Methadone vs buprenorphine-naloxone in opioid overdose survivors

Importance — Individuals who experience opioid overdoses are at high risk of death, and data are limited about the comparative effectiveness of medications for this high-risk population.

Objective — To compare the effectiveness of methadone and buprenorphine-naloxone in individuals with a past-year history of opioid overdose.

Design, setting, and participants — This retrospective cohort study with target trial emulation included individuals who started methadone or buprenorphine-naloxone treatment between January 1, 2017, and December 31, 2023, in Ontario, Canada. Participants had a past-year emergency department visit for an opioid overdose and no use of either medication for 7 days. Individuals starting each medication were matched on a 1:1 basis using propensity scores, sex, and index date. Data were analyzed from November 2024 to June 2026.

Main outcomes and measures — The primary outcome was death from any cause within 1 year, analyzed using an initiator approach with Cox proportional hazards regression. The secondary outcomes were opioid overdose and treatment discontinuation. Per-protocol analyses evaluating outcomes before medication discontinuation were also conducted.

Results — A total of 5882 individuals were included in the matched cohort (mean [SD] age, 35.8 [10.8] years; 1888 female [32.1%]). In the methadone-initiating group, 165 individuals (5.6%) died within 1 year, compared with 210 (7.1%) in the buprenorphine-naloxone–initiating group (HR, 0.78; 95% CI, 0.63-0.95; P = .01), with an E-value of 1.9. Median time to treatment discontinuation was 25 days (IQR, 7-131 days) in the methadone group compared with 16 days (IQR, 5-69 days) in the buprenorphine-naloxone group (HR, 0.78; 95% CI, 0.74-0.82; P < .001). There was no significant difference in the time to first opioid overdose (HR, 1.07; 95% CI, 0.97-1.18; P = .17). In the per-protocol analyses, there was no significant difference in the hazard of death among individuals starting methadone and buprenorphine-naloxone (HR, 0.84; 95% CI, 0.48-1.47; P = .55) and an increased hazard of opioid overdose among individuals starting methadone (HR, 1.52; 95% CI, 1.19-1.93; P < .001).

Conclusions and relevance — In this matched cohort of individuals with a past-year opioid overdose, starting methadone was associated with a reduced risk of death during the subsequent year compared with starting buprenorphine-naloxone. Treatment durations were short, particularly among individuals starting buprenorphine-naloxone, and there was no significant difference in mortality during receipt of opioid agonist treatment. These findings may be explained by unmeasured confounding, and epidemiologic studies in other jurisdictions are needed to confirm these findings.