Cannabis use and psychosis

“Cannabis increases risk of psychosis in teens” is the headline in The Daily Telegraph today. Users of cannabis have a “higher average number of symptoms associated with a risk of psychosis”, the newspaper adds. The results of a study of more than 6,000 young people in Finland also suggests that teen users had a greater risk of the “prodromal”, or warning symptoms, of psychosis than older users.

The study behind the stories is a cross-sectional study of adolescents enrolled in a larger study. The design of this study means that it cannot prove that the link between these warning symptoms and cannabis use is a causal one. However, the findings add to the evidence that there is a link between use of cannabis and mental health. It is important that other factors which may affect the relationship are taken into account in such studies; this is not an easy task. Before definitive answers are found through research, it seems wise to avoid smoking cannabis, not only because of the debate around mental health, but also because of the well-known link between smoking and a host of other diseases, including lung cancer and heart disease.

Where did the story come from?

Dr Juoko Miettunen and colleagues from the University of Oulu in Finland carried out this study. The research was funded by the Academy of Finland, the Signe and Ane Gyllenberg Foundation, the Sigrid Juselius Foundation and the Thule Institute at the University of Oulu in Finland. It was published in the (peer-reviewed) medical journal: the British Journal of Psychiatry .

What kind of scientific study was this?

The study is a survey of young people who are enrolled in a prospective cohort study with their mothers. In 2001–2002, when the participants were aged about 15 or 16 years, they were invited to a clinical check-up where they were also given a questionnaire to assess “prodromal” (early warning of psychosis) symptoms and drug use.

Of the 9,340 children in the original cohort, 6,298 of them provided answers to questions about cannabis use, and could be included in the final analysis. The researchers used a shortened version (12 questions) of a longer questionnaire (21 questions) called the PROD-screen to assess prodromal symptoms of psychosis over the previous six months. The questions asked about whether the subject had a feeling that something strange or inexplicable is taking place in oneself or in the environment; feelings that one is being followed or influenced in some special way. From this, the researchers were able to discern which children were “at risk of developing a psychotic disorder”.

The researchers also had access to information about the young people’s early emotional and behavioural symptoms through questionnaires completed by their teachers when the participants were eight years old. They took these into account when analysing the relationship between early symptoms of psychosis and drug use. They also considered other factors that may have an effect on the relationship, such as gender, parent social class, tobacco use and use of other drugs, as well as parent substance misuse.

What were the results of the study?

The majority of adolescents reported that they never used cannabis (5,948/6,298). However, 352 (6%) participants reported having used cannabis (once or more). Girls were more likely than boys to have used cannabis. The study also found that those who had tried cannabis had a higher mean number of prodromal (early warning of psychosis) symptoms (3.11 v 1.88), and that compared with those who had never used, those who had tried cannabis (once or more) were twice as likely to have three or more prodromal symptoms (OR 2.23, 95% CI 1.70 to 2.94). This result took into account other factors that may have had an effect (e.g. age, gender, smoking, parental substance misuse etc.). The researchers also found that more intensive cannabis use was more strongly associated with these symptoms.

What interpretations did the researchers draw from these results?

The researchers conclude that lifetime cannabis use is associated with the incidence of early warning symptoms of psychosis.

What does the NHS Knowledge Service make of this study?

There are limitations to a study such as this that should affect the interpretation of results, particularly where causation is being claimed:

• As the researchers collected data on cannabis use and early warning symptoms at one point in time, this is a cross-sectional study. Due to their design, cross-sectional studies cannot establish causation. At best, the researchers can say that cannabis use is “associated with” or “linked to” the prodromal symptoms. Other factors that may be involved become important when interpreting results of studies such as these.
• Although the researchers took into account early emotional and behavioural problems around age eight, they did not account for mental problems that may have occurred between ages eight and 16 years. 
• Importantly, a positive “score” on the PROD-screen questionnaire does not diagnose psychosis. It is used to indicate whether a person is entering the period of early symptoms or changes in function that may come before psychosis. However, even for this, the score is not 100% accurate at predicting psychosis, or even proven as a tool for diagnosing the prodrome. The researchers used a shortened form of the original PROD-screen questionnaire (they reduced it from 21 questions to 12). It is unclear what effect this would have on the overall accuracy of the screening test. If it was too inclusive, i.e. there were a high number of false-positives, the relationship between drug use and symptoms would have been overstated.
• The researchers grouped all adolescents who had ever used cannabis into one category for analysis (i.e. they do not differentiate between adolescents who have tried cannabis once and those who are regular users).

Overall, this study points to an area that requires more research, but because of its design, the study does not prove that cannabis causes psychosis. Confirmation of the usefulness and accuracy of the PROD-screen in predicting increased risk of psychosis will also be important. When the findings are considered in light of a growing body of evidence of a link between cannabis use and mental health problems such as schizophrenia, it seems wise to limit the use of the drug. This is not only because of considerations of the effects on mental health, but also the well-established risks for cancer and other diseases that are associated with smoking.

Sir Muir Gray adds...

Dangerous drug, cannabis.