Long term smoking and quitting among people with severe mental illness: 3-year follow-up of the SCIMITAR+ Trial

Abstract Background People with severe mental illnesses (SMIs) are three times more likely to smoke than the wider population, contributing to widening health inequalities. Here we report the long term 3 year outcomes for the SCIMITAR+ trial ( ISRCTN72955454 ), which compared usual care to a bespoke smoking cessation package. Methods We recruited 526 heavy smokers with bipolar illness or schizophrenia who were randomly allocated to a bespoke smoking cessation intervention (n=265) or to usual care (n=261) between October 2015 and December 2016. We measured long-term quitting by carbon monoxide-verified smoking status at 3 years post randomisation, and a range of secondary outcomes. Results 261 of the original 526 were followed up. 209 participants (80.1% of followed-up; 39.7% of randomised) could be defined as a smoker or non-smoker for the 36-month primary analysis. Forty-three participants (16.5% of 261) were determined to be a non-smoker via self-reporting and CO confirmation; 21 in the intervention arm (16.3% of 129; 19.6% of those who provided both self-reported and CO measure) and 22 in the control arm (16.7% of 132; 21.6% of those who provided both self-reported and CO measure) – adjusted OR of 0.89 (95% CI 0.45, 1.77, p = 0.74). For secondary outcomes, there was no sustained between group difference in reduction in nicotine dependence (measured using the Fagerstrom Test for Nicotine Dependence) or motivation to quit. Some short-term improvements in physical health (measured by the Short Form 12) were present at 36 months. Conclusions A bespoke intervention represents the model for care for mental health services in the UK, but long term quit rates cannot be assumed. Sustained attention to smoking and relapse is likely to be needed to ensure that short term gains are maintained. The certainty of these results is tempered by loss to follow up and low statistical power. Funding This study was funded by NIHR Health Technology Assessment Programme (Project number 11/136/52) Plain English Summary Smoking rates are very high amongst people who use mental health services. This makes a significant contribution to health inequalities and reduced life expectancy. Earlier research has shown that smoking cessation services are effective when the specific needs of people with mental ill health are taken into account. This forms the basis of guidelines issued by the National Institute for Health and Care Excellence. In this study we looked at the longer-term impact of an intervention to help people to quit smoking. At three years under half of the people who took part in the original study were able to provide information on their use of tobacco. The rates of smoking were similar between people who received the intervention three years ago, and those who received only usual care. Since we already know that the intervention is effective in helping people to quit in the short term, it would be prudent to ensure that people continue to be offered care to help them to quit if they are at risk of relapse. The longer-term impact of a brief intervention on longer term quit rates cannot be assumed, since the strength of this conclusion is limited because we were only able to follow up under half of the people who took part in the study. Further research is needed into the optimum way in which the health benefits of quitting can be sustained in the longer term.