Both in the UK and internationally, tobacco smoking continues to be an important cause of morbidity and mortality. For example, smoking related illnesses are estimated to kill over 100,000 people in the UK each year [1, 2], usually later in life. Yet smoking is a behaviour that is, in general, taken up between the ages of 10 and 20 years. A variety of different research studies have supported the idea that the vast majority of smokers take up this habit as adolescents [3–5] with 40% of adult smokers having started before they reached 16 years of age . The General Household Survey  reported that 38% of adult regular smokers took up the habit before the age of 15 years. This appears to be the case despite the fact that health promotion messages have ensured that awareness of the health consequences of smoking is now widespread, even among the young. In the UK, while the rates of regular smoking at 11 years of age are only 0.5%, this rapidly rises to 15% by 15 years of age, and then more gradually to around 20% among young adults . The recent UK Department of Health  plan is to reduce the rates of regular smoking in 15 year olds to 12% or less by the end of 2015. Two important ways to tackle smoking-related harm are interventions to help individuals (usually adults) quit smoking and interventions to help individuals (usually adolescents) not to initiate smoking. The present paper reports the protocol for an intervention designed to reduce smoking initiation in adolescents as potentially the most effective way to reduce smoking-related harm.
A worthwhile intervention to reduce smoking initiation in the groups most likely to take up this habit (i.e., adolescents) would need to have at least three important characteristics: strong effects on reducing initiation; wide reach; and low cost. Promising initial data have been collected  (i.e., an explanatory trial) on the efficacy of one intervention technique that may have all of these characteristics. The study outlined here would aim to collect further data on the outcome effectiveness and cost-effectiveness of this technique in the form of a cluster randomised controlled trial (i.e., a pragmatic trial). Given the nature of the existing data in this area, this has been cast as a pragmatic trial  that could inform the potential roll out of this intervention for widespread use. The intervention in question is the formation of repeated implementation intentions about how to refuse offers of cigarettes. Implementation intentions are simple ‘if-then’ plans  about how to respond to environmental cues in order to help achieve a goal such as not taking up smoking, e.g., If offered a cigarette I will say ‘no thanks, I don’t smoke’. Such plans can be formed before the opportunity to act presents itself and have been found to be an effective means to change a range of behaviours . Our research has shown that the repeated formation of implementation intentions about how to refuse offers of cigarettes can have a strong effect on reducing smoking initiation rates among adolescents [8, 12]. In addition, this is a simple intervention that could be deployed across most schools (i.e., has wide reach) in order to help tackle smoking initiation in adolescent groups. Finally, this intervention is relatively low cost, requiring around 30 minutes per session to implement (including anti-smoking messages and completing an implementation intention questionnaire). Such an intervention could offer ‘value for money’ and be rolled out to significant proportions of the adolescent population in the UK.
Adolescence is the period during which the vast majority of smokers take up this habit in the UK [1, 3]. As such it represents a key period during which to intervene. A considerable number of interventions have been tested in this age group. The vast majority of such interventions have been school-based but have met with only mixed evidence of effectiveness (for reviews see [13, 14]). In part this may be attributable to lack of thorough evaluation. Nevertheless, even among high quality randomised controlled trials of school-based interventions the evidence is mixed. For example, information-based interventions have generally been ineffective, while social influence interventions have tended to show rather mixed effectiveness . The most highly regarded and longest trial of this type of intervention was the Hutchinson Smoking Prevention Project which reported no evidence for effectiveness over a period of 8 years . The ASSIST project is another well conducted trial carried out in the UK to evaluate a peer-led intervention to reduce smoking initiation . This RCT did report support for a peer-led intervention over both a one- and a two-year period. More mixed evidence has been reported for interventions that target social competence or test multi-modal interventions , although there are far fewer such studies in this area. A further problem has been that many of these interventions have been quite intensive and costly to implement. For example, the Hutchinson Smoking Prevention Project  included a total of 65 lessons provided over a period of 8 years. Similarly, the ASSIST intervention involved two-day training events for the peer supporters. Such interventions may be costly and difficult to scale up into population level interventions that could be conducted in all schools across the UK at reasonable cost and with relative ease. In contrast, an intervention involving forming a repeated implementation intention about how to refuse offers of cigarettes would appear to be efficacious in reducing smoking initiation rates [8, 12] and is also simple, easy to administer, cheap, and would appear to have the potential to be readily scalable up into a population-level intervention delivered in the vast majority of UK classrooms.
Gollwitzer  has defined an implementation intention as a plan of how, where and when to commit a behaviour (see  for a review). This type of plan establishes an ‘if-then’ link between a situation and a planned behaviour (e.g., between the offer of a first cigarette and a refusal strategy). Implementation intentions have proved to be effective yet simple means of changing a range of different health behaviours . Gollwitzer and Sheeran’s  meta-analysis showed that across 94 independent studies in both laboratory and field settings, implementation intentions had an average effect size of d+ = 0.65, although only 6 out of the 94 studies reviewed by Gollwitzer and Sheeran  investigated health-risk behaviours and none examined smoking. More recently, research has begun to emerge to suggest that implementation intentions may be effective in promoting smoking cessation. For example, in two field experiments, Armitage [18, 19] found that implementation intentions caused significantly more smokers to quit (up to 19% quit) compared with smokers randomly allocated to active control conditions (2% quit), suggesting that the technique has utility in this domain.
Implementation intentions have also been used in relation to reducing smoking initiation. Two studies have assessed the impact of forming an implementation intention about what to say to refuse the offer of a cigarette on subsequent smoking initiation. Higgins & Conner  reported that among adolescents who formed such an implementation intention (on a single occasion) 0% went on to initiate smoking in the next two months, whereas 6% did so in a control condition. Conner & Higgins  more recently reported the results of an explanatory cluster randomised controlled trial of repeated implementation intentions in a more appropriately powered study. Classes of children were randomly allocated to complete implementation intentions about how to refuse offers of cigarettes (intervention) or complete their homework (control) on 7 occasions between the ages of 11–12 and 13–14 years (both groups also read simple anti-smoking messages on each occasion). Research assistants delivered the intervention in classroom time to whole classrooms of children in approximately 30 minute sessions. After controlling for various known predictors of smoking initiation (e.g., gender, attitudes to smoking, friends and family smoking) and the multi-level nature of the data, rates of self-reported and objectively assessed smoking at age 15–16 years were significantly lower in the intervention compared to the control condition (d+ = 0.24 and 1.04 for self-reported smoking and objectively assessed smoking respectively). For self-reported smoking, the unadjusted reduction in smoking initiation was approximately 7% between the intervention and control conditions, while the unadjusted difference for objectively assessed smoking was approximately 10%.
The following research questions will be addressed in this trial:
Can repeated implementation intentions related to refusing offers of cigarettes reduce smoking initiation rates in 11–16 year olds relative to a control group of adolescents?
What is the cost effectiveness of such an intervention?