Assessing inequities in buprenorphine treatment across the care cascade

Describe buprenorphine treatment gaps across the care cascade by race, ethnicity, age, sex and key clinical characteristics.

Observational cohort study of new opioid use disorder (OUD) treatment episodes, 1/1/15-12/31/21 PARTICIPANTS: Individuals 16-89 years at Boston Medical Center and affiliated Boston-area clinics with OUD or buprenorphine prescription following 90-day washout MAIN OUTCOMES AND MEASURES: Buprenorphine initiation (by 14d), engagement (second prescription by 34d), and retention (continuous buprenorphine through 180d without >14d gap). Covariates: sex, race and ethnicity, age, past 12-month buprenorphine, past 12-month overdose, co-occurring substance use disorder (SUD) and psychiatric diagnosis.

24,165 episodes (14,182 individuals) included the following characteristics: female (34.1 %); Black (21.5 %), Latino (16.2 %), White (59.2 %); 16-25 years (4.1 %), 65-89 years (4.9 %). Initiation, engagement and retention were low: 39.3 %, 19.3 %, and 12.4 %. In adjusted models, compared with males, females' lower probability of overall retention reflected lower engagement conditional on initiation (41 % vs. 53 %; OR:0.61, 95 %CI:0.56-0.67, p < 0.001). Black individuals' lower probability of retention overall reflected lower initiation than White individuals (28 % vs 44 %; OR:0.50, 95 %CI:0.46-0.55, p < 0.001). The retention disadvantage for younger and older groups reflected cascade-wide gaps. Lower overall retention associated with co-occurring SUDs reflected lower initiation (OUD only:46 %; OUD plus: alcohol:19 % OR:0.27, 95 %CI:0.21-0.34, p < 0.001; sedatives:21 %, OR:0.31, 95 %CI:0.20-0.49, p < 0.001; stimulants:25 %, OR:0.40, 95 %CI:0.33-0.48, p < 0.001; ≥3 SUDs:25 %, OR:0.40, 95 %CI:0.37-0.43, p < 0.001).

Inequities in buprenorphine use emerged across the care cascade with unique patterns by sociodemographic and clinical subgroup. Health systems aiming to reduce buprenorphine inequities should identify the optimal cascade step based on the population of interest.