Cannabis use and suicide: a case-control study based on integrative data analysis

Cannabis use has been identified as a risk factor for depression, suicidal ideation, and suicide attempts. However, the direct relationship between cannabis use and suicide death has not been adequately investigated due to data and methodological challenges. We assessed the association between cannabis use and suicide using the case-control design and integrative data analysis techniques.

Cases consisted of suicide decedents aged 16 years and older, selected from the 2013 National Violent Death Reporting System (NVDRS). Controls were respondents from the 2013 National Survey on Drug Use and Health (NSDUH). To mitigate potential misclassification in the NSDUH respondents' self-reported cannabis use, we employed a data fusion approach by integrating the NSDUH data with data from the 2013 National Roadside Survey of Alcohol and Drug Use by Drivers (NRS), which included both toxicological testing and self-reported data on cannabis use. For statistical analysis, adjusted odds ratios (aORs) and 95% confidence intervals (CIs) of suicide associated with cannabis and alcohol use were estimated using weighted multivariable logistic regression models.

Multiply imputed toxicological testing data indicated that 17.1% of the cases, and 7.2% of the NSDUH controls used cannabis as indicated by testing positive for delta-9-tetrahydrocannabinol. Weighted multivariable logistic modeling revealed that cannabis use was associated with 83% increased odds of suicide (aOR = 1.83; 95% CI: 1.36, 2.31) and that alcohol use was associated with 20-fold increased odds of suicide (aOR = 20.53; 95% CI: 11.83, 29.24). Other factors associated with significantly increased odds of suicide were male sex, White race, being 35-49 years of age, and having less than a high school education.

Cannabis use is a significant risk factor for suicide, independent of alcohol use and demographic characteristics. Policy makers should take into consideration the excess risk of suicide associated with cannabis use when assessing the externalities of legalizing cannabis.