Opioid agonist therapy during residential treatment of opioid use disorder: Cohort study on access and outcomes.

Objective To determine access to opioid agonist therapy (OAT) for those entering residential treatment for opioid use disorder; to report on treatment outcomes for those taking OAT and those not taking OAT; and to determine the association between OAT use and residential treatment completion. Design Retrospective cohort study. Setting Ontario. Participants Patients with opioid use disorder admitted to publicly funded residential treatment programs in the province of Ontario between January 1, 2013, and December 31, 2016. Main outcome measures Access to OAT during residential treatment using descriptive statistics. Treatment outcomes (ie, completed the program, voluntarily left early, involuntary discharged, and other) for the entire cohort and for the OAT and non-OAT groups using descriptive statistics. Association between OAT use at admission and treatment completion (a binary outcome) using bivariate and multivariate models. Results Among an identified cohort of 1910 patients with opioid use disorder, 52.8% entered programs that permitted access to OAT. Overall, 56.8% of patients completed treatment, 23.3% voluntarily left early (eg, were no-shows, dropped out), 17.0% were involuntarily discharged, and 2.9% were discharged early for other reasons. Those taking OAT were as likely to complete treatment as those not taking OAT (53.9% vs 57.5%, respectively; adjusted odds ratio of 1.07, 95% CI 0.77 to 1.38). Conclusion This study demonstrates 2 large gaps in care for patients with opioid use disorder. First, these patients have poor access to OAT-the first-line treatment of opioid use disorder-while in publicly funded residential treatment programs; and second, many are involuntarily discharged from treatment. Additionally, this study indicates that patients taking OAT have similar likelihood of completing residential treatment as those not taking OAT do. Limitations of this study are that it is based on observational data for patients who self-selected before admission to use OAT or not, and it is likely not all confounders were accounted for.