rethinkTHC Search
Menu
Study breakdown

Genetic risk for schizophrenia did not predict cannabis use disorder in the general population

ObservationalStrong evidence
The takeaway

In a study of 88,637 individuals, polygenic risk scores for schizophrenia did not predict cannabis use disorder in controls, challenging the idea that shared genetics explain the cannabis-schizophrenia link.

Psychiatric genetics researchers and clinicians interpreting the cannabis-psychosis relationship.

CUD predicted schizophrenia (HR=4.91) but schizophrenia genetics did not predict CUD

What the researchers found

Schizophrenia polygenic risk scores did not predict CUD in controls (HR=1.16, not significant) or in patients with other psychiatric disorders. However, ADHD polygenic risk did predict CUD (HR=1.27 per SD increase, and HR=2.02 for highest decile). CUD was a strong predictor of schizophrenia (HR=4.91), and adjusting for various polygenic risk scores did not weaken this association.

Why it matters

If shared genetics explained the cannabis-schizophrenia association, then genetic risk for schizophrenia should predict who develops CUD. It did not. This finding strengthens the case that cannabis use may have a more direct role in schizophrenia risk, rather than both being driven by the same genetic vulnerability.

The numbers in context

88,637 individuals. Schizophrenia PRS predicting CUD in controls: HR=1.16 (95% CI 0.95-1.43, not significant). ADHD PRS predicting CUD: HR=1.27 (95% CI 1.08-1.50). CUD predicting schizophrenia: HR=4.91 (95% CI 4.36-5.53).

How the study worked

Linked Danish nationwide registers with genetic information from neonatal bloodspots in 88,637 individuals. Compared polygenic risk scores for schizophrenia, ADHD, autism, and anorexia as predictors of CUD diagnosis across controls, schizophrenia patients, and patients with other psychiatric disorders.

What this study cannot tell us

Polygenic risk scores capture only a fraction of genetic risk. Negative findings could reflect insufficient statistical power for schizophrenia PRS. Danish population may not generalize globally. Registry-based CUD diagnosis may miss milder cases.

How to read the evidence

Very large genetically informed sample with national registry linkage. Strong methodology for testing the shared-genetics hypothesis.

When this study was published

2021 study using Danish national registers and neonatal genetic data.

The bigger picture

The ADHD genetic finding is notable: ADHD genetic risk predicted CUD while schizophrenia genetic risk did not. This suggests that the pathways leading from genetic vulnerability to cannabis problems may involve impulsivity and attention regulation rather than psychosis susceptibility.

Questions still open

  • Why does ADHD genetic risk predict CUD while schizophrenia genetic risk does not? Does this pattern hold across different populations? What does the persistent CUD-schizophrenia association (HR=4.91) represent if not shared genetics?

Common questions

Does this prove cannabis causes schizophrenia?
Not directly, but it removes one alternative explanation. The finding that schizophrenia genetic risk does not predict CUD argues against shared genetics being the primary explanation for the cannabis-schizophrenia link.
Why did ADHD genetics predict cannabis use disorder?
ADHD involves impulsivity and executive function challenges that may increase vulnerability to developing problematic substance use patterns, including cannabis use disorder.

Read the original research

No evidence of associations between genetic liability for schizophrenia and development of cannabis use disorder.

Psychological medicine, 51(3), 479-484

Citation

Hjorthøj, Carsten; Uddin, Md Jamal; Wimberley, Theresa; Dalsgaard, Søren; Hougaard, David M; Børglum, Anders; Werge, Thomas; Nordentoft, Merete. (2021). No evidence of associations between genetic liability for schizophrenia and development of cannabis use disorder.. Psychological medicine, 51(3), 479-484. https://doi.org/10.1017/S0033291719003362

Explore the wider topic