Tobacco Smoking and Psychotic-Like Experiences: Are They Linked?

Tobacco smoking is highly prevalent around the world. It is one of the main causes of premature mortality and is the primary cause of preventable disease and death. Its impact on the risk of cardiovascular diseases and cancers is acknowledged by physicians and the general population, but little is known about its correlation with mental health. Although the association between tobacco smoking and psychosis remains unclear, it is known that people with schizophrenia are more likely to smoke than the general population.

Many environmental factors are involved in the occurrence of psychosis, and, among them, tobacco use has emerged as a potential suspect. For decades, other substances (especially cannabis) have been a focus of research on psychosis and schizophrenia. Until recently, the status of tobacco smoking has not been considered in epidemiologic studies except as a comorbidity. The (false) assumption that tobacco use alleviates the symptoms of schizophrenia or even compensates cognitive deficiencies (so called “self-medication hypothesis”) has been defended for many years by research at least in part funded by the tobacco industry. While patients with schizophrenia are more prone to smoke because they fail in quitting, this does not account for the huge proportion of patients smoking (around 62%, according to a worldwide meta-analysis).

In this context, we recently published an epidemiologic study examining the association of tobacco smoking with psychotic-like experiences (PLEs). PLEs are defined as subclinical expression of psychosis and are reported in nonclinical settings, including the general population. They represent a valuable indicator of mental health in the spectrum for psychosis.

We studied data from the Wave 2 of the NESARC, a large US population-based, nationally representative sample, conducted with 34,653 adults. Participants were assessed in a face-to-face interview. Among other questions, twenty-two items covered the broad spectrum of psychosis. Participants were stratified according to their smoking status and other uses of tobacco. We found a significant association between smoking status and 14 of the 22 assessed PLEs. These associations remained significant after adjustment for common known environmental risk factors for psychosis: sociodemographic variables (eg, urbanicity, ethnicity), lifetime drug use disorder, and past-year cannabis use. While about one-quarter of nonsmokers reported at least one PLE, this prevalence was slightly higher in former smokers and rose as high as 40% in current smokers. All of the 22 PLEs had a higher prevalence in smokers than in former smokers or lifetime abstainers. About 9% of smokers reported at least 5 PLEs, compared to 3.5% of lifetime abstainers.

In conclusion, smoking status was associated with various PLEs, and this association was not explained by other known risk factors for PLEs or schizophrenia, which raises both research and clinical questions. There is a need to identify the potential neurobiological mechanisms by which tobacco smoking and mental health are linked both to provide help for individual patients and to inform the public health perspective.

Financial disclosure:Drs Mallet, Dubertret, and Le Strat have no relevant personal relationships to report.​