Pharmacotherapy and overdose risk following new opioid use disorder diagnosis among Medicaid beneficiaries

Abstract

Background: Timely initiation and continuation of medications for opioid use disorder (MOUD) following a new opioid use disorder (OUD) diagnosis reduces overdose risk. Objectives: To examine patient- and community-level factors associated with timely MOUD receipt and the association of these factors and MOUD modality (buprenorphine, methadone, naltrexone) with subsequent overdose. Methods: This cohort study included Medicaid enrollees in 44 states aged 18–64 with a new OUD diagnosis between April 2016 and December 2019. Multivariable Cox Proportional Hazards models assessed time (in days) to MOUD receipt and medically-treated nonfatal overdose stratified by involvement of heroin/synthetic vs. other (primarily prescription) opioids only. Models adjusted for patient, county, and state-level covariates, and models examining overdose additionally adjusted for time-varying (daily) MOUD receipt. Results: Of 1,172,200 Medicaid enrollees with a new OUD diagnosis, 51.8% were female, 55.8% White, and 42.2% aged 30–44. Most (69.2%) did not receive MOUD within 180 days. MOUD receipt was lower among Black (Hazard Ratio [HR] = 0.68, 95% Confidence Interval [CI] 0.67–0.69) and Hispanic (HR = 0.91, 95%CI = 0.90–0.92) compared to White enrollees. During follow-up, 3.2% of enrollees experienced heroin/synthetic opioid overdose, with reduced risk more strongly associated with methadone (HR = 0.14, 95%CI = 0.12–0.15) and buprenorphine (HR = 0.23, 95%CI = 0.22–0.25) than naltrexone (HR = 0.70, 95%CI 0.64–0.77). Conclusion: Use of methadone and buprenorphine, and to a lesser extent naltrexone, are associated with lower overdose risk among Medicaid beneficiaries newly diagnosed with OUD. State and local policies aimed at increasing timely, equitable, and sustained treatment receipt are needed to address the opioid epidemic and reduce disparities.