Disparities in Use of Novel Diabetes Medications by Insurance: A Nationally Representative Cohort Study

Background: Minority racial and ethnic populations have the highest prevalence of type 2 diabetes mellitus but lower use of sodium-glucose co-transporter-2 inhibitors (SGLT2i) and glucagon-like peptide-1 receptor agonists (GLP1ra), novel medications that reduce morbidity and mortality. Observed disparities may be due to differences in insurance coverage, which have variable cost-sharing, prior authorization, and formulary restrictions that influence medication access. Objective: To assess whether racial/ethnic differences in SGLT2i and GLP1ra use differ by payer. Design: Cross-sectional analysis of 2018 and 2019 Medical Expenditure Panel Survey data. Participants: Adults >= 18 years old with diabetes. Main Measures: We defined insurance as private, Medicare, or Medicaid using >= 7 months of coverage in the calendar year. We defined race/ethnicity as White (non-Hispanic) vs non-White (including Hispanic). The primary outcome was use of >= 1 SGLT2i or GLP1ra medication. We used multivariable logistic regression to assess the interaction between payer and race/ethnicity adjusted for cardiovascular, socioeconomic, and healthcare access factors. Key Results: We included 4997 adults, representing 24.8 million US adults annually with diabetes (mean age 63.6 years, 48.8% female, 38.8% non-White; 33.5% private insurance, 56.8% Medicare, 9.8% Medicaid). In our fully adjusted model, White individuals with private insurance had significantly more medication use versus non-White individuals (16.1% vs 8.3%, p < 0.001), which was similar for Medicare beneficiaries but more attenuated (14.7% vs 11.0%, p = 0.04). Medication rates were similar among Medicaid beneficiaries (10.0% vs 9.0%, p = 0.74). Conclusions: Racial/ethnic disparities in novel diabetes medications were the largest among those with private insurance. There was no disparity among Medicaid enrollees, but overall prescription rates were the lowest. Given that disparities vary considerably by payer, differences in insurance coverage may account for the observed disparities in SGLT2i and GLP1ra use. Future studies are needed to assess racial/ethnic differences in novel diabetes use by insurance formulary restrictions and out-of-pocket cost-sharing.