AbstractBackground and aimsUncertainty about optimal treatment duration for buprenorphine opioid agonist therapy may lead to substantial variation in provider and payer decision‐making regarding treatment course. We aimed to identify distinct trajectories of buprenorphine use and examine outcomes associated with these trajectories to guide health system interventions regarding treatment length.DesignRetrospective cohort study.SettingUS Pennsylvania Medicaid.PatientsA total of 10 945 enrollees aged 18–64 years initiating buprenorphine treatment between 2007 and 2012.MeasurementsGroup‐based trajectory models were used to identify trajectories based on monthly proportion of days covered with buprenorphine in the 12 months post‐treatment initiation. We used separate multivariable Cox proportional hazard models to examine associations between trajectories and time to first all‐cause hospitalization and emergency department (ED) visit within 12 months after the first‐year treatment.FindingsSix trajectories [Bayesian information criterion (BIC) = −86 246.70] were identified: 24.9% discontinued buprenorphine < 3 months, 18.7% discontinued between 3 and 5 months, 12.4% discontinued between 5 and 8 months, 13.3% discontinued > 8 months, 9.5% refilled intermittently and 21.2% refilled persistently for 12 months. Persistent refill trajectories were associated with an 18% lower risk of all‐cause hospitalizations [hazard ratio (HR) = 0.82, 95% confidence interval (CI) = 0.70–0.95] and 14% lower risk of ED visits (HR = 0.86, 95% CI = 0.78–0.95) in the subsequent year, compared with those discontinuing between 3 and 5 months.ConclusionsSix distinct buprenorphine treatment trajectories were identified in this population‐based low‐income Medicaid cohort in Pennsylvania, USA. There appears to be an association between persistent use of buprenorphine for 12 months and lower risk of all‐cause hospitalizations/emergency department visits.