Intended for healthcare professionals

Research News

Smoking cigarettes may increase risk of schizophrenia, study shows

BMJ 2015; 351 doi: https://doi.org/10.1136/bmj.h3773 (Published 10 July 2015) Cite this as: BMJ 2015;351:h3773
  1. Nigel Hawkes
  1. 1London

Smoking may increase the risk of developing schizophrenia, an analysis by a team from King’s College, London has found.1

An association between cigarettes and psychotic symptoms has been reported before, but it has been generally assumed that this arises because people with the symptoms take up smoking as a way to relieve distress, counteract the symptoms, or manage the side effects of drugs used to treat them.

The King’s College team, from the Institute of Psychiatry, Psychology and Neuroscience, put this link to the test by examining the timing of smoking and the onset of symptoms.

“If smoking is a form of self medication, one would expect newly diagnosed patients to have normal smoking rates which would then rise in response to the symptoms,” James MacCabe, one of the authors, told a briefing at the Science Media Centre in London. “But if smoking is causative, one would expect that, at the time of the first psychotic episode, people would already be smoking more than average.”

The team also looked at whether regular smokers tended to develop psychotic symptoms earlier than non-smokers and whether early uptake of smoking was linked to earlier onset of symptoms. They used data from 61 observational studies that included 15 000 smokers and 273 000 non-smokers.

They found that 57% of people having a first episode of psychosis were smokers, three times the rate in the general population. In daily smokers, onset of the condition occurred on average a year earlier than in non-smokers; and, among those with established psychotic symptoms, smoking had started an average of six months earlier than in those without symptoms.

Summarising the findings Sameer Jauhar, a coauthor, said, “We found an increased risk of schizophrenia in daily smokers, a far higher smoking prevalence among those with a first psychotic episode, earlier onset in regular smokers, and an earlier start to smoking among those with symptoms.” (The last of these was not statistically significant.)

While all of these findings point to smoking being a cause of psychosis, the team said that it cannot entirely rule out reverse causation. MacCabe said, “While it is always hard to determine the direction of causality, our findings indicate that smoking should be taken seriously as a possible risk factor for developing psychosis, and not dismissed simply as a consequence of the illness.”

Since many smokers in Europe are also cannabis users the findings also raised the possibility that a link between cannabis use and schizophrenia could be partly accounted for by tobacco. But another coauthor, Robin Murray, said that he did not think this undermined the link between cannabis and schizophrenia, since it is also found in the United States and Australia, where cannabis users are less likely to be cigarette smokers. “It could be that if you smoke tobacco with your cannabis, you get a double whammy,” he said. “If you just smoke cannabis the risk could be more modest. That would be my view.”

Michael Bloomfield, a psychiatrist at the MRC Clinical Sciences Centre at University College London, who was unconnected with the study, said, “This new study combines previously published scientific data into a statistical analysis which found that smoking cigarettes appears to modestly increase the risk of developing schizophrenia in later life.

“There are a number of plausible ways to explain how this may be happening, such as by heavy cigarette smoking increasing the ability to make the chemical dopamine in part of the brain called the striatum, which is in turn thought to play an important role in the development of schizophrenia.

“However, much more research is needed before scientists can say for certain that smoking definitely increases the risk of schizophrenia, since it remains possible that people who would go on to develop schizophrenia are more likely to start smoking.”

Notes

Cite this as: BMJ 2015;351:h3773

References

View Abstract