Prevention of adolescent smoking: A prospective test of three models of intervention

Abstract

Objective: The onset of smoking in adolescence leads to significant health problems in later life and so adolescent smoking prevention is a crucial concern of health psychology. Yet the evidence on smoking prevention in adolescence is not encouraging. The objective of this study was to examine the relative long‐term efficacy of three specifically focussed prevention strategies (health‐oriented, fitness‐oriented and social skills/stress management‐oriented) directed at the onset of adolescent smoking.

Design: A longitudinal intervention study.

Method: A large sample of adolescents aged 11–17 years was assessed for smoking behaviour and then assigned through group randomization to one of the three intervention programmes listed above. Following intervention, based on four standard classroom sessions, smoking behaviour was then reassessed immediately after intervention. Final follow‐up of smoking behaviour at 12 months after intake yielded completed data for n = 1,694 (62.3%) of the intake cohort. Data on smoking behaviour were then compared both across intervention strategies and with another large “control” cohort of adolescents who had been identically assessed in a previous study by DGB.

Results: Application of a health‐oriented strategy was significantly better than the other two strategies in controlling smoking behaviour immediately following intervention. The intervention strategy emphasizing social skills and stress management in the face of peer pressure to smoke was, however, superior to either the health‐ or fitness‐oriented strategies, or to the “control” group in controlling smoking behaviour at 12‐month follow‐up.

Conclusions: While the health message cannot be dismissed as a focus for adolescent smoking prevention, a strategy that assists young people to resist the effects of peer pressure through social skills and stress management seems to provide the most enduring means of controlling smoking behaviour in adolescence.

• Adolescence
• smoking
• prevention
• control

Acknowledgements

The authors thank the schools and their staff and students who agreed to participate in this study. The authors are also grateful to two members of an earlier research team, Mrs Gillian Philpott and Mrs Monika Reinhart, who were closely involved in the development and implementation of the smoking‐prevention strategies and in the collection of the data. Their highly professional efforts were invaluable in the construction of the intervention strategies and the conduct of the implementation. The study was funded by a grant to DGB by the (then) Smoking and Health Research Foundation of Australia.

Notes

1. Canberra is the national capital of Australia. Schools were selected to reflect socio‐economic diversity across the city, although it is noted for its relative socio‐economic uniformity (Australian Bureau of Statistics, 1989).

2. Negotiation with participating schools and with the Ethics Committee of the Education Department determined that active programmes of intervention must be offered to all participating students volunteering for the study. It was not possible, therefore, to collect control data as part of the intervention study itself.

3. Smoking data were collected by self‐report. While it has been suggested that this procedure lacks validity, recent evidence indicates that self‐reported smoking behaviour in adolescents approaches 100% concordance with expired air carbon monoxide levels (Becona & Vasquez, 1998; Prokhorov, Murray & Whitbeck, 1993). A recent review of 28 studies comparing self‐reported smoking behaviour to a range of biological measures of smoking among adolescents concluded that self‐report provided a completely valid measure of adolescent smoking (Dolcini, Adler, & Ginsberg, 1996).

4. In fact, as will be evident from Table III, this latter group was quite small relative to the former group.