Race Concordance in Medicare Part-B Summary Data, with Emphasis on African American and Hispanic Family Physicians

Context: Concordance, or the sense of shared identity between two entities based on an attribute, is increasingly being emphasized as a mechanism to address health disparities. A growing body of literature suggests that race-concordance in patient-provider relationships can positively impact patient experience and health outcomes through myriad mechanisms. We sought to investigate the existence of race-concordance of family physicians (FPs) in Medicare Part-B.

Objective: To assess the extent of race-concordance between FPs and their patient panels in Medicare Part-B, with emphasis on two groups of FPs: African Americans and Hispanics.

Study Design and Analysis: A cross-sectional study using the 2020 Medicare Part-B Physician and Other Practitioners Public Use File (PUF): race-ethnicity of FPs from the 2020 Board Certification and Recertification data were linked by NPI to the race composition of patient panels in the PUF. Analysis of Variance and Generalized Linear Models were used to assess the extent to which differences in race-concordance between FPs were statistically significant. Linear regression was used to estimate the share of beneficiaries with a race-concordant Black or Hispanic FP.

Population Studied: Family physicians who sat for Board Certification, Recertification exams in 2020, and had billed Medicare FFS Part-B in 2020.

Outcome Measures: Average Share of Patient Panel that is African American

Average Share of Patient Panel that is Hispanic

Average Share of Patient Panel that is Dual-Eligible

Social Deprivation Index

Hierarchical Condition Category (HCC) Percentile

Results: There were 4818 FPs linked to the PUF. Across seven race-ethnicities, African American FPs had the highest percentages of African American beneficiaries in their patient panels at 38.3%. Similarly, Hispanic FPs had the highest percentages of Hispanic beneficiaries at 41.5%. These two groups of FPs also had the highest HCC percentiles, and practiced in areas with the highest levels of social deprivation relative to the other FP race-ethnic groups.

Conclusions: Race concordance can be plausibly demonstrated for family physicians of African American and Hispanic heritage, using Part-B PUF data. These two groups of physicians also had patient panels with higher patient complexity, and practice in areas with higher levels of unmet social needs. It is a reminder that a diverse physician workforce is needed to meet healthcare demand of the vulnerable segments of our society.