Characterizing opioid agonist therapy engagement and its association with mental disorders pharmacotherapy among people living with HIV and co-occurring opioid use and mental disorders : a methodological and population-based assessment

Background: Opioid agonist therapy (OAT) is the gold-standard treatment for opioid use disorder (OUD) and is associated with reduced mortality and improved HIV-related outcomes among people living with HIV (PLWH). However, its role in managing co-occurring mental disorders remains unclear. While a growing body of research has leveraged administrative health data to examine OAT outcomes, a key methodological gap persists: existing approaches do not adequately capture the dynamic nature of OAT exposure, limiting precise characterization of OAT treatment engagement and its impact on co-occurring mental disorders. Using population-based administrative health data, this dissertation aimed, among PLWH with co-occurring OUD and mental disorders, to: (1) develop a framework for constructing OAT treatment episodes; (2) characterize OAT engagement; and (3) examine its association with mental disorder pharmacotherapy. Methods: We used data from the population-based Comparative Outcomes And Service Utilization Trends (COAST) cohort in British Columbia, including PLWH aged ≥19 years (2012−2020). From this cohort, we identified individuals with co-occurring OUD and one of three mental disorder categories: mood and anxiety disorders (MAD-Only), psychotic disorders (PsyD-Only), or both (MAD-PsyD). OAT episodes were constructed using Allen’s interval algebra, temporal margins (Ɛ), and drug-specific permissible gaps, and episodes were classified as monotherapy, transition, multitherapy, or transition-multitherapy. This framework enabled the characterization of OAT engagement (induction completion and adherence) and the examination of its association with mental disorder pharmacotherapy using mixed-effects negative binomial models. Differences across mental disorder groups were evaluated using Holm–Bonferroni adjustments. Results: Among 17,119 PLWH, 1,486 (8.7%) had co-occurring OUD and a mental disorder, of whom 79.2% received OAT. Across 7,232 episodes, monotherapy predominated (>90%). Induction completion (<30%) and adherence (<50%) were low, with inadequate engagement among PsyD-Only and MAD-PsyD groups (adjusted p<0.01). OAT exposure was strongly associated with psychotropic dispensing (rate ratio=3.11), with greater engagement corresponding to increased dispensing; induction completion and adherence ≥90% were associated with over twofold and sixfold increases, respectively. Conclusions: This dissertation advances measurement of OAT exposure and shows that, although OAT was strongly associated with mental disorder pharmacotherapy, engagement remained suboptimal, particularly among individuals with psychotic disorders, highlighting the need for integrated, tailored, adherence-supportive OAT models.