Rationale, design and conduct of a comprehensive evaluation of a school-based peer-led anti-smoking intervention in the UK: the ASSIST cluster randomised trial [ISRCTN55572965]

BACKGROUND: To date, no school-based intervention has been proven to be effective in preventing adolescent smoking, despite continuing concern about smoking levels amongst young people in the United Kingdom. Although formal teacher-led smoking prevention interventions are considered unlikely to be effective, peer-led approaches to reducing smoking have been proposed as potentially valuable. METHODS/DESIGN: ASSIST (A Stop Smoking in Schools Trial) is a comprehensive, large-scale evaluation to rigorously test whether peer supporters in Year 8 (age 11–12) can be recruited and trained to effect a reduction in smoking uptake among their fellow students. The evaluation is employing a cluster randomised controlled trial (RCT) design with secondary school as the unit of randomisation, and is being undertaken in 59 schools in South East Wales and the West of England. Embedded within the trial are an economic evaluation of the intervention costs, a process evaluation to provide detailed information on how the intervention was delivered and received, and an analysis of social networks to consider whether such a peer group intervention could work amongst schoolchildren in this age group. Schools were randomised to either continue with normal smoking education (n = 29 schools, 5562 students), or to do so and additionally receive the ASSIST intervention (n = 30 schools, 5481 students). No schools withdrew once the trial had started, and the intervention was successfully implemented in all 30 schools, with excellent participation rates from the peer supporters. The primary outcome is regular (weekly) smoking, validated by salivary cotinine, and this outcome has been obtained for 94.4%, 91.0% and 95.6% of eligible students at baseline, immediate post-intervention, and one-year follow-up respectively. DISCUSSION: Comprehensive evaluations of complex public health interventions of this scale and nature are rare in the United Kingdom. This paper demonstrates the feasibility of conducting cluster RCTs of complex public health interventions in schools, and how the rigour of such designs can be maximised both by thorough implementation of the protocol and by broadening the scope of questions addressed in the trial by including additional evaluative components.