Mental health

More evidence on cannabis psychosis

“Young people who use cannabis are doubling their risk of developing psychotic symptoms,” the Daily Mail has reported. Mental health problems may also persist among continual users, the newspaper added.

The news is based on a study involving nearly 2,000 German adolescents and young adults. It found that new cannabis use almost doubled the risk of psychotic symptoms in the years after use. The study also found that these users had been free from psychotic symptoms prior to smoking cannabis. Previously, it has not been clear whether cannabis use leads to psychotic symptoms or whether young people with psychotic symptoms use cannabis to “self-medicate”.

It should be noted that self-reported psychotic symptoms were assessed rather than clinically-diagnosed psychotic problems. Psychotic symptoms are not uncommon in the general population. But overall, this large, well-designed study supports the results of previous research on the matter, suggesting there is an association between cannabis use and psychotic symptoms.

Where did the story come from?

The study was carried out by researchers from Maastricht University, Netherlands; the University of London, UK; the University of Basel, Switzerland; the Max Plank Institute of Psychiatry, Germany, and Technical University Dresden, Germany. The study was funded by the German government and was published in the peer-reviewed British Medical Journal.

Generally, the study was reported accurately in the papers. The Daily Telegraph reported comments from external experts, one of whom pointed out that the study did not distinguish between different types of cannabis. However, some newspapers featured the specific claim that cannabis use doubles the risk of psychosis, which could be considered to be inaccurate as the study found that using cannabis doubled the risk of self-reported psychotic symptoms rather than clinically-diagnosed psychotic illness. None of the papers pointed out that the study relied on young people recalling psychotic symptoms, rather than on clinical diagnosis.

What kind of research was this?

This was a prospective cohort study of nearly 2,000 German adolescents and young people who were followed-up for a period of 10 years, to determine whether use of cannabis in adolescence increases the risk of “subclinical” psychotic symptoms (i.e. symptoms below the level required for a clinical diagnosis). It looked at both ‘incident’ (i.e. new) psychotic symptoms in cannabis users and non-users. It also looked at whether psychotic symptoms persisted in those who used cannabis.

The authors say that cannabis use is already associated with an increased risk of psychotic disorders. However, it is not known whether cannabis use itself increases the risk or if the association is due to people with pre-existing psychotic symptoms tending to use cannabis as a form of “self-medication”. Also, the mechanism by which cannabis use might cause psychotic symptoms is not yet understood. In this longitudinal study, the researchers set out to look at first-time cannabis use in relation to first-time psychotic symptoms.

What did the research involve?

This was a population-based cohort study that took place in Germany, involving 1,923 participants from the general population. The participants were aged between 14 and 24 at the start of the study. The sample had been drawn from an earlier study, which collected data on mental disorders in a random sample of adolescents and young adults.

The researchers collected information from participants on both cannabis use and “subthreshold” psychotic symptoms at the start of the study (baseline). They also collected information at three further points in time: on average, these were 1.6 years (T1), 3.5 years (T2) and 8.4 years (T3) after the study began. In collecting information, they used a validated diagnostic interview which assesses symptoms, syndromes and diagnoses of various mental disorders in accordance with internationally agreed definitions.

The interviews were conducted by trained clinical psychologists. The diagnostic interview also included questions about substance use. The presence of psychotic experiences, as defined by the diagnostic interview, included symptoms such as delusions, hallucinations, feelings of persecution and thought interference.

Within the same interview, participants were also asked if they had used cannabis five times or more. Cannabis use on at least five occasions was used to define cannabis exposure, which was recorded as either “yes” or “no”.

The researchers then used standard statistical methods to assess the relationship between cannabis use and both new and persistent psychotic symptoms. The results were adjusted for “confounders” that might have influenced results, such as sex, age, social and economic status, and other drug use. However, researchers were unable to take account of family history of psychosis, which could have influenced the risk of psychotic symptoms.

What were the basic results?

They found that:

  • In young people who had not reported psychotic symptoms or cannabis use at baseline, starting to use cannabis between baseline and the T2 phase increased the risk of later having new (incident) psychotic symptoms over the period from T2 to T3 (adjusted odds ratio [AOR] 1.9, 95% confidence interval 1.1 to 3.1).
  • Continued use of cannabis increased the risk of persistent psychotic symptoms over the period from T2 to T3 (AOR 2.2, 95% CI 1.2 to 4.2).
  • 31% (152) of people who had been exposed to cannabis reported psychotic symptoms over the period from baseline to T2, compared to 20% (284) of individuals who had not been exposed.
  • Over the period from T2 to T3, 14% (108) of those exposed to cannabis reported psychotic symptoms, compared to 8% of the participants who were not exposed.

How did the researchers interpret the results?

The researchers say their study has found that cannabis use is a risk factor for the development of incident (i.e. new) psychotic symptoms, and that cannabis use precedes the onset of psychotic symptoms. They also say that continued cannabis use increased the risk of symptoms persisting. Therefore it might increase the risk of psychotic disorder.

The researchers also suggest that the increased risk seen with cannabis use may be due to repeated exposure to THC (the main psychoactive component of cannabis), although there is currently a lack of evidence for this in humans.


This large, well-designed study suggests that new cannabis use carries a risk of later psychotic symptoms in young people who had not previously had psychotic symptoms. It also suggests that continuing cannabis use might lead to persistent psychotic symptoms and that this could increase the risk of developing psychotic illness. However, the study had several limitations, some of which the authors have noted:

  • It relied on self-reported information about both psychotic symptoms and cannabis use. This could potentially introduce error, although the authors say this possibility was minimised through their interviews being conducted by trained clinical psychologists.
  • The study did not adjust for family history of psychosis, a possible confounding factor. The authors say they may have indirectly adjusted for this to some degree, though.
  • The authors say they used a “broad outcome measure” to represent psychotic experiences, rather than clinically relevant psychotic disorder. However, they say that psychotic experiences show “continuity” with psychotic disorders.
  • The study may have had its results influenced by “selective recall” about cannabis use and psychotic symptoms, i.e. the participants may have intentionally or unintentionally modified their answers to support their personal views on the matter. The long-term nature of this study may increase the risk of this occurring as the participants would have known the purpose and methods of the study and could have modified their answers at later interviews.

In conclusion, this study’s findings are a valuable addition to the research on the possible association between cannabis use and psychotic symptoms, particularly because it was able to show that use of cannabis preceded psychotic symptoms. However, further research is required into any association between use of cannabis and more long lasting, clinically diagnosed psychotic disorders.

NHS Attribution