Program Profile: A Stop Smoking in Schools Trial (ASSIST) Program

This in-school smoking prevention program was designed to spread and sustain norms of non-smoking behavior among 12–13 year olds, using influential peer opinion leaders. The program is rated Ineffective. Youths who received the intervention did not differ statistically significantly from youths who did not receive the intervention in their odds of smoking in the last week, at 2 years post-intervention.

An Ineffective rating implies that implementing the program is unlikely to result in the intended outcome(s).

This program's rating is based on evidence that includes at least one high-quality randomized controlled trial.

Program Goals/Target Population

A Stop Smoking in Schools Trial (ASSIST) program aims to reduce adolescent smoking prevalence by using influential peer supporters to spread new behavioral norms related to smoking through informal conversations with peers. The program targets adolescents aged 12 to 13 years.

Program Components

The ASSIST program has five components: 1) nomination of peer supporters, 2) recruitment of peer supporters, 3) training of peer supporters, 4) implementation of intervention, and 5) acknowledgement of peer supporters’ contributions. To identify influential peers who can serve as peer supporters, all students are invited to respond to a questionnaire with the following questions: “Who do you respect in Year 8 at your school?”, “Who are good leaders in sports or other group activities in Year 8 at your school?”, and “Who do you look up to in Year 8 at your school?”  Approximately 20 percent of the most influential students are invited to attend the recruitment meetings, during which nominees learn about the role of peer supporters, get clarification about the program, and agree to attend the training course. Students who smoked should only be peer supporters if they are committed to trying to stop smoking.

Peer-supporter training is conducted over a 2-day period, at a venue away from school premises, and facilitated by a team of external trainers. Trainers use a variety of participatory-learning activities, including role plays, student-led research, small group work, and games to 1) provide information about the short-term risks of smoking and the health, environmental, and economic benefits of not smoking, 2) develop communication skills, including verbal and non-verbal communication, listening, expression of feelings and ideas, group work, team building, cooperation and negotiation, ways of giving and receiving information, and conflict resolution, and 3) enhance students’ personal development, including their confidence and self-esteem, empathy and sensitivity to others, assertiveness, decision-making and prioritizing skills, attitudes toward risk taking, and exploration of personal values.

After the training, peer supporters engage in informal conversations with their peers about smoking during their free time, including to and from school, during breaks, at lunchtime, and after school, over a 10- to 14-week period (or a period equivalent to a school term). Peer supporters keep a log of these conversations. Trainers perform four to six, 1-hour follow-up visits to schools to meet with peer supporters and provide support, trouble shooting, and monitoring of logs. At the end of the intervention period, peer supporters are acknowledged for their contributions with certificates and those who submit their logs also receive additional recognition (e.g., gift vouchers).

Key Personnel

Key program personnel for the ASSIST program include a program organizer, health promotion specialists, and health promotion trainers. The program organizer is generally responsible for setting up and coordinating the program, including identifying suitable trainers, ensuring that the program is delivered in line with the ASSIST program manual, ensuring that quality standards are met by the trainers, and liaising with schools taking part in the program.

Health promotion trainers conduct the training of peer supporters, with oversight from health promotion specialists, and provide ongoing support and supervision.

Program Theory

The ASSIST program is designed to change the norms and behaviors surrounding smoking within a school community by using the potential protective influence of peers. This program has its foundations in diffusion of innovations theory (Rogers 1983), which presumes that using trained peer supporters to communicate smoking-related information during informal contacts with other students will positively influence smoking-related attitudes and behaviors. Additionally, the use of peers and schools as significant social contexts for influencing behavior is consistent with Bronfenbrenner’s (1979) ecological model of human development and Stokol’s (1996) social ecological theory, which emphasize the significance of personal attributes, social contexts, and environmental conditions in influencing individual behavior. The intervention’s goal of addressing smoking uptake in adolescence is also supported by empirical evidence that shows adolescence as a time of high susceptibility to nicotine addiction as well as links between smoking uptake in adolescence and frequency of smoking in adulthood (Campbell et al. 2008).

Study

Campbell and colleagues (2008) studied the effects of the A Stop Smoking in Schools Trial (ASSIST) intervention on students in Year 8 (12–13 years old) in 59 schools in west England and southeast Wales. Since the focus of the intervention was on peer groups within a school-year group, the study used a cluster randomized design; therefore the unit of randomization for the evaluation was at the school level. Of 113 schools that responded positively to the invitation to participate in a randomized controlled trial, 66 schools were selected to participate in the study by random sampling, with stratification by county, type of school, mixed-sex or single-sex, English-speaking or Welsh-speaking, size of school, and proportion of students eligible for free school meals. A final total of 59 schools agreed to participate in the study (30 schools were randomly assigned to the intervention group and 29 schools were randomly assigned to the control group). Schools in the intervention and control groups agreed to continue implementing their existing smoking cessation programming. In addition, those schools in the intervention group would also receive the ASSIST program.

The average number of eligible students in each school was 187. Influential students were identified in both intervention and control schools through student nominations. Only students in the intervention schools were invited to train as peer supporters. A total of 835 of the 5,358 eligible students completed the training and agreed to work as peer supporters. Of those, 687 submitted a completed diary at the end of the intervention period. Peer supporters were generally representative of their year group in terms of gender, ethnicity, smoking status, and intent to remain in full-time education after age 16 (the specific student characteristics were not provided in the study).

At baseline, the intervention schools had 5,358 eligible students and the control schools had 5,372 eligible students. There were no differences between intervention and control schools at baseline in school characteristics (e.g., type of school, size of school, mixed-sex or single-sex). There was, however, a difference between students in the intervention schools and those in the control schools in prevalence of smoking and in socioeconomic status. More students in the control schools reported smoking every week than students in the intervention group (7 percent versus 5 percent). Students in the control group also had significantly lower levels of family-affluence scores than the treatment group (27 percent of the control group had a family-affluence score of 0–2 versus 23 percent of the treatment group) and were significantly less likely to have families that own more than one vehicle (49 percent of the control group had families that owned two or more cars versus 57 percent of the treatment group), suggesting more students in the control group came from less affluent backgrounds. 

Participating youths completed in-class surveys that included questions about their smoking behavior at baseline, immediately after the intervention, and at 1-year and 2-years post-intervention. Students also provided saliva samples at baseline and at 1-year follow up. Saliva samples were also collected at 2-year follow up from students at 24 of the schools (12 intervention schools and 12 control schools). These were used to assess the amount of misreporting among participants; only 1 percent of students who reported not smoking had saliva cotinine concentrations indicative of smoking. 

The primary outcome was likelihood of smoking in the past week. Random effects logistic regression was used to estimate the effects immediately after the intervention and at 1-year and 2-years post-intervention, with school as a random effect. Analyses controlled for smoking behavior among participating students in the school at baseline, as well as for the following school characteristics: independent or state, mixed-sex or single-sex, English- or Welsh-speaking, size of school, and eligibility for free school meals. Analysis used an intent-to-treat model.

ASSIST is available by purchasing a standard 3-year license from DECIPHer IMPACT. The license provides a range of products and services, including the use of all ASSIST materials and methodology, staff training, materials for up to 10 schools per year, quality assurance, seminars and conferences, and helpdesk. <br/><br/> A cost-effectiveness study conducted by Hollingworth and colleagues (2012) estimated that the mean cost of the intervention per school was £5,662 (approximately $7608) or £32 per student (approximately $43).

DECIPHer IMPACT provides interested schools with the necessary training, support, advice, resources, and quality assurance to set up and deliver ASSIST appropriately, as well as ongoing events and updates to the program.

Subgroup Analysis

Campbell and colleagues (2008) found that students in the intervention group had a lower prevalence of smoking than those in the control group at all three follow-up periods (immediately after the intervention, 1-year post-intervention, and 2-years post-intervention). However, only the difference between students in the intervention and control groups at the 1-year follow up was statistically significant; the difference was not statistically significant either immediately after the intervention or at the 2-year follow up.

Campbell and colleagues (2008) also tested for intervention effects among youths identified as at high risk of smoking uptake at baseline (i.e., those who were occasional, experimental, or ex-smokers). There were no statistically significant differences between intervention and control schools in the proportion of students who were at high risk for regular smoking uptake. Findings showed that the intervention did not have a more beneficial effect for this group of students. 

Additional sub-group analyses tested for differential intervention effects by gender, peer supporter status, deprivation (whether the proportion of students in the school eligible for free school meals was above or below the median), and by school location (whether or not the school was located in a community in the south Wales valleys). Statistically significant differences in intervention effects were detected by school location, but not by gender, peer-supporter status, or deprivation. The intervention had a more beneficial effect for students in intervention schools located in the south Wales valleys compared with students in intervention schools located in towns or cities.

These sources were used in the development of the program profile:

These sources were used in the development of the program profile:

Audrey, Suzanne, Kathleen Cordall, Laurence Moore, David Cohen, and Rona Campbell. 2004. “The Development and Implementation of a Peer-Led Intervention to Prevent Smoking Among Secondary School Students Using Their Established Social Networks.” Health Education Journal 63: 266–84.

http://doi.org/10.1177/001789690406300307

Bronfenbrenner, Urie. 1979. The Ecology of Human Development. Cambridge, Mass.: Harvard University Press.

DECIPHer IMPACT website. No date.

Hollingworth, William, David Cohen, James Hawkins, Rachael Hughes, Laurence Moore, Jo Holliday, Suzanne Audrey, Fenella Starkey, and Rona Campbell. 2012. “Reducing Smoking in Adolescents: Cost-Effectiveness Results From the Cluster Randomized ASSIST (A Stop Smoking In Schools Trial).” Nicotine and Tobacco Research 14(2):161–68.

https://www.ncbi.nlm.nih.gov/pubmed/22180581

Rogers, Everett. 1983. Diffusion of Innovations. 3rd ed. New York, N.Y.: The Free Press.

Starkey, Fenella, Laurence Moore, Rona Campbell, Mark Sidaway, and Michael Bloor. 2005. “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.” BMC Public Health 5:43–52.

http://www.biomedcentral.com/1471-2458/5/43

Stokols, Daniel. 1996. “Translating Social Ecological Theory into Guidelines for Community Health Promotion.” American Journal of Health Promotion 10(4): 282–98.