Rosemarie Martin, Ryan Carpenter, Ariel Hoadley, Amelia Bailey, Augustine Kang, Josiah Rich, Damaris Rohsenow, Linda Hurley, Jennifer Clarke
This study presents the first person-level, statewide evaluation of a comprehensive MOUD program within a U.S. correctional system, revealing a significant and sustained reduction in post-release overdose deaths over a five-year period. Using robust quasi-experimental methods, we demonstrate durable, population-level benefits extending beyond individuals who directly received MOUD. The divergence in overdose trends between justice-involved individuals and the general population, despite parallel fentanyl exposure, suggests that comprehensive MOUD implementation is associated with protective, system-wide effects (e.g., improved staff competencies, stigma reduction) that reshape the post-incarceration risk environment. These protective effects may benefit incarcerated individuals who both do and do not receive MOUD.
BACKGROUND: The criminal legal system represents a critical intervention point for reducing opioid-related overdose deaths. This study describes the population-level impact of providing all three FDA-approved medications for opioid use disorder (MOUD) during incarceration in a statewide carceral system.
METHODS: Using a retrospective cohort design and controlled interrupted time-series analysis, we compared statewide overdose death rates among individuals with and without recent incarceration across three time periods: 1) pre-implementation of the comprehensive MOUD program (2014-2016), 2) the implementation phase, and 3) post-implementation (2016-2022).
RESULTS: Across the observation period, there were 373 overdose deaths among individuals with incarceration history and 2260 among those without. During the pre-implementation period, the monthly overdose rate increased in the incarceration-history group (b = 0.18, 95% CI: [0.05, 0.31], p = .008). After program implementation, the overdose rate in this group decreased significantly (b = -4.57, 95% CI: [-7.36, -1.78], p = .002), but not among those without incarceration history. The post-intervention slope in the incarceration-history group remained negative but non-significant (b = -0.11, 95% CI: [-0.24, 0.03], p = .130).
CONCLUSIONS: This study presents the first person-level, statewide evaluation of a comprehensive MOUD program within a U.S. correctional system, revealing a significant and sustained reduction in post-release overdose deaths over a five-year period. Using robust quasi-experimental methods, we demonstrate durable, population-level benefits extending beyond individuals who directly received MOUD. The divergence in overdose trends between justice-involved individuals and the general population, despite parallel fentanyl exposure, suggests that comprehensive MOUD implementation is associated with protective, system-wide effects (e.g., improved staff competencies, stigma reduction) that reshape the post-incarceration risk environment. These protective effects may benefit incarcerated individuals who both do and do not receive MOUD.