Aims: To assess if cannabis use is a risk factor for future psychotic symptoms, and vice versa, in adolescents and young adults from the general population.
Design: Cohort study.
Setting/participants: 'Zuid Holland' study, a 14-year follow-up study of 1580 initially 4-16-year-olds who were drawn randomly from the Dutch general population. Because cannabis use is generally condoned in the Netherlands, false-negative reports of cannabis use may occur less frequently than in countries with stricter drug policies, which supports the value of the present study.
Measurements: Life-time cannabis use and psychotic symptoms, assessed with the Composite International Diagnostic Interview (CIDI).
Findings: Cannabis use, in individuals who did not have psychotic symptoms before they began using cannabis, predicted future psychotic symptoms (hazard ratio = 2.81; 95% confidence interval = 1.79-4.43). However, psychotic symptoms in those who had never used cannabis before the onset of psychotic symptoms also predicted future cannabis use (hazard ratio = 1.70; 95% confidence interval = 1.13-2.57).
Conclusions: The results imply either a common vulnerability with varying order of onset or a bi-directional causal relationship between cannabis use and psychosis. More research on patterns and timings of these relationships is needed to narrow down the possibilities.