Can population-based tobacco-control policies change smoking behaviors of adolescents from all socio-economic groups? Findings from Australia: 1987-2005

OBJECTIVE: To examine whether socio-economic status (SES) was associated with changes in smoking prevalence among Australian adolescents during three phases of tobacco-control activity between 1987 and 2005. METHODS: Triennial cross-sectional national studies of representative random samples of secondary students aged 12-17 years have been conducted since 1987. Numbers range from 19,203 in 1987 to 29,853 in 1996. Self-report anonymous surveys assessed cigarette use in the past month, week (current), and on at least three of the previous seven days (committed). Students' residential postcode was collected and the Index of Relative Socio-Economic Disadvantage (IRSD) associated with each postcode determined SES quartiles. RESULTS: Between 1987 and 2005, smoking prevalence decreased in all SES groups. Tobacco-control activity level was associated with changes in smoking prevalence and whether changes were consistent across SES groups. In a period of low tobacco-control funding (1992-1996) and activity, smoking prevalence increased among 12- to 15-year-olds, the increase being greatest among low SES students. In a period of high tobacco-control activity (1997-2005) smoking decreased and reductions were generally consistent across SES groups. CONCLUSIONS: Well-funded, population-based tobacco-control programs can be effective in reducing smoking among students from all SES groups.