Medicaid Expansion and Moud Use and Opioid Admissions In Residential Treatment: A National Difference-In-Differences Study

Background and Purpose: In recent years, opioid overdose deaths have remained at historically high levels, contributing to a substantial economic and societal burden. Medications for opioid use disorder (MOUD) are effective in reducing overdose risk and supporting recovery; however, most individuals with opioid use disorder (OUD) do not receive medication treatment. Compared to outpatient settings, residential treatment programs have been slower to adopt MOUD, due in part to differences in treatment philosophy and insurance-related barriers. Medicaid Expansion, implemented under the Affordable Care Act in 2014, represents a potential mechanism to increase access to MOUD in residential treatment. However, limited research has examined its impact in these settings. This study uses a national dataset to examine the effect of Medicaid Expansion on MOUD use and opioid-related treatment admissions in short- and long-term residential treatment.

Methods: We employed a staggered-adoption difference-in-differences approach to estimate three linear probability models using Treatment Episode Data Set–Admissions data from 2006 to 2023. In Model 1, the dependent variable indicated whether a residential treatment admission involved receipt of MOUD. In Models 2 and 3, the dependent variable indicated whether the admission involved OUD, focusing on short-term (Model 2) and long-term (Model 3) residential treatment, respectively. Each model compared states that expanded Medicaid to those that did not, using a pre-post framework with the year prior to expansion as the reference. Parallel trends were assessed using event study analyses, supporting claims of causality from our findings. All models included state and year fixed effects and adjusted for sex, age, race, ethnicity, prior treatment experience, employment status, co-occurring mental illness, and health insurance type. Model 1 additionally adjusted for residential treatment type and OUD-related mortality, while Models 2 and 3 adjusted for overdose mortality. Missing data were addressed using multiple imputation with chained equations.

Results: Model 1 included 1,802,378 treatment episodes; Model 2 included 3,172,856 episodes; and Model 3 included 2,429,375 episodes. In Model 1, 13.82% of episodes involved receipt of MOUD. In Models 2 and 3, 35.16% and 33.41% of admissions involved OUD, respectively. Medicaid Expansion resulted in a statistically significant increase in MOUD receipt (marginal effect = 0.127, SE = 0.023, p < .001) relative to non-expansion states in the year prior to expansion. However, Medicaid Expansion resulted in no changes in OUD-related admissions in short-term (p = .331) or long-term (p = .418) residential treatment.

Conclusions and Implications: These findings provide evidence that Medicaid Expansion improves access to MOUD within residential treatment settings. Results highlight the importance of addressing insurance-related barriers to expand access to evidence-based treatment for OUD. Future research should examine whether these gains are equitably distributed among populations historically underrepresented in the treatment system.