The single largest contributor to the 15- to 20-year life expectancy gap between people with schizophrenia and the general population is not suicide. It is not the side effects of medication. It is smoking. Roughly 60 to 80 percent of people with schizophrenia smoke, compared with around 12 percent of the general adult US population, according to data from the CDC and the SAMHSA National Survey on Drug Use and Health (samhsa.gov). Most of those who smoke are heavy smokers. Most have tried to quit. Most have been told, implicitly or explicitly, that quitting is too much to ask of them.
Quitting smoking is one of the most impactful health changes a person with schizophrenia can make — and modern medications and behavioural support can roughly triple the odds of success.
Why are rates so high?
Several overlapping explanations have evidence behind them:
- Self-medication of cognition and negative symptoms. Nicotine acutely improves attention, working memory, and processing speed. People with schizophrenia often describe feeling more "switched on" after smoking. Brain-imaging work suggests nicotinic receptor systems are altered in schizophrenia, which may make smoking subjectively more rewarding.
- Social and environmental exposure. Inpatient psychiatric units historically permitted smoking, smoking remains common in supported housing, and many social networks of people with severe mental illness include high rates of smoking.
- Shared genetic vulnerability. Some genetic variants (CHRNA5, for example) appear linked to both nicotine dependence and risk for psychotic illness.
- Lower priority in clinical care. Clinicians have often viewed smoking as a coping resource not to be disturbed during fragile recovery — a view increasingly challenged by the evidence.
The health cost
Smoking is the dominant driver of cardiovascular disease, COPD, and lung cancer in people with schizophrenia. It compounds antipsychotic-related metabolic risk. It also has a direct interaction with antipsychotic treatment: tobacco smoke (not nicotine itself) induces the liver enzyme CYP1A2, which metabolises clozapine and olanzapine. A heavy smoker may need a substantially higher dose of clozapine to achieve the same blood level as a non-smoker — and quitting suddenly without telling the prescriber can cause clozapine levels to rise into toxic ranges.
If you take clozapine or olanzapine, tell your psychiatrist before you stop smoking. They can monitor blood levels and adjust the dose downward as needed. Quitting is still strongly recommended — it just needs to be coordinated.
Does quitting destabilise psychosis?
This was the long-standing fear, and it has largely not held up. The EAGLES trial (Anthenelli et al., The Lancet, 2016) — the largest randomised trial of smoking cessation pharmacotherapies in people with and without psychiatric disorders — found that varenicline and bupropion were both significantly more effective than nicotine patch or placebo, and that neither produced a meaningful increase in neuropsychiatric adverse events compared with placebo, including in the subgroup with serious mental illness. Earlier black-box warnings on varenicline have since been removed by the FDA.
What works
1. Pharmacotherapy
Three medications have strong evidence in this population:
- Varenicline (Chantix) — the most effective single agent in EAGLES. It partially activates the same nicotinic receptors that nicotine binds to, reducing both cravings and the reward of smoking.
- Nicotine replacement therapy (NRT) — patches plus a short-acting form (gum, lozenge, inhaler, or spray) is the standard. Using two forms together is consistently more effective than one.
- Bupropion — an antidepressant with cessation indication. It can be combined with NRT. It does lower the seizure threshold modestly, which matters for some patients (especially those on clozapine).
All three roughly double or triple quit rates compared to no medication. Combinations (varenicline plus NRT, or bupropion plus NRT) often outperform single agents.
2. Behavioural support
Pharmacotherapy alone produces fewer quits than pharmacotherapy plus structured behavioural support. The strongest evidence is for multi-session counselling delivered by a trained smoking-cessation specialist. National quit lines (1-800-QUIT-NOW in the US) and free apps (smokefree.gov) are reasonable starting points if dedicated counselling is not accessible.
3. Vaping — a complicated topic
The Cochrane review on e-cigarettes for smoking cessation (Hartmann-Boyce et al., updated regularly) finds moderate-certainty evidence that nicotine e-cigarettes help more people quit than NRT. They are clearly less harmful than smoking, but they are not harmless, and long-term effects are not known. For someone who has failed multiple attempts on standard treatments, switching to a vape can be a reasonable harm-reduction step. It is best done with input from a clinician who knows your full medication list.
Why quit attempts fail — and what to do differently
- Trying to quit "cold turkey" with no medication. This is the lowest-success approach for any smoker, and lower still in this population.
- Underdosing NRT. Heavy smokers often need higher-strength patches, sometimes combined with short-acting NRT, to actually control cravings.
- Quitting during an acute episode. Stability first; cessation works better when other variables are controlled.
- Going alone. Quit rates rise with social support, even informal — a friend, group, or app community.
- Stopping too early. Combination treatment for at least 12 weeks (often longer) is the standard.
What clinicians should do (and what to ask for)
Best practice — recommended by NICE, the American Psychiatric Association, and the Surgeon General — is that every clinical contact include a brief discussion of smoking, and that pharmacotherapy and counselling be offered routinely. If your clinician does not raise it, you can ask: "Can we talk about a quit attempt with medication support?"
The bigger picture
For decades, smoking cessation was treated as an extra burden to be deferred indefinitely in people with schizophrenia. The data no longer support that view. Quitting is the single highest-impact change for long-term health in this population, and modern treatments make it much more feasible than the older literature implied. It does not have to be done alone, in one attempt, or without medical support.
This article is for educational purposes only and is not medical advice, diagnosis, or treatment. Always consult a qualified mental health professional. If you or someone you know is in crisis, call or text 988 in the US, or your local emergency number.