In this report, we show that the association between state medical marijuana policy and suicide risk was not statistically significant, nor even suggestive of a protective effect, after adjustment for key covariates. Though an earlier report demonstrated an apparent correspondence between the legalization of medical marijuana and a decrease in suicide rates among men (Anderson et al., 2014), incorporation of demographic variables and additional state characteristics into the regression models reduced the magnitude of this association such that it was no longer consistent with even a modest protective effect.
