Abstract
IMPORTANCE: In 2017, British Columbia guidelines changed the preferred first-line treatment for opioid use disorder from methadone to buprenorphine/naloxone. Systematic reviews have consistently shown better treatment retention associated with methadone, while studies have reported a reduced risk of mortality associated with buprenorphine/naloxone, causing confusion in interpretation for clinical guidelines.
OBJECTIVE: To estimate the population-level health benefits and harms associated with buprenorphine/naloxone vs methadone for treatment of opioid use disorder, accounting for differences in treatment retention and risk of mortality while receiving treatment.
DESIGN, SETTING, AND PARTICIPANTS: This decision analytic model analysis was conducted from January 1, 2010, to March 17, 2020, using a semi-Markov cohort model and linked population-level health administrative data on all individuals presenting for opioid agonist treatment in British Columbia, Canada. Hazard ratios were estimated for the association of buprenorphine/naloxone vs methadone with treatment discontinuation (breaks in dispensations lasting ≥5 days for methadone and ≥6 days for buprenorphine/naloxone) and mortality while receiving treatment (per protocol). Treatment episodes in primary analysis were based on the medication initiated (initiator analysis). Probabilistic (mean estimates and 95% credible intervals from 10 000 simulations) and deterministic sensitivity analyses were conducted. Data were analyzed between August 2023 and October 2024.
EXPOSURE: Alternative treatment policies in which buprenorphine/naloxone or methadone were exclusively available to individuals presenting for opioid agonist treatment.
MAIN OUTCOME AND MEASURES: Incremental life-years, fatal overdoses, and all-cause deaths.
RESULTS: The study population included 40 461 individuals with 109 126 cumulative person-years of follow-up (median [IQR] age, 33 [23-43] years; 66.0% [range, 56.0%-76.0%] male). A policy of exclusively buprenorphine/naloxone was estimated to have -1602 incremental life-years (95% credible interval, -3249 to -549 life-years) compared with methadone, with an additional 221 fatal overdoses (95% credible interval, 119 to 376 overdoses) and 303 all-cause deaths (95% credible interval, 120 to 589 deaths) over a 10-year period.
CONCLUSIONS AND RELEVANCE: Results from this study suggest that any advantages from a reduced risk of mortality with treatment with buprenorphine/naloxone were outweighed by deficits in treatment retention. Evaluated in a jurisdiction where both medications were available in office-based settings, these findings do not support recommendations of buprenorphine/naloxone as first-line treatment over methadone.