Instability of Insurance Coverage in Patients on Buprenorphine and Naltrexone for Opioid Use Disorder: Evidence From Three US Health Systems
Recommended Citation
Nguyen A, Binswanger I, Goodrich G, Campbell C, Xu S, Loree A, Glanz J. Instability of Insurance Coverage in Patients on Buprenorphine and Naltrexone for Opioid Use Disorder: Evidence From Three US Health Systems. Drug Alcohol Depend 2025; 267.
Document Type
Conference Proceeding
Publication Date
2-1-2025
Publication Title
Drug Alcohol Depend
Keywords
benzodiazepine derivative, buprenorphine, naltrexone, opiate, Caucasian, cohort analysis, comorbidity, conference abstract, controlled study, drug dependence, drug therapy, female, follow up, Hispanic, human, major clinical study, male, medicaid, medicare, mental disease, retrospective study, treatment duration
Abstract
Drug Category: Opiates/Opioids Topic: Health Services Abstract Detail: Clinical - Epidemiology Abstract Category: Original Research Aim: We examined health plan disenrollment and associated factors in patients on medication for opioid use disorder (MOUD). Methods: In this retrospective cohort study of patients age ≥16 on buprenorphine or naltrexone between 2012-2021 from 3 health systems, we calculated the frequency of health plan disenrollment within 12 and 24 months after the first observed MOUD (index) treatment episode. We fit Poisson models to identify patient characteristics associated with disenrollment rates, defined as the number of disenrollments per person-years of follow-up. Characteristics examined included patient demographics, insurance type, medications (e.g., opioid analgesics and benzodiazepines), substance use disorders, mental health disorders, and other comorbidities. Results: The cohort included 22,938 MOUD patients with a mean (median) age of 38.7 (36), of whom 62% were men, 5% Black, 78% White, 17% other or missing race, and 19% Hispanic of any race. Further, 84% had commercial insurance, 12% had Medicare, 6% had Medicaid, and 6% had other insurance (not mutually exclusive). The mean (median) length of treatment episodes was 506 (130) days. Within 12 and 24 months of the index treatment episode, 6,161 (26.9%) and 9,525 (41.5%) patients, respectively, had ≥1 disenrollments on or off treatment. Additionally, 1,898 (8.3%) and 2,698 (11.5%) patients had ≥1 disenrollments while on treatment within 12 and 24-months from the index episode, respectively. In an adjusted model, the estimated risk of disenrollment was higher for ages 16-25 (IRR=1.80, 95% CI 1.63-1.99) and 26-45 (IRR=1.40, 95% CI 1.40-1.66) compared to ages 46-64. Further, the estimated risk was lower for patients with Medicare compared to those without Medicare (IRR=0.73, 95% CI, 0.62-0.84). Results were similar for disenrollment while on treatment. Conclusions: In this cohort of patients on MOUD, more than 40% disenrolled from health plans within two years of treatment. Findings suggest that disenrollment may contribute to poor treatment retention, especially in younger patients. Financial Support: This study is supported by a grant from the National Drug Abuse Treatment Clinical Trials Network (CTN-0141) of the National Institute on Drug Abuse (3UG1DA040314-08S4).
Volume
267
