Data support the removal of “first-line” terminology from British Columbia’s (BC’s) 2023 guideline update on the treatment for opioid use disorder (OUD). Clinicians and patients should engage in shared decision-making, according to the researchers.
The net health gains associated with buprenorphine/naloxone (BNX), namely reduced mortality risk, are outweighed by lower treatment retention compared with methadone, according to a decision analytic model analysis.
Though data from multiple systematic reviews and retrospective cohort studies have found lower mortality risk with BNX and better treatment retention with methadone, confounders such as short follow-up periods have tended to limit comparative, empirical evidence. In Canada, both agents are available in office-based and specialized treatment settings and covered under the universal healthcare system.
When it comes to opioid agonist treatment (OAT), “there is limited and conflicting population-level evidence on the net effects of treatment alternatives to inform clinical recommendations, particularly in the fentanyl era,” wrote Benjamin Enns, a health economist at the Centre for Advancing Health Outcomes in Vancouver, and colleagues.
- BNX vs methadone: net 10-year life-years ↓ 1.6% (-1602).
- BNX associated with +221 fatal overdoses; +303 all-cause deaths.
- BNX retention lower: 43.9% person-time vs methadone 58.8%.
- Per-protocol: BNX still worse; -1229 life-years, +170 overdoses.
- Shared decision-making advised; remove ‘first-line’ terminology.
The findings were published on December 26, 2025, in JAMA Network Open.
Seeking Answers
The researchers analyzed linked population-level health administrative data for all patients presenting for OAT in BC between January 1, 2010, and March 17, 2020. The study population included 40,461 participants. The median age of the participants was 33 years, and 66% of the population was male.
The researchers aimed to determine the cumulative difference in incremental life-years and population-level benefits and harms (eg, fatal overdoses and all-cause deaths) between three OUD treatment regimens. The first included all patients initiating or reinitiating any OAT between January 2010 and March 2020 based on BC health data. The second included patients initiating or reinitiating methadone exclusively during that period. The third included patients initiating or reinitiating BNX exclusively during that period.
The study was designed to address confounding by indication at baseline, time-varying confounding resulting from suboptimal dosing, and the effects of take-home dosing. The researchers estimated hazard ratios (HRs) on treatment breaks lasting for 5 or more days for methadone and for 6 or more days for BNX (ie, discontinuations) and on mortality while receiving treatment. They adjusted the data for treatment setting and baseline and treatment characteristics. Overdose risk was modeled to account for the change in fentanyl prevalence and rapid increase in fatal overdoses from 2015 to 2017.
Mortality Risk, Treatment Retention
Exclusively prescribing BNX was associated with a 1.6% decrease in total projected life-years (-1602 incremental life-years) over 10 years compared with methadone. It also was associated with an additional 221 fatal overdoses and 303 all-cause deaths. Net health losses were observed across all simulation exercises.
These findings were only slightly attenuated in the per-protocol analysis, which demonstrated an overall loss of 1229 incremental life-years, 170 additional fatal overdoses, and 231 all-cause deaths with BNX vs methadone. Sole receipt of BNX was associated with decreased time in treatment compared with methadone (43.9% of cumulative person-time vs 58.8%).
The two-way sensitivity analysis on treatment discontinuation vs mortality risk while in treatment showed a 40% reduced HR for mortality (ie, improved mortality outcomes) among patients receiving BNX.
Practice Implications
The 2023 BC guideline update on treatment for OUD maintained that BNX might still be favored in the absence of patient preference or contraindications. The Centre for Addiction and Mental Health and the Canadian Research Initiative in Substance Matters took similar positions the following year.
“The findings validate the changing recommendations and change in clinical practice to put BNX and methadone on equal ground,” Alice Ordean, MD, an addiction medicine consultant at St. Joseph’s Health Centre in Toronto, told Medscape News Canada. Ordean, who was not involved in the study, noted that the data validate how she has practiced for over two decades.
When discussing options with patients, pharmacology matters, especially considering the growing problem of fentanyl and overdose deaths, said Ordean.
“Because of its pharmacological properties, BNX became a medication that we could engage people on more easily, but it wasn’t necessarily the most effective medication, which is why we saw a loss in treatment retention in the study. But if you look at the fentanyl issue, we can go higher in dosage on methadone than in the maximum buprenorphine dose, which may be more beneficial for patients using fentanyl in the new era.”
Prescribing requirements are likewise important. “The hardest selling feature is how these medications are prescribed and what that means for the patient on a day-to-day basis,” said Ordean. An employed patient might not want to go to the pharmacy every day for methadone, so they may opt for BNX because the prescribing guidelines are more flexible.
The patient’s preference is crucial. “While individual patients may have different likelihoods of benefiting from one treatment or another, individuals should ultimately be empowered to choose their treatment via shared decision-making with their clinician,” wrote the authors.