Search   |  Advanced

Research Summary

Does Early Cannabis Use Lead to Psychosis?

Studies have suggested an association between cannabis use and psychosis-related outcomes. In this sibling-pair analysis of an Australian birth cohort, researchers interviewed 3801 young adults (53% of the original cohort) to assess age-of-onset of cannabis use as well as nonaffective psychosis, hallucinations, and Peters et al. Delusions Inventory (PDI) score at 21-year follow-up. The sample included 228 sibling pairs.

  • Those with 6 or more years since initiation of cannabis had an increased risk of nonaffective psychosis (adjusted odds ratio [AOR], 2.2), scoring in the highest quartile on the PDI (AOR, 4.2), and hallucinations (AOR, 2.8).
  • Within the sibling pairs, there was a modest association between years since first cannabis use and PDI score.
  • Notably, participants who reported hallucinations at the 14-year follow-up were more likely than those who didn’t to have longer times since first cannabis use and to use cannabis more frequently at the 21-year follow-up.


Unfortunately, design issues raise concerns about these results. Although sibling-pair analysis reduces the influence of genetic and/or environmental factors, it does not address recall bias (individuals with psychotic symptoms might be more likely to report early cannabis use); protopathic bias (individuals with preclinical manifestations of psychosis, such as hallucinations, might be more likely to initiate cannabis); or bias introduced by differential loss to follow-up. Since a randomized trial is not feasible, other ongoing studies should use prospective data and econometric methods to reach a more definitive conclusion. Peter D. Friedmann, MD, MPH


McGrath J, Welham J, Scott J, et al. Association between cannabis use and psychosis-related outcomes using sibling pair analysis in a cohort of young adults. Arch Gen Psychiatry. 2010;67(5):440–447.