Buprenorphine Adherence Trajectories, Persistence, and Healthcare Expenditures Among Commercially Insured Patients With Opioid Use Disorder

OBJECTIVES: Despite the proven clinical benefits, evidence for buprenorphine adherence over the recommended minimum of 180 days for opioid use disorder (OUD) is fragmented, with a majority focusing on publicly insured patients. Our study aimed to bridge this evidence gap by providing national estimates of longitudinal adherence and persistence to buprenorphine among commercially insured patients, identifying their predictors and associated cost consequences.

METHODS: Our retrospective longitudinal cohort study utilized the Merative MarketScan Commercial database to identify patients diagnosed with OUD from 2017-2019 and who initiated buprenorphine treatment. We employed Group-based trajectory models to identify distinct adherence patterns over 180 days. Predictors of adherence trajectories were identified using a multinomial logistic regression model. Time to buprenorphine discontinuation (defined as a gap of >30 days in prescription refill) and associated predictors were assessed using Kaplan-Meier- and multivariable Cox regression models. Generalized linear models were employed to estimate marginal healthcare expenditures associated with adherence trajectories.

RESULTS: We identified three distinct buprenorphine adherence trajectories among our study population of 5,263 patients aged ≥12 years: Declining adherence (29.8%), Steadily increasing adherence (13.1%), and Adherent (57.1%). The overall mean (SD) time to buprenorphine discontinuation was 154.00 (0.77) days. Strong predictors (>10% increase in effect size) of declining adherence included having cancer, respiratory diseases, mental health disorders, and being female. Decreasing buprenorphine average daily dose (ADD) and increasing comorbidity burden were associated with steadily increasing adherence. Predictors of early buprenorphine discontinuation included prior alcohol use disorder and HIV. The non-adherent groups had higher total expenditures (mean, 95% confidence interval) [(Declining adherence vs. adherent: $21,037.81($8,463.79–$33,611.82), Steadily increasing adherence vs. adherent: $37,800.14($16,421.05–59,179.22).

CONCLUSIONS: We found buprenorphine adherence to be associated with reduced health expenditures in patients with OUD. The identified predictors of non-adherence and early discontinuation can guide the development of tailored interventions to improve buprenorphine utilization.