Racial Disparities in Medication Adherence and the Patient-Provider Relationship: Does Racial/Ethnic Concordance Matter?

Objective: To examine the impact of patient-provider racial/ethnic concordance on adherence to a prescribed medication regimen in marginalized populations with a focus on health issues related to hypertension, heart condition/disease, elevated cholesterol, and diabetes. Study Setting and Design: Applying the Andersen-Newman Behavioral Model of Health Service Use, we estimate multivariate linear models to analyze the number of prescriptions filled by patients within a calendar year using publicly available data from the Medical Expenditure Panel Survey (MEPS), a set of large-scale surveys of families and individuals, their medical providers, and employers across the United States. Data Sources and Analytic Sample: Data from MEPS on patient race/ethnicity and provider race/ethnicity were collected from survey years 2007 to 2017 as well as data to control for demographic, socioeconomic, and health factors. Our sample includes 238,355 observations, including 46.1% White respondents, 27.1% Hispanic respondents, 19.3% Black respondents, and 7.5% Asian respondents. There are 52,069 (about 22%) cases of patient-provider concordance. Principal Findings: We find a positive association between adherence to a prescribed medication regime and racial/ethnic patient-provider concordance. Patients identifying as non-White fill their prescriptions approximately three times less often than White patients. Relative to White patients in racial/ethnic concordance with their providers, there is an increase in the number of filled prescriptions for Black patients in racial/ethnic concordance with their providers (coef = 0.715; p = 0.02). For patients with hypertension, being in a racial/ethnic concordant relationship with their providers increases the number of prescription refills (White: coef = 1.884, p < 0.001; Black: coef = 2.360, p < 0.001; Hispanic: 1.925, p < 0.001; Asian: 1.461, p = 0.003). The number of prescription refills also increases for White (coef = 1.665, p < 0.001), Hispanic (coef = 3.469, p < 0.001), and Asian (3.796, p < 0.001) patients with heart condition/disease and in racial/ethnic concordance with their providers. Conclusions: The results provide evidence supporting that patients in racial/ethnic concordant relationships with their providers have a greater predisposition to medication adherence even after controlling for enabling, need, and other predisposing factors, particularly for patients with certain chronic medical diseases. Health policy implications are discussed.