June 18, 2012

What Should We Do About the Link Between Smoking and Psychosis?

Author Picture

Hannah Newall, MBBS

The Queen Elizabeth Hospital, Woodville West, SA, Australia


In The Journal of Clinical Psychiatry, my colleagues and I published a meta-analysis entitled “Tobacco Use Before, At, and After First-Episode Psychosis: A Systematic Meta-Analysis.” The aim of this meta-analysis was to explore the complex relationship between tobacco use and the early phase of psychotic illness. We hoped that defining when tobacco use begins in relation to the onset of psychosis and the course of tobacco use after the development of psychotic illness would help to explain why tobacco use is so prevalent in this population and might also help improve our understanding of the illness itself.

The meta-analysis confirmed and quantified one point that is widely acknowledged, ie, the majority of patients who present with first-episode psychosis already have a well-established tobacco habit. In fact, tobacco use at the time of presenting for treatment of first-episode psychosis was 59%, or 6 times the rate of tobacco use in age- and gender-matched controls.

What was astonishing, though, was that regular tobacco use in these patients preceded onset of psychosis by a mean of about 5 years. The proportion of tobacco users in this population changed little over time (in up to 10 years of follow-up).

These findings raise many questions. In this blog, we invite discussants to consider 2 questions:

  1. What is the nature of the relationship between smoking and psychosis?
  2. What ought we to do to minimize the rate of smoking among those with early psychosis?

Neither of these questions has clear answers. Is it possible that smoking is a cause of psychosis? Smoking certainly predates treatment for psychosis by many years, making theories about using tobacco as self-medication for positive symptoms or to combat side effects of antipsychotics less likely. Critical prospective studies on this relationship, however, have yielded conflicting results, and, unlike with cannabis use, tobacco use is not associated with an earlier onset of psychosis. It is perhaps surprising that little attention has been paid to tobacco as a potential etiologic factor in psychiatric illness, perhaps because clinicians have accepted smoking among their patients in a way that generalists have not.

Smoking, however, contributes to the high rates of vascular disease and malignancy experienced by mentally ill people that, along with suicide and misadventure, are major causes of premature mortality. But what, then, should we do? Certainly, we should try to motivate our patients to quit. This should be done as early as possible, during early psychosis and among those at high-risk for psychosis. Medical options such as nicotine replacement therapy should not be ignored; however, we must also acknowledge that the effective smoking cessation strategies being used in the general population have not had the same success in this population, leaving another vast gap in health equality.

The link between tobacco smoking and psychosis is undebatable, and the associated impacts on health are drastic. Should we continue investing research dollars into investigating the cause of the link or instead focus our efforts on combating this major health problem with unique strategies designed for this population?

Financial disclosure:Dr Newall had no relevant personal financial relationships to report. ​

Category: Psychosis , Schizophrenia
Link to this post:
Related to "What Should We Do About the Link Between Smoking and Psychosis?"

11 thoughts on “What Should We Do About the Link Between Smoking and Psychosis?

  1. I play some role in the care of someone with bipolar 1 disorder that experiences psychotic mania. During each manic episode I have witnessed she develops an addiction to tobacco that appears to be real. Then when the psychosis ends (after treatment in hospital with antipsychotic medication) her addiction to tobacco vanishes, she stops smoking immediately, and experiences no withdrawal. Indeed, when depressed or euthymic she has an aversion to tobacco smoke.

    Aside from the obvious link between dopamine and nicotine and dopamine and psychosis a review of the literature suggests that PKC signalling that is associated with manic psychosis is also implicated in addiction.

  2. I have Bipolar 1. When diagnosed 20 years ago, I was a non-smoker and cycling was distinctively depressive or manic/hypomanic with episodes of psychosis. I was not an ultradian rapid-cycler until I started smoking but became so and experienced severe mixed episodes with psychosis. My psychiatrist prescribed Chantix and I was able to quit within a week after beginning to take it. Now, 5 years later, I have not had a mixed state, hypomania, or psychosis, though SAD has been a problem. I’m wondering if the association between my degree of cycling and tobacco use had anything to do with even just a small bit of hypoxia or CO2 concentrations? Tobacco is bad enough, but I’m afraid it has so many additives that it will be tough to boil it down to the real culprit(s).
  3. Everything I have read about smoking, including all studies of its impact on CYP 50 all agree on its harmfulness…BUT then, what is psychosis but the inability or the unwillingness to not be obdurate or obstreperous to overwhelming evidence about the wrongness of our perception of reality ?
    Is that not why we ban smoking in all of our public places EXCEPT the casinos in IOWA where psychotic odynophiliacs congregate?
  5. Many of my pts tell me it that it not only helps them to relax, it also helps derease the auditory hallucintions. Many would sacrifrice food before giving up on smoking. Despite my efford to help pt stop smoking, I have rarely succeeded.
  6. Tobacco use is highly prevalent in general population, particulary among minors. To evaluate a possible causal relation, a follow up study of young smokers vs non smokers and psychosis is needed.
  7. i am not sure we can really get a straight answer. the issue on causal relationship between smoking and psychosis is mixaed; first, a patient not yet diagnosed of any psychotic disorder may be smoking to reduce hitherto pschotic symptoms like hallucination, second, a patient with a diagnosed psychotic disorder may be smoking for the same reason or to actually reduce or take care of the side effects of antipsychotics and lastly it may just be a co-occcurence i.e both condition may not be related. therefore approach should be individualized and not generalized.
  8. Most of these pts who have moderate to severe symptoms of psychosis and mania , also have significant amount of anxiety whose source is variable to different patients . These patients seem to trying deal with this anxiety with smoking . We can help these patients with smoking if we can relieve their associated anxiety with these conditions.
  9. I a a psychiatrist and i have conductd study on 500 sudent .i found acorrelation between self reprtd psychiatic symptoms ( I applied general health questionnaire) and somking .
  10. I while in residency had multiple patients that use/used multiple drugs e.i., nicottine, Thc, etc.., and the percentage of the number of mentally ill patients and the use of substances was higher than the literature 50-75%
    but i would always educated my pt’s regarding the health risks despite of running the risk of “distabilizing the patient”. Nicottine is an inducer increased #’s of cigarettes lowers the effect of the antipsychotic but on one of my cases my pt reduced and stopped all smoking then she decompen. positive symptoms florid despite of no issues on nonadherance and multiple atypical plus typical antipsych. We have to continue to educate our pt’s despite the risk of decompensation.

Leave a Reply


Browse By Author



Browse By Author

Sign-up to stay
up-to-date today!


Already registered? Sign In

Case Report

Safety and Tolerability of Concomitant Intranasal Esketamine Treatment With Irreversible, Nonselective MAOIs: A Case Series

Three cases suggest that concomitant use of intranasal esketamine with an irreversible, nonselective MAOI is safe in...