AHA Scientific Sessions 2026 Abstract · Nov 2, 2026
State Medicaid GLP-1RA Coverage Decisions and Population-Level Severe Obesity: A Quasi-Experimental Analysis
By Marinacci L, Liu M, Wadhera RK
Introduction: Severe obesity is rising among low-income working-age adults in the United States. GLP-1 receptor agonists (GLP-1RAs) are highly effective anti-obesity agents. Medicaid insures ~38 million low-income adults, but historically has not covered GLP-1RAs for obesity treatment. However, starting in 2023, several state Medicaid program decided to cover GLP1-RAs for weight loss indications. Spending and fills increased, but the impact of these policies on population-level obesity measures remains uncertain. Objective: To evaluate the impact of Medicaid GLP-1RA coverage for weight loss on the prevalence of severe obesity among low-income working-age adults. Methods: Using Behavioral Risk Factor Surveillance System data (2020-2024), we conducted a difference-in-differences analysis comparing 7 states that expanded Medicaid coverage of GLP-1RAs for weight loss between 1/1/2023 and 1/1/2024 (intervention) with 23 states that did not (control). We included nonpregnant adults aged 26-64 with household incomes ≤138% of the federal poverty level, the Medicaid eligibility threshold in all study states. The pre-period was 2020-2022, the post-period was 2024; 2023 was excluded. The primary outcome was prevalence of severe obesity (BMI≥40 kg/m ² ); secondary outcomes were obesity prevalence (BMI≥30 kg/m ² ) and mean BMI. We fit survey-weighted linear probability models adjusted for age and sex with state and year fixed effects. Interaction terms between group (intervention or control) and period (pre or post) represented the difference-in-differences estimates. Given potential spillover effects, adults with diabetes were included in the primary analysis and excluded in a secondary analysis. Parallel trends were assessed visually and via pre-period group-by-year interaction analyses. Results: The unweighted sample size was 87,924 adults (mean age 43.8, 55.2% female). Pre-period trends were parallel (Figure) . Following Medicaid GLP1-RA coverage for weight loss, severe obesity declined from 10.6% to 8.5% in the intervention states, while increasing from 9.7% to 10.1% in control states, resulting in a significant differential reduction in severe obesity (adjusted DiD -2.5 pp [95% CI -4.6 , -0.5]). Expanded GLP-1RA coverage for weight loss was also associated with a differential reduction in mean BMI (DiD -0.65 kg/m ² [-1.2, -0.09]), but not overall obesity ( Table ). Results were similar after excluding adults with diabetes (severe obesity DiD -2.1 pp [-4.2, -0.04]). Conclusion: Medicaid GLP-1RA coverage for weight loss led to a reduction in severe obesity and mean BMI among low-income working-age adults, suggesting that Medicaid GLP-1RA coverage decisions may have significant implications for population-level cardiometabolic health. However, the observed gains may be lost as states reverse coverage in response to rising costs.