Policy elements as predictors of smoking and drinking behaviour: the Dutch Cohort Study of secondary schoolchildren

Changes in smoking and drinking behaviour in relation to policy and economic variables were investigated among Dutch secondary schoolchildren in two measurements. The response rate was 82% during the second measurement. In bivariate analyses the policy variables 'pressure by tobacco advertising' and 'incorrect knowledge about dangers of tobacco consumption' and the economic variables 'money' and 'free availability from parents' predicted weekly smoking (among all respondents), and 'the experienced anti-smoking campaigns' predicted weekly smoking (among non-smokers during the first measurement). Only the economic variables predicted weekly drinking. A general and a specific theory of different mechanisms were designed to investigate whether these predictors are independent of other influences. We assumed that changes in smoking and drinking behaviour are predicted at three levels: (a) previous behaviour, (b) socio-demographic variables and (c) possible causes and intermediary factors. In multivariate analyses, only the prediction of weekly smoking by 'experienced anti-smoking campaigns' and weekly drinking by economic variables remained significant. The findings are relevant when policy priorities are developed.