Opioid agonist treatment and risk of mortality in French primary care: a nationwide, retrospective cohort study (2026)

Background: There remains little evidence on whether opioid agonist treatment (OAT) is associated with a reduction in all-cause mortality in the context of widespread buprenorphine use. We aimed to study the risk of mortality in French primary care patients using OAT according to time on or off OAT treatment.
Methods: In this population-based, retrospective cohort study, we used data from the French National Health Data System (Système National de Données de Santé; SNDS). We included patients aged 15 years and older who began OAT between Jan 1, 2010, and Dec 31, 2022, and had at least two consecutively dispensed opioid agonists. Covariate data were also retrieved from the SNDS and included age, sex, deprivation index, medical comorbidities, psychiatric disorders, and concurrent medication. The primary endpoint was all-cause mortality by 1 year; all-cause mortality by 2, 5, and 7 years and specific mortality were the secondary endpoints. Multivariate Cox models accounting for time-dependent exposure (on OAT vs off OAT) were used to assess the association between OAT and mortality.
Findings: We included 175 191 individuals using OAT, of whom 131 444 (75.0%) were male and 43 747 (25.0%) were female. The median follow-up period was 3320 days (IQR 2055-4328). At inclusion, most patients had received buprenorphine (114?247 [65.2%]) or buprenorphine-naloxone (5895 [3.4%]). OAT use was associated with a lower risk of all-cause death at 1 year (adjusted hazard ratio [HR] 0.41 [95% CI 0.37-0.44]; absolute risk difference per 1000 person-years: 21.45 [19.32-23.58]), and this association persisted at 2 years (adjusted HR 0.36 [95% CI 0.34-0.39]), 5 years (0.36 [0.34-0.38]), and 7 years (0.40 [0.37-0.42]). OAT was also associated with lower cause-specific mortality, including injury or poisoning, drug-related deaths, accidental overdoses, infectious causes, and suicide, with similar patterns over time. The association was especially marked for buprenorphine, which had a 6-fold lower risk of all-cause death at 1 year (adjusted HR 0.16 [0.14-0.18]).
Interpretation: In this large, nationwide cohort of incident participants using OAT, we identified a large decrease in the risk of all-cause death associated with treatment. The association was especially marked for buprenorphine, even if this might result from indication bias.
Funding: EPI-PHARE Scientific Interest Group in Epidemiology of Health Products from the French National Agency for the Safety of Medicines and Health Products and the French National Health Insurance. [Author's abstract]