Factors Associated With Initial Treatment Choice, Engagement, and Discontinuation for Patients With Opioid Use Disorder.

Abstract

Pharmacotherapy for opioid use disorder is effective but underused from a clinical perspective, and average treatment duration is shorter than current recommendations. In this analysis, the authors examined factors associated with initiation of, engagement in, and duration of treatment among patients with opioid use disorder. Using the OptumLabs Data Warehouse (a large, national, deidentified database of commercial or Medicare Advantage plan enrollees), the authors identified a sample of 204,225 patients with opioid use disorder between July 1, 2010, and April 1, 2019. Factors associated with initial treatment type were identified with multinomial logistic regression. The odds of treatment engagement, defined as two claims for treatment and a treatment episode of ≥30 days, were estimated with logistic regression. The hazard ratios for treatment discontinuation were estimated with a Cox proportional hazards model. Treatment initiation with pharmacotherapy (alone or in combination with psychosocial therapy) was associated with higher odds of treatment engagement and a lower hazard of treatment discontinuation. Patients with certain behavioral health conditions (e.g., anxiety or mood disorders) had higher odds of initiating treatment with pharmacotherapy and engaging in treatment and a lower hazard of discontinuing treatment. Patients with certain painful general health conditions (e.g., fibromyalgia or musculoskeletal disorders) had lower odds of initiating and engaging in treatment. Treatment initiation with pharmacotherapy was associated with treatment engagement and duration. Previous contact with behavioral health treatment may support initiating, engaging in, and remaining in treatment. Patients with painful conditions may benefit from provider support in initiating treatment for opioid use disorder.