Recent studies have shown that cannabis use acts as a specific risk factor provoking the onset of psychosis in vulnerable individuals. Association of adolescent cannabis use and psychosis risk was studied after adjustment with prodromal symptoms.

To assess possible causality between cannabis use and the risk of psychosis.

To examine associations between cannabis use and the risk of psychosis in 10 years follow-up while taking into account the prodromal symptoms of psychosis in a prospective general population sample.

The sample (N=6258) composed of a prospective Northern Finland Birth Cohort 1986. Questionnaire on prodromal symptoms for psychosis (PROD-screen) and on drug use was conducted when the cohort members were 15-16 years old. The participants were asked if they had tried cannabis: never, once, 2-4 times, 5 times or more. Information on psychoses was gathered from registers until age 27 years.

In total 102 new psychoses emerged. The proportion of psychoses in the groups 'never”, 'once”, '2-4 times”, '5 times or more” were 1.5%, 2.8%, 3.6%, and 8.5%, respectively. The hazard ratio (HR) for risk of psychosis in subjects who had tried cannabis 5 times or more was 5.9 (95% CI 2.4-14.4) when compared to non-users. The association remained statistically significant when adjusted for prodromal symptoms and parental psychosis (HR 2.6, 1.0-6.6). When gender and smoking was taken into account association was no longer significant (HR 2.3, 0.9-6.0).

Adolescent cannabis use associates with increased risk of first-episode psychosis even after controlling for baseline prodromal symptoms.