BACKGROUND: The association between cannabis use and the risk of psychosis has been studied extensively but the temporal order still remains controversial. Aims To examine the association between cannabis use in adolescence and the risk of psychosis after adjustment for prodromal symptoms and other potential confounders. METHOD: The sample (n = 6534) was composed of the prospective general population-based Northern Finland Birth Cohort of 1986. Information on prodromal symptoms of psychosis and cannabis use was collected using questionnaires at age 15-16 years. Participants were followed up for ICD-10 psychotic disorders until age 30 years using nationwide registers. RESULTS: The risk of psychosis was elevated in individuals who had tried cannabis five times or more (hazard ratio, (HR) = 6.5, 95% CI 3.0-13.9). The association remained statistically significant even when adjusted for prodromal symptoms, other substance use and parental psychosis (HR = 3.0, 95% CI 1.1-8.0). CONCLUSIONS: Adolescent cannabis use is associated with increased risk of psychosis even after adjustment for baseline prodromal symptoms, parental psychosis and other substance use. Declaration of interest None.