Associations between COVID impact domains and time to buprenorphine treatment disengagement

<i>Background:</i> People with opioid use disorder (OUD) experienced worse outcomes from the COVID-19 pandemic, which disrupted general medical care.<i>Objectives:</i> To quantify COVID-19 impact on OUD treatment disengagement among patients in office-based addiction treatment (OBAT) with buprenorphine.<i>Methods:</i> We recruited 112 outpatients actively on buprenorphine at enrollment from July 2021 to 2022 for telephone surveys within a prospective cohort study. Exposures were six COVID-19 impact domains: personal or family COVID-19 infection, difficulty accessing healthcare/medications, economic stressors, worsening physical or mental health, social isolation, and conflicts and disruptions in the home. The primary outcome was time to OBAT disengagement. We examined associations between each COVID-19 impact domain score and the composite COVID-19 pandemic impact score (a sum of all scores) with time to disengagement using Cox regression models.<i>Results:</i> Average age was 48 years, 56.3% male, 71.2% white/non-Hispanic, 67% Medicaid insurance and 92.9% housed. The majority (74/112) had started in OBAT prior to the pandemic. Median time to disengagement was 271 days. Neither unadjusted nor adjusted analyses showed significant associations between COVID-19 impact domains and time to disengagement. The adjusted hazard ratio for composite COVID-19 impact score was 1.00 (95% CI: 0.99, 1.02).<i>Conclusions:</i> Negative COVID-19 impacts were prevalent among patients in OBAT, but none were significantly associated with time-to-care disengagement. Most patients had multiple protective factors (male, white/non-Hispanic, housed, and Medicaid insurance) and this OBAT rapidly adopted telemedicine, which may have minimized treatment disruption. Preserving patient insurance access and telehealth may maximize OUD treatment retention despite pandemic stressors.