For information about COVID-19, including symptoms and prevention, please read our COVID-19 patient guide. Please also consider supporting Weill Cornell Medicine’s efforts against the pandemic.

Physicians Caring for Higher Proportions of Socially Disadvantaged Patients Score Lower in Medicare’s Value-Based Payment System

Physicians who care for the most socially disadvantaged patients were more likely to receive lower scores in a Centers for Medicare & Medicaid Services (CMS) program that financially rewards high-value care than physicians who care for more affluent populations, according to a study by Weill Cornell Medicine researchers. Understanding the factors driving this disparity is essential, the investigators say, as financial penalties and rewards tied to these scores will increase substantially in the coming years.

Established in 2017, the Merit-based Incentive Payment System (MIPS) is Medicare’s latest program for linking physicians’ reimbursement to the value of care they provide to patients. Eligible physicians receive rewards or penalties based on their performance two years earlier across four domains: quality of care according to performance measures created by stakeholder groups and the CMS; improvement activities; advancing care information, such as using electronic health records and sharing test results and treatment plans with other healthcare providers; and cost of care, which was added in 2018.

“There has been a growing concern that value-based programs for modifying physician reimbursement, such as MIPS, may exacerbate health disparities because they do not adequately adjust scores for patient social risk,” said lead author Dr. Dhruv Khullar, an assistant professor of population health sciences and of medicine at Weill Cornell Medicine. “The program is budget neutral, meaning that penalties collected from physicians with lower MIPS scores are redistributed as bonuses to physicians with higher MIPS scores.”

The study, published Sept. 8 in the Journal of the American Medical Association, was the first to examine the effect of patients’ social risk on physician reimbursement. Dr. Khullar and colleagues analyzed publicly available data for 284,544 physicians who participated in MIPS in 2017. They found that physicians with the highest proportion of patients with dual Medicare and Medicaid eligibility had an average MIPS score of 64.7, 11 points lower than a score of 75.9 for physicians with the lowest proportion of socially disadvantaged patients. Individuals with dual Medicare and Medicaid eligibility experience high rates of chronic illness, social risk factors and long-term care needs. MIPS scores range from zero to 100, with a higher score reflecting better performance.

The discrepancy between patient social risk and MIPS performance held across specialties. Lower scores reflected lower ratings across all three domains of quality, improvement activities and advancing care information that comprised the MIPS score for 2017. Adjustments to physicians’ reimbursement ranged from a 4 percent penalty to a 1.9 percent bonus.

Dr. Khullar cautioned that the study evaluated the performance scores for physicians participating in the MIPS program—not necessarily the quality of care that physicians provide.  Further research is required to determine if the discrepancy in scores is related to differences in the quality of care, or to other factors such as the ability to collect, analyze, and report data. One hypothesis is that safety net practices caring for vulnerable populations provide quality care but have difficulty complying with the administrative burden of the MIPS program.

Adjustments for physician reimbursement in the MIPS program are codified in law under the Medicare Access and CHIP Reauthorization Act. For the 2017 program, physician reimbursements were raised or lowered by 4 percent. By 2022, the adjustment rate will be 9 percent.  

“We hope that our analysis of MIPS generates more questions and research on physician reimbursement,” Dr. Khullar said. “Policymakers who are considering levers to adjust the program from year to year may wish to smooth out some of these differences or restructure the program accordingly.”