Importance: People experiencing homelessness have been disproportionately affected by the opioid overdose crisis. To mitigate morbidity and mortality, several office-based addiction treatment (OBAT) programs designed for this population have been established across the US, but studies have not yet evaluated their outcomes.

Objective: To evaluate treatment retention and mortality in an OBAT program designed specifically for individuals experiencing homelessness with opioid use disorder (OUD).

Design, Setting, and Participants: A retrospective cohort study was conducted in the Boston Health Care for the Homeless Program (BHCHP). Participants included all adult patients (N = 1467) who had 1 or more OBAT program encounter at BHCHP from January 1 through December 31, 2018. Data analysis was conducted from January 13 to December 14, 2020.

Exposures: Sociodemographic, clinical, and addiction treatment-related characteristics were abstracted from the BHCHP electronic health record.

Main Outcomes and Measures: The primary outcome was all-cause mortality, identified by linkage to the Massachusetts Department of Public Health vital records. Multivariable Cox proportional hazards regression analyses were performed to evaluate baseline and time-varying variables associated with all-cause mortality. Secondary addiction treatment-related outcomes were abstracted from the electronic health record and included (1) BHCHP OBAT program retention, (2) buprenorphine continuation and adherence verified by toxicology testing, and (3) opioid abstinence verified by toxicology testing.

Results: Of 1467 patients in the cohort, 1046 were men (71.3%) and 731 (49.8%) were non-Hispanic White; mean (SD) age was 42.2 (10.6) years. Continuous retention in the OBAT program was 45.2% at 1 month, 21.7% at 6 months, and 11.3% at 12 months. Continuous buprenorphine adherence was 41.5% at 1 month, 17.6% at 6 months, and 10.2% at 12 months, and continuous opioid abstinence was 28.3% at 1 month, 6.1% at 6 months, and 2.9% at 12 months. The all-cause mortality rate was 29.0 deaths per 1000 person-years, with 51.8% dying from drug overdose. Past-month OBAT program attendance was associated with lower mortality risk (adjusted hazard ratio, 0.34; 95% CI, 0.21-0.55).

Conclusions and Relevance: Mortality rates were high in this cohort of addiction treatment-seeking homeless and unstably housed individuals with OUD. Although continuous OBAT program retention was low, past-month attendance in care was associated with reduced mortality risk. Future work should examine interventions to promote increased OBAT attendance to mitigate morbidity and mortality in this vulnerable population.