Smoking prevention in young people: A cluster randomised controlled trial of implementation intentions

This research will test an intervention to reduce smoking initiation in adolescents initially aged 11-12 years over a 4 year time period. The intervention will employ a simple but effective strategy to refuse offers of cigarettes (if-then plans about how to say no to offers of cigarettes plus reading anti-smoking messages) and test it against a control condition (forming if-then plans about completing homework plus reading pro-homework messages). A total of 18 schools (with approximately 3 classes in each) will be randomly selected to implement the intervention and 18 other schools (again with approximately 3 classes in each) will be randomly selected to implement the control condition. The intervention will be implemented by trained teachers and be conducted in schools in classroom time on nine occasions (ie. six months between each intervention). The main outcome measure will be smoking behavior assessed by objective measures (breath carbon monoxide measures) at baseline (when participants were 11-12 years of age) and at 12, 24, 36 and 48 months post baseline (ie., final follow-up when participants were 15-16 years of age). A secondary outcome measure will be self-reported smoking measured at the same time points. Analysis will test for differences in smoking rates at each of the follow-ups after controlling for any differences at baseline.

Reducing smoking initiation in adolescents is an important way to tackle smoking-related harm. This research will test the effectiveness of a simple intervention shown previously to reduce smoking initiation rates (forming repeated implementation intentions in relation to how to refuse offers of a cigarette plus reading anti-smoking messages) in a Phase III cluster randomised controlled trial with a large sample of adolescents (N = 4320). A total of 36 schools will be randomised to intervention or control conditions. Those in the intervention condition will complete implementation intentions in relation to refusing offers of cigarettes on nine occasions (each separated by six month intervals) between the ages of 11-12 and 15-16 years. Those in the control condition will complete implementation intentions in relation to completing homework at the same time points. Adolescents in both conditions will also read information designed to increase motivation to either not initiate smoking or complete homework just before completing each implementation intention. The intervention will be implemented by trained teachers. Objectively assessed smoking (by breath carbon monoxide monitor), self-reported smoking and predictors of smoking will be assessed at baseline (at age 11-12 years) and at 12, 24, 36 and 48 months post-baseline. Objective smoking at 48 months post-baseline when the participants are aged 15-16 years will be the key outcome measure (self-reported smoking at the same timepoint will be a secondary outcome). The study is powered to detect a difference in smoking rates between the intervention (9% smoking) and control (14% smoking) groups of approximately 5%. We will also conduct a cost-effectiveness analysis. These data will be used to inform the potential roll out of the intervention to the larger population.

There are various beneficiaries to the proposed research. The research will generate theoretical and practical knowledge about the effectiveness of a smoking prevention intervention in young people. This knowledge will be useful to those interested in health behaviour change and prevention of smoking initiation in particular. We will make this knowledge available through publications in high quality, high impact journals such as the BMJ by the end of the project. Such knowlege will help inform the development of health behaviour change interventions based on implementation intentions or related techniques with other behaviours and other populations.More specifically the research will benefit those wishing to reduce smoking initiation rates in adolescents. Should this effectiveness trial prove successful then it can inform the roll out of the developed and tested intervention to a large proportion of adolescents across the UK. Given the present trial will develop and test materials for use by teachers to perform this intervention it will be possible to develop a toolkit including all materials necessary to the implementation of this intervention (available on a website). This will include a brief theoretical rationale for the intervention, step-by-step description of all procedures to be followed, all classroom material to administer the intervention, and a help guide including frequently asked questions.Widespread use of the intervention in schools across the UK could be expected to contribute significantly to reducing smoking initiation rates in adolescents. Such a reduction in smoking initiation rates will contribute to reductions in smoking-related harm and so reduce the healthcare costs associated with smoking. We will seek to ensure the findings of the research reach those wishing to reduce adolescent smoking in a number of ways. First, we plan to ensure our findings are published in user journals within the lifetime of the project. Second, we will develop and publicise a website reporting the findings and containing the intervention materials and instructions on how to use them before the end of the project. Third, we will host a stakeholder workshop in Leeds towards the end of the project in order to explore the best means to roll out the intervention in schools across England. The stakeholders invited will include those involved in tobacco control and the Department of Health and we will liaise with our collaborators in tobacco control to ensure that all appropriate stakeholders are invited to this workshop.