The reason(s) underpinning the strong association between cannabis use disorder (CanUD) and schizophrenia remains a passionately debated issue in psychiatry. Does cannabis use increase the risk for schizophrenia, or does schizophrenia promote cannabis use? Or could there be a shared liability, due to common genetic and environmental factors, that explains why these disorders so frequently co-occur? To shed light on this debate, Johnson and colleagues [1] investigated the role of tobacco use, which is strikingly common among people with schizophrenia and those with CanUD. As substance use disorder contributes to the reduced life expectancy in people with schizophrenia, and as more jurisdictions move toward cannabis legalization, finding answers to these questions is more important than ever.

Twin-based heritability estimates for CanUD and schizophrenia are ~50% and ~80%, respectively, indicating a strong genetic underpinning. Similarly, tobacco smoking is highly heritable (~50% [2]). In support of a shared liability, large genome-wide association studies (GWASs) of CanUD, schizophrenia, and tobacco smoking have identified several shared loci [3, 4], yet few studies have investigated the genetic interplay between the three traits. Mendelian Randomization (MR), a causal inference approach in which genetic information is used to ‘randomize’ individuals to an ‘exposure’, has brought us a step closer in understanding potential causal links. With MR, hypotheses in both directions can be tested (i.e. does CanUD cause schizophrenia, and/or does schizophrenia cause CanUD?). Results from MR studies have been mixed, however, with some showing greater support for schizophrenia causing CanUD, and others providing support for reciprocal causal effects (i.e. CanUD may also cause schizophrenia). To extend this work, Johnson and colleagues [1] used a suite of analytic techniques to disentangle genomic risk factors, and unveil potential causal relationships, between CanUD, schizophrenia, and tobacco smoking.

Log in or create a free account to read this content

Gain free access to this article, as well as selected content from this journal and more on nature.com

Access through your institution

or

Sign in or create an account
Continue with Google
Continue with ORCiD