• Danielle R Fine, Katherine Hart, Natalia Critchley, Yuchiao Chang, Susan Regan, Andrea Joyce, Emily Tixier, Nora Sporn, Jessie Gaeta, Joe Wright, Gina Kruse, and Travis P Baggett.
• Division of General Internal Medicine, Massachusetts General Hospital, 100 Cambridge Street, 16th Floor, Boston, MA, 02114, USA. drfine@mgh.harvard.edu.
• J Gen Intern Med. 2024 Nov 1; 39 (15): 292729342927-2934.

BackgroundThe opioid overdose epidemic disproportionately impacts people experiencing homelessness. Outpatient-based opioid treatment (OBOT) programs have been established in homeless health care settings across the USA, but little is known about the success of these programs in engaging and retaining this highly marginalized patient population in addiction care.ObjectiveTo evaluate predictors of initial engagement and subsequent attendance in a homeless-tailored OBOT program.DesignProspective cohort study with 4 months of follow-up.ParticipantsA total of 148 homeless-experienced adults (≥18 years) who newly enrolled in the Boston Healthcare for the Homeless Program (BHCHP) OBOT program over a 1-year period (1/6/2022-1/5/2023).Main MeasuresThe primary outcomes were (1) initial OBOT program engagement, defined as having ≥2 additional OBOT visits within 1 month of OBOT enrollment, and (2) subsequent OBOT program attendance, measured monthly from months 2 to 4 of follow-up.Key ResultsThe average age was 41.7 years (SD 10.2); 23.6% were female, 35.8% were Hispanic, 12.8% were non-Hispanic Black, and 43.9% were non-Hispanic White. Over one-half (57.4%) were initially engaged. OBOT program attendances during months 2, 3, and 4 were 60.8%, 50.0%, and 41.2%, respectively. One-quarter (24.3%) were initially engaged and then attended the OBOT program every month during the follow-up period. Participants in housing or residential treatment programs (vs. unhoused; adjusted odds ratios (aORs) = 2.52; 95% CI = 1.17-5.44) and those who were already on or initiated a medication for opioid use disorder (OUD) (aOR = 6.53; 95% CI = 1.62-26.25) at the time of OBOT enrollment had higher odds of engagement. Older age (aOR = 1.74 per 10-year increment; 95% CI = 1.28-2.38) and initial engagement (aOR = 3.50; 95% CI = 1.86-6.59) conferred higher odds of attendance.ConclusionsIn this study, over half initially engaged with the OBOT program, with initial engagement emerging as a strong predictor of subsequent OBOT program attendance. Interventions aimed at enhancing initial OBOT program engagement, including those focused on housing and buprenorphine initiation, may improve longer-term outcomes in this marginalized population.© 2024. The Author(s), under exclusive licence to Society of General Internal Medicine.

3 keywords...

hide…