Quantifying the impact of a large-scale opioid agonist treatment program on suicide prevention in New South Wales, Australia: A data-modeling study.

Aims: This study aimed to quantify the population-level impact of a large-scale opioid agonist treatment (OAT) program on suicide-related mortality among people with opioid use disorder (OUD) in New South Wales (NSW), Australia.

Design: This is the first study to use dynamic mathematical modeling to explore the population-level impact of OAT on suicide mortality. The study used a two-part approach. First, we analyzed cohort data (2001-2017) to calculate incidence rate ratios (IRRs) and other model parameters related to OAT and suicide risk. Second, findings were applied to model outputs to estimate suicides averted by the NSW OAT program (2001-2020).

Setting and participants: A cohort of 46 845 individuals who received OAT between 2001 and 2017 in community and prison settings in New South Wales, Australia.

Measurements: IRRs for suicide and other model parameters were calculated for individuals on versus off OAT in community and prison settings (2001-2017). These estimates, along with model outputs, were used to determine the number and proportion of suicides averted by the OAT program (2001-2020).

Findings: Receiving OAT was associated with an IRR for suicide of 0.32 [95% confidence interval (CI) = 0.25-0.40] in the community and 0.34 (95% CI = 0.10-1.10) in prison for cohort data analyses (2001-2017). Between 2001 and 2020, the OAT program in NSW averted an estimated 338 suicides [95% credible interval (CrI) = 213-492), with 325 (95% CrI = 202-476) averted in the community and 13 (95% CrI = 0-46) in prison, corresponding to a 35% (95% CrI = 27%-43%) reduction in suicides among those accessing OAT.

Conclusions: The opioid agonist treatment program in New South Wales, Australia, was associated with a 35% reduction in suicide mortality among individuals with opioid use disorder receiving treatment between 2001 and 2020, providing novel evidence of its population-level impact on suicide prevention.