rethinkTHC Search
Menu
Study breakdown

Twin Study Shows Cannabis Use Disorder Leads to Psychotic Experiences Through Both Genetic and Direct Effects

Cross SectionalStrong evidence
The takeaway

In nearly 2,800 twins, the link between cannabis use disorder and psychotic experiences was explained by both shared genetics and a direct causal pathway from cannabis to psychosis, not the reverse.

Read this if you want to understand whether cannabis causes psychosis or whether people prone to psychosis simply use more cannabis.

Direction of causation modeling supported cannabis causing psychotic experiences, not the reverse

What the researchers found

This twin study of 2,793 young adults disentangled the genetic and environmental contributions to the cannabis-psychosis relationship.

Cannabis use disorder symptoms were strongly associated with psychotic-like experiences, with an incidence rate ratio of 6.3. Within twin pairs (controlling for shared genetics and family environment), the association remained significant at 3.5, confirming that factors beyond family background contribute.

Heritability of cannabis use disorder symptoms was 88% in both men and women. Heritability of psychotic experiences was 77% in men and 43% in women, revealing a striking sex difference.

The genetic correlation between cannabis use disorder and psychotic experiences was 0.55, and the environmental correlation was 0.52, meaning both shared genes and shared environments contribute to the association.

Critically, the model testing whether cannabis causes psychotic experiences fit the data significantly better than models proposing the reverse direction (psychosis leading to cannabis use) or reciprocal causation. This supports a directional effect from cannabis to psychosis.

Why it matters

Twin studies are among the most powerful methods for disentangling genetic and environmental influences. This study provides strong evidence for a causal direction from cannabis to psychosis, not the other way around, while also showing that shared genetics play a substantial role.

The numbers in context

2,793 twins. Lifetime cannabis use: 10.4%. PLEs prevalence: 0.1-2.2%. CUD-PLE incidence rate ratio: 6.3 total, 3.5 within pairs. CUD heritability: 88%. PLE heritability: 77% (men), 43% (women). Genetic correlation: 0.55. Environmental correlation: 0.52.

How the study worked

Population-based twin study of 2,793 young adults (63.5% female, mean age 28.2). Cannabis use disorder symptoms and psychotic-like experiences assessed via CIDI. Item Response Theory modeled latent risk. Co-twin control analysis assessed familial confounding. Biometric models estimated heritability, genetic/environmental correlations, and direction of causation.

What this study cannot tell us

Cross-sectional assessment limits causal inference despite the direction-of-causation modeling. Self-reported cannabis use and psychotic experiences may be subject to bias. The predominantly female sample (63.5%) may affect generalizability. Psychotic-like experiences are subclinical and may not predict full psychotic disorders.

How to read the evidence

Large population-based twin study with advanced biometric modeling. Strong because twin methodology controls for genetic and shared environmental confounding.

When this study was published

Published in 2017.

The bigger picture

The directional finding (cannabis to psychosis, not reverse) is significant because it counters the argument that people predisposed to psychosis simply self-medicate with cannabis. Both shared genetics and a direct effect contribute, but the arrow points from cannabis use disorder toward psychotic experiences.

Questions still open

  • Could genetic testing identify individuals most vulnerable to cannabis-induced psychotic experiences? Does the sex difference in heritability of psychotic experiences explain why men are more affected by cannabis-psychosis associations? Would the directional effect hold in clinical psychosis rather than subclinical experiences?

Common questions

Does cannabis cause psychosis or do psychosis-prone people use cannabis?
This twin study found evidence for both shared genetics and a direct effect from cannabis to psychosis. The model where cannabis causes psychotic experiences fit the data better than the reverse direction, supporting a causal role for cannabis.
How much of the cannabis-psychosis link is genetic?
The genetic correlation was 0.55, meaning shared genetic factors explain a substantial portion of the association. However, the within-twin-pair analysis and direction-of-causation modeling showed that direct effects of cannabis also contribute beyond genetics.

Read the original research

Genetic and Environmental Contributions to the Association Between Cannabis Use and Psychotic-Like Experiences in Young Adult Twins.

Schizophrenia bulletin, 43(3), 644-653

Citation

Nesvåg, Ragnar; Reichborn-Kjennerud, Ted; Gillespie, Nathan A; Knudsen, Gun Peggy; Bramness, Jørgen G; Kendler, Kenneth S; Ystrom, Eivind. (2017). Genetic and Environmental Contributions to the Association Between Cannabis Use and Psychotic-Like Experiences in Young Adult Twins.. Schizophrenia bulletin, 43(3), 644-653. https://doi.org/10.1093/schbul/sbw101

Explore the wider topic