Cannabis use linked to psychosis later in lifeBMJ 2007; 335 doi: http://dx.doi.org/10.1136/bmj.335.7613.228-c (Published 02 August 2007) Cite this as: BMJ 2007;335:228
Young people should be warned that cannabis will probably increase their risk of a psychotic illness, conclude the authors of a comprehensive meta-analysis. Two linked comments agree (p 292, p 293), and one estimates that 800 cases of schizophrenia could be prevented each year in the UK if all cannabis consumption ceased.
After pooling data from seven cohort studies, the authors found that people who had ever used cannabis increased their odds of a psychotic illness or symptoms by around 40%, compared with those who had never used it (adjusted odds ratio 1.41, 95% CI 1.20 to 1.65). They also found a clear dose-response effect. The odds ratio for the heaviest users was 2.09 (1.54 to 2.84). There may also be a link between cannabis and other disorders such as anxiety and depression, but the data were weaker and less consistent.
We will never know for certain whether the reported association between cannabis and psychosis is causal, because randomised trials are impossible. But these authors looked carefully for other explanations, adjusted for dozens of confounding variables, and still the association persisted. If cannabis causes later psychosis in even a small number of users, the effect on public health would be substantial. Around 40% of young people in the UK have used it at least once.