Smoking behaviour is a high-risk factor for mortality, not only for people with psychosis but also for the general population. Certainly, that’s not new to the research community. De Leon and Diaz (2005) found that people with schizophrenia were three times more likely to smoke compared to the general population, while Olfson et al (2015) indicated a higher risk of premature mortality due to chronic obstructive pulmonary disease.

Evidence-based studies indicate that cognitive functioning is gradually decreasing in people with psychosis (Fioravanti et al., 2005). But how is this associated with smoking? Well, although nicotine may have positive effects on peoples’ cognitive functioning (e.g. enhanced attention), it doesn’t apply to those with schizophrenia. Often, patients have a dysregulated nicotinergic system, which is implicated in cognitive functioning (Parikh et al., 2016).

Studies focusing on the long-term effects of nicotine on cognitive functioning indicated that elderly and middle-aged smokers were at higher risk of cognitive decline compared to non-smokers (Anstey et al., 2007; Sabia et al., 2008). Interestingly, the risk of poor cognition tends to be decreased among people who had quit smoking when compared to smokers. Does the same theory apply to people with psychosis? Nevertheless, the results on long-term smoking and cognition on patients are contradictory, while longitudinal evidence of an association between smoking and specific cognitive functions is lacking.

So what did Vermeulen et al. (2018) aim to explore in this study?

• If peoples’ current smoking behaviour is associated with their performance on specific cognitive tasks.
• If a longitudinal association between smoking behaviour and cognitive functioning exists when comparing people with non-affective psychosis, their unaffected siblings and healthy people.
• If smoking cessation can improve peoples’ cognitive functioning.

Methods

The study was based on the Naturalistic Multicentre Genetic Risk and Outcome of Psychosis (GROUP) cohort. The sample consisted of 1,119 patients with a diagnosis of non-affective psychosis, 1,059 unaffected siblings and 586 controls. Their recruitment was conducted by clinicians in the premises of four University sites and their collaborating mental health care facilities, located in the Netherlands and Belgium. Following baseline, all participants were invited for two follow-ups: at three and six years.

A variety of measures has been used to assess smoking behaviour and cognitive functioning, including a shortened version of the Composite International Diagnostic Interview (CIDI), the MATRICS Consensus Cognitive Battery Test, subtests of the WAIS-III, the Auditory Verbal Learning Test and the Continuous Performance Test. Positive smoking status was given to people who were smoking daily during a month or more.

Differences in demographics and clinical outcomes among smokers and non-smokers in all groups were explored through t-tests, chi-square tests, and one-way ANOVA. The association between smoking and cognitive functioning was investigated through linear mixed-effects models. Initially, the models were applied to the raw scores and raw change scores for each cognitive domain. In order to reduce the likelihood of type I errors and counteract the problem of multiple comparisons, the Bonferroni correction method was used. The mixed model was adjusted by fixed effects, such as smoking status during the last year, time and a set of a priori covariates (including age, gender, years of education, cannabis use, antipsychotic medication use, and the severity of psychopathology) and by random effects, such as intercepts for subjects and random slopes for the effect of time. Additionally, the researchers ran post-hoc linear mixed-model analyses for every significant difference that was found between smokers and non-smokers in relation to a specific task. Finally, the association between smoking severity and cognitive functioning was assessed by comparing smoking status at baseline to 3-years follow-up and smoking status at 3-years follow-up compared to 6-years follow-up. The model was applied to the raw scores for each cognitive domain and change scores for the number of cigarettes smoked per day. The model-building process followed the same parameters as described above.

Results

Out of 1,904 patients with non-affective psychosis, 65.5% were diagnosed with schizophrenia. At baseline, 66.6% of patients were smoking, compared to 38.3% siblings and 25.2% controls, while the number of smoked cigarettes per day was also significantly higher.

Smoking behaviour and cognitive performance

• Among patients, the results suggested an association between smoking and lower performance on processing speed, while a similar association was also indicated for the number of cigarettes per day.
• Among unaffected siblings, the findings indicated negative associations between smoking status and working memory, reasoning and problem-solving. Post-hoc analyses also showed negative associations between the number of cigarettes per day and perceptual organisation.
• Finally, for the control group, results showed a significant association between smoking behaviour and number of cigarettes smoked per day and lower performance on processing speed.

Smoking cessation and cognitive recovery

Overall, all subgroups demonstrated a significantly greater performance on the majority of the cognitive tasks during follow-up periods. Although smoking behaviour remained the same in the majority between follow-up periods, gradually more individuals quit smoking. Results indicated associations between smoking cessation and better performance on processing speed among patients who quit smoking. Additionally, post-hoc analyses indicated a significant association between changes in the number of cigarettes per day and improvement in processing speed among patients.

Discussion

Key points to remember:

• Smoking was associated with lower performance in many cognitive functions in all groups.
• A dose-response relationship was found between the number of cigarettes per day and processing speed, reasoning and problem-solving.
• Smoking cessation was significantly associated with better processing speed in patients.

Strengths and limitations

• A number of methodological strengths can be identified, such as the large sample size, comparison groups, and long-term nature of the study.

but…

• The observational design has its own disadvantages, as we cannot rule out the reverse causation.
• Besides, the study focused on spontaneous changes in smoking status, so the sample of people who quit smoking might not be representative.
• Furthermore, we lack information on the smoking history and blood levels of cotinine or carbon monoxide, which might have an effect.

Implications for research and policy

Current evidence is in-line with the literature and enhances our understanding regarding the relationship between smoking and cognitive functioning; including people with psychosis, family members and healthy subjects. While we gained a deeper understanding regarding the differences between smokers and non-smokers irrespectively their mental health, it is important to further explore the relationship between the exposure of smoking and cognitive performance. Do family history, years of smoking or passive smoking have a similar effect?

Evidence-based psychosocial interventions could be developed and delivered to people with psychosis on the grounds that smoking is associated with cognitive decline. Both individual and family psychoeducation will be beneficial to understand the short-term and long-term negative effects of smoking on cognitive performance. Furthermore, the potential effectiveness of smoking cessation treatment is highlighted. People with psychosis will not only be able to ameliorate their cognition, but also their mental and physical health outcomes.

Conflicts of interest

None to declare.

Links

Primary paper

Vermeulen J, Schirmbeck F, Blankers M, van Tricht M, Bruggeman R, & van den Brink W. et al. (2018). Association Between Smoking Behavior and Cognitive Functioning in Patients With Psychosis, Siblings, and Healthy Control Subjects: Results From a Prospective 6-Year Follow-Up Study. American Journal Of Psychiatry, 175(11), 1121-1128. https://doi.org/10.1176/appi.ajp.2018.18010069

Other references

de Leon J and Diaz FJ (2005). A meta-analysis of worldwide studies demonstrates an association between schizophrenia and tobacco smoking behaviors. Schizophr Res 2005; 76:135–157 [PubMed abstract]

Olfson M, Gerhard T, Huang C. et al (2015). Premature mortality among adults with schizophrenia in the United States. JAMA Psychiatry 2015; 72:1172–1181

Fioravanti M, Carlone O,Vitale B. et al (2005). A meta-analysis of cognitive deficits in adults with a diagnosis of schizophrenia. Neuropsychol Rev 2005; 15:73–95 [Abstract]

Parikh V, Kutlu MG and Gould TJ (2016). nAChR dysfunction as a common substrate for schizophrenia and comorbid nicotine addiction: current trends and perspectives. Schizophr Res 2016; 171:1–15

Anstey KJ,von Sanden C, Salim A. et al (2007). Smoking as a risk factor for dementia and cognitive decline: a meta-analysis of prospective studies. Am J Epidemiol 2007; 166:367–378 [PubMed abstract]

Sabia S, Marmot M, Dufouil C. et al (2008). Smoking history and cognitive function in middle age from the Whitehall II study. Arch Intern Med 2008; 168:1165–1173

Photo credits

• Photo by Diego López on Unsplash
• Photo by Hosein Emrani on Unsplash
• Morgan CC BY 2.0
• Thanks for 1.9 million views CC BY 2.0