Findings from this study suggest that amongst patients with a psychiatric comorbidity, women, unemployed individuals, and those with mood disorders are at a significantly heightened risk of attempting suicide. Heaviness of cannabis use was also found to be associated with an increased risk of suicide attempt in men in a subset of the study sample. Contrastingly, no association was found between the risk of attempting suicide and cannabis use, marital status, age, or having an anxiety or psychotic disorder. Interestingly, the negative impact of a mood disorder and the protective effect of employment were more pronounced in men compared to women with psychiatric disorders.

While cannabis use in the general population leads to an increased risk of suicide attempts as seen in some studies, our findings suggest that this association does not hold true in a large cohort of psychiatric patients who are at an already heightened risk of attempting suicide [6–9]. Heaviness of cannabis use was, however, found to have a slight but significant association with suicide attempt in men in a subset of our study sample. The pathophysiological link between cannabis use and suicidal behavior in the general population has yet to be established, though both direct and indirect associations between cannabis use and suicide attempt have been hypothesized [9]. One hypothesis suggests that tetrahydrocannabinol (THC, the active ingredient of cannabis) has direct neurophysiological effects that lead to impaired cognition and behavior, whereas another hypothesis suggests that cannabis use indirectly increases one’s risk for suicide attempt because cannabis users tend to have other predisposing social factors that increase their risk of attempting suicide [9]. Having a psychiatric disorder including other substance use is likely associated with altered neurophysiological state and therefore the effect of cannabis is less well known and difficult to isolate from the underlying psychopathology [32, 33].

It is also likely that the association between cannabis use and suicide risk described in the literature varies by the type of suicidal behavior. We have previously shown that certain epidemiological observations of suicide risk factors, such as obesity, differ for suicide ideation, attempts, and completed suicide [34]. In addition, we have also found that genetic risk factors associated with suicide also varied by the type of suicidal behavior [35]. Taken together, these observations call for homogenous definitions of suicidal behaviors in order to infer conclusions with any certainty on the association between certain risk factors and suicide.

Our significant findings related to suicide attempts and risk factors are mostly supported by existing literature. Previous studies have established that while men are at an increased risk of completing suicide, the prevalence of attempted suicide is significantly higher in women amongst the general population [16, 17]. Amongst psychiatric disorders, mood disorders are the most strongly associated with suicidal attempts [36]. Similarly, unemployment has been linked to a heightened risk of suicide. Whether this association is directly causal or indirectly associated with suicide by precipitating other risk factors for suicide, such as depression, has yet to be established [17, 37, 38]. Interestingly, the association between having a mood disorder and attempting suicide was twice as prominent in men as it was in women. Similarly, employment was more protective against suicide attempt in men compared to women. While such differences have not been previously reported, a plausible explanation may be the fact that women are already at a heightened risk of attempting suicide such that these additional predisposing factors may play a smaller role. A similar rationale may explain the significant association between heaviness of cannabis use and suicide attempt in men but not in women. This finding has not previously been reported in this population, and previous literature has actually revealed that even amongst cannabis users, women are more likely than men to attempt suicide [9]. It is possible that, given that women experience deleterious effects of cannabis use at lower doses and more frequently than men as we describe above, the dose response relationship is less pronounced [22, 24]. Ultimately, our study adds to the existing body of literature showing that the elevated suicidal behavior risk amongst women, the unemployed, and those with a mood disorder seen in the general population remains consistent in a cohort of psychiatric patients and that the effects of unemployment, mood disorders, and heavy cannabis use are more pronounced in men compared to women.

Our current study findings advance knowledge regarding suicide risk in psychiatric patients and warrant further investigation into the effect of cannabis on specific psychiatric disorders. With the WHO’s plans to reduce suicide rate by 10% by 2020, it is imperative that we establish clear risk factors to assist in stratifying individuals’ suicidal risk [18]. While such data exist on a general population level, they are lacking amongst psychiatric patients who comprise a large proportion of patients with suicidal behavior. The prevalence of suicide attempt in our patient sample (29.7%) falls on the higher end of the spectrum of what has previously been reported for psychiatric patients [39]. Given that 50% of patients who attempt suicide are known to have a concurrent substance use disorder, this finding is expected as the majority of patients in our sample have a substance use disorder [40].

Our study reveals that in a large population of psychiatric patients, women, those who are unemployed, those who have a mood disorder, and men with heavier cannabis use are at a heightened risk of attempting suicide and may therefore require closer follow-up, additional counseling, and/or screening for underlying mental health processes to mitigate the risk of SB [41]. Contrastingly, cannabis use as a whole does not seem to add to the suicide risk in this patient population despite its established association with suicide risk in the general population, though this effect may vary in certain subgroups and based on the amounts used. Nonetheless, cannabis use has been associated with other psychopathology such as mood and psychotic symptoms which may in turn increase the risk of suicide and therefore cannabis use in this high-risk population should not be overlooked [42].