As the global prevalence of cannabis use disorder increases, there are concerns about its association with self-harm and suicide. We conducted a retrospective population-level study in Ontario, Canada of all individuals aged 15-105. Deliberate self-harm and deaths due to suicide were compared between individuals with an incident emergency department (ED) visit or hospitalization for cannabis use, matched with members of the general population (primary analysis) or people with an ED visit for alcohol use (secondary analysis), using adjusted hazard models. Models were adjusted for age and sex and further adjusted for pre-specified variables including sociodemographics, co-morbid mental and chronic health conditions and substance use. We followed 11,320,897 people of whom 85,108 (0.75%) had an incident cannabis-related hospital contact. We included a total of 861,291 people with a median follow-up of 5 (IQR 3-7) years in our primary matched analysis. People with a cannabis-related hospital contact had a 5.4-fold (adjusted Hazard Ratio [aHR] 5.35 95%CI5.04-5.67) increased risk of deliberate self-harm and a 9.2-fold (aHR 9.22 95%CI5.24-16.23) increased risk of death by suicide within 3 years relative to the general population. Among individuals with a cannabis-related hospital contact, the risk of self-harm was elevated in those with and without co-morbid mental health or substance use disorder. Sensitivity analyses excluding those with co-morbid mental health or substance use disorders continued to show increased risk of deliberate self-harm relative to the general population (aHR 7.18 95%CI6.26-8.23). In the secondary analysis, people with an alcohol related hospital contact were at a higher risk of self-harm relative to cannabis (aHR 1.22 95%CI1.18-1.26). These findings highlight that the growing prevalence of cannabis use disorder may have important implications for mental health, self-harm and suicide.
