In two European cohorts, cannabis use predicted psychotic-like experiences even after accounting for genetic risk for schizophrenia, with no evidence of gene-environment interaction.
Read this if you want to understand whether the cannabis-psychosis link is truly causal or just reflects genetic predisposition.
Independent riskCannabis predicted psychotic experiences regardless of genetic vulnerability to schizophrenia
What the researchers found
In the IMAGEN cohort (n=1,740) and replicated in the Utrecht cohort (n=1,223), both polygenic risk for schizophrenia (PRS-Sz) and lifetime cannabis use at age 16 independently predicted psychotic-like experiences. PRS-Sz predicted cannabis use (p=0.027) and psychotic-like experiences (p=0.004). Cannabis use remained significantly associated with psychotic-like experiences after accounting for PRS-Sz and covariates (p=0.007). Critically, there was no evidence that cannabis mediated or moderated the genetic risk — the two risk factors operated independently.
Why it matters
A common counterargument to cannabis causing psychosis is that genetically predisposed individuals simply use more cannabis. This study directly tests and refutes that hypothesis: cannabis remains a risk factor for psychotic symptoms regardless of genetic vulnerability.
The numbers in context
- IMAGEN cohort: n=1,740
- Utrecht replication cohort: n=1,223
- PRS-Sz predicted cannabis use: p=0.027
- PRS-Sz predicted PLEs: p=0.004
- Cannabis predicted PLEs adjusting for PRS-Sz: p=0.007
- No mediation or moderation effects detected
How the study worked
Analysis of the European IMAGEN cohort with replication in the Utrecht cannabis cohort. Tested mediation and moderation models between polygenic risk score for schizophrenia, lifetime cannabis use at age 16, and psychotic-like experiences (CAPE-42). Sensitivity analyses included PRS for cannabis use as covariate.
Who was studied
1,740 young adults from the IMAGEN European cohort; 1,223 from the Dutch Utrecht cohort
What this study cannot tell us
Psychotic-like experiences measured by self-report questionnaire, not clinical assessment. Lifetime cannabis use is a binary measure that does not capture frequency or potency. European cohorts may not generalize globally.
How to read the evidence
Two-cohort genomic epidemiology study with rigorous statistical testing of mediation and moderation. Strong evidence for independent risk.
When this study was published
Published in 2023. Uses state-of-the-art polygenic risk scoring to disentangle genetic and environmental factors.
The bigger picture
This study strengthens the case that cannabis is an independent environmental risk factor for psychosis, not merely a marker of pre-existing genetic vulnerability. This has direct implications for prevention strategies: reducing cannabis exposure could reduce psychosis risk even in people without genetic predisposition.
Questions still open
- Would dose-dependent cannabis measures show a stronger effect?
- Does cannabis-related psychosis risk vary across different ancestral backgrounds?
- Are there other genetic variants beyond PRS-Sz that might interact with cannabis?
Read the original research
Independent contribution of polygenic risk for schizophrenia and cannabis use in predicting psychotic-like experiences in young adulthood: testing gene × environment moderation and mediation.
Psychological medicine, 53(5), 1759-1769
Citation
Elkrief, Laurent; Lin, Bochao; Marchi, Mattia; Afzali, Mohammad H; Banaschewski, Tobias; Bokde, Arun L W; Quinlan, Erin Burke; Desrivières, Sylvane; Flor, Herta; Garavan, Hugh; Gowland, Penny; Heinz, Andreas; Ittermann, Bernd; Martinot, Jean-Luc; Martinot, Marie-Laure Paillère; Nees, Frauke; Orfanos, Dimitri Papadopoulos; Paus, Tomáš; Poustka, Luise; Hohmann, Sarah; Fröhner, Juliane H; Smolka, Michael N; Walter, Henrik; Whelan, Robert; Schumann, Gunter; Luykx, Jurjen; Boks, Marco P; Conrod, Patricia J. (2023). Independent contribution of polygenic risk for schizophrenia and cannabis use in predicting psychotic-like experiences in young adulthood: testing gene × environment moderation and mediation.. Psychological medicine, 53(5), 1759-1769. https://doi.org/10.1017/S0033291721003378