and entitled to Medicare owing to disability or end-stage kidney disease.

Hispanic enrollees were 11.8 percentage points less likely than White enrollees to reside in a double-bonus county (17.4% vs 29.2%, respectively; difference, 95% CI, −19.5% to −4.0% percentage points) (Table 2). Mean payments were $320 higher for plans in double-bonus vs non–double bonus counties (eAppendix in the Supplement). Taken together and holding quality performance constant across plans, the double-bonus policy increased mean plan payments by $56 for Hispanic enrollees and by $93 for White enrollees, a disparity of $38 per enrollee per year (Table 2). In aggregate, White MA populations gained $551 million more than Hispanic populations from 2012 through 2019 through double bonuses.

In this national study of the MA double-bonus policy, Hispanic enrollees were substantially less likely than White enrollees to live in counties eligible for double bonuses, resulting in lower payments to plans caring for Hispanic populations. Because changes to MA plan payments are partially passed through to enrollees,⁶ bonus disparities likely reduce health care benefits and raise premiums for Hispanic enrollees. Our analysis is limited by the RTI race and ethnicity variable, which undercounts Hispanic enrollees.⁵ Our study also held quality constant to examine structural payment disparities embedded in double bonuses. Given that the average quality performance is worse for Hispanic than White enrollees in MA, the actual payment disparities for plans serving more Hispanic enrollees is likely even larger.^1,2 As Hispanic enrollment in MA continues to increase, Medicare should eliminate double bonuses and other structural payment disparities in the MA program.

Accepted for Publication: December 27, 2021.

Published: March 4, 2022. doi:10.1001/jamahealthforum.2021.5281

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2022 Markovitz AA et al. JAMA Health Forum.

Corresponding Author: Andrew M. Ryan, PhD, Department of Health Management and Policy, School of Public Health, University of Michigan, 1415 Washington Heights, M3124 SPH II, Ann Arbor, MI 48109 (amryan@umich.edu).

Author Contributions: Drs Ryan and Markovitz had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Markovitz, Ayanian, Ryan.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Markovitz, Montás, Ryan.

Critical revision of the manuscript for important intellectual content: Markovitz, Warrier, Ayanian, Ryan.

Statistical analysis: Markovitz, Ryan.

Administrative, technical, or material support: Warrier.

Conflict of Interest Disclosures: Dr Ayanian reported grants from the National Institute on Aging during the conduct of the study, grants from Michigan Department of Health and Human Services, grants from the National Library of Medicine, personal fees from the JAMA Network, personal fees from the New England Journal of Medicine, personal fees from Harvard University, personal fees from the University of Chicago, personal fees from the University of California San Diego, nonfinancial support from the National Academy of Medicine, nonfinancial support from the National Institutes of Health, and nonfinancial support from AcademyHealth outside the submitted work and is a Physicians Health Plan board member representing the University of Michigan. No other disclosures were reported.

Funding/Support: Dr Markovitz was supported by the Agency for Healthcare Research and Quality (grant 5T32H000053-209). Dr Ayanian was supported by the National Institute on Aging (grant 2P01AG032952-11).

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: Dr Ayanian is Editor in Chief of JAMA Health Forum, but he was not involved in any of the decisions regarding review of the manuscript or its acceptance.