Study aim and methods: We set out to investigate the impact of smoking cessation on mental health in general and psychiatric populations. We included longitudinal studies of adults from any population and reported in any language, we also sought out additional data where means and SDs were not presented in publications. We searched five databases and our search strategy resulted in over 10,000 records; we narrowed this down to about 250 full text articles, 26 of which were eligible for meta-analysis. We used a generic inverse variance meta-analysis to compare the difference in mean change in mental health outcomes between smokers who stopped and smokers who continued smoking (figure 1).
Study findings and implications: The included studies examined change in mental health outcomes: anxiety, depression, mixed anxiety and depression, stress, psychological quality of life and stress. On average participants were aged 44, smoked about 20 cigarettes a day, and were followed up for six months. We included studies from the general population, people with mental health disorders, people with physical disability and pregnant women. The results showed that smoking cessation was associated with significant improvements in symptoms of depression and anxiety, and significant increases in positive affect and psychological quality of life. The association was similar between general and psychiatric populations and there was no evidence that these findings were the results of confounding factors (we conducted over 10 sensitivity analyses). To interpret these findings we compared the size of the association, known as the effect estimate, with that of anti-depressant treatment for depression and anxiety disorders; we found that the effect of stopping smoking was equal to or greater than the effect of anti-depressant treatment for mood disorders.
The data in this study were observational and therefore cannot prove causality. One could argue that it is an improvement in mental health which causes people to stop. We argued against this hypothesis as the majority of the studies included were secondary analyses of data derived from trials in which everyone attempted to stop smoking; therefore, whatever happened to their mental health occurred after they stopped smoking. Secondly, the review included one randomised trial, from which cause-effect associations can be derived. In this randomised trial there were modest improvements in anxiety and depression in the stop-smoking arm, while the continue-smoking arm experienced very little change.
Many mental health patients report that smoking helps them with their stress and psychological disturbances; however smoking may actually worsen some of these symptoms. It is a common practice to postpone addressing patients’ smoking behaviour until after their mental health improves. On the other hand we know for many mental health patients their problems will likely be ongoing, so when is the right time? As the evidence suggests, there is no confirmation of harm from cessation; rather, the evidence suggests cessation may be therapeutic.
The publication process: This paper was submitted for publication in the BMJ in October 2012. We had major changes and were instructed to revise and resubmit. We re-analysed the data accordingly, including all sensitivity analyses. The paper was resubmitted and accepted in January 2013, ready for publication in February. This was my first ever publication, so I was very pleased and excited to have a publication in the BMJ. The results from the study led to significant media attention, including interviews with BBC and coverage in the daily telegraph and other major newspapers in the UK and internationally. The team were also asked to submit a video abstract for the BMJ. This is available on the BMJ website alongside the publication, available at http://dx.doi.org/10.1136/bmj.g1151. The article is amongst the highest ever scored in the BMJ (ranked #42 of 18,782), and was awarded ‘best research paper of the year’ for 2014. I was awarded a certificate and a prize from the BMJ in December 2014, at the BMA house in London. I was very honoured and proud that all the hard work paid off.
Figure 1. Meta-analysis outcome: the difference in mean change in mental health outcomes between smokers who stopped, and smokers who continued smoking.
Written by Gemma Taylor
