Cannabis is the most widely consumed illicit psychoactive substance globally, and is associated with both schizophrenia and a range of externalising behaviours (Pasman et al. 2018). Current hypotheses regarding the causal relationship between cannabis use and schizophrenia are reliant on observational studies, genetic correlation estimates and Mendelian Randomisation studies, but any assumption of causality remains controversial (Pasman et al. 2018; Vaucher et al. 2018). Genetic studies of causality have conflicted on both the direction and magnitude of the causal relationship, and so far have not fully explained this complex relationship or investigated the majority of other influences such as externalising behaviours (Gage et al. 2017; Pasman et al. 2018).
We have previously determined that up to 42% of the association between schizophrenia and cannabis use is explained by the influence of the externalising spectrum, but these analyses did not allow us to make inferences about a causal relationship (Wormington et al. 2022). This suggests that the externalising spectrum may partially mediate the causal relationship between schizophrenia and cannabis use. The aim of our current work is to investigate to what extent the causal relationship between cannabis use and schizophrenia is mediated by externalising behaviours. We have found that the externalising spectrum reduces the causal influence of schizophrenia on cannabis use by about 40%, but this reduction was not seen for the causal influence of cannabis use on schizophrenia. Risk-taking alone also reduced the causal influence of schizophrenia on cannabis use by about half, suggesting it may represent a modifiable risk factor to reduce likelihood of cannabis use in those with a genetic predisposition to schizophrenia.