Exposure to point-of-sale tobacco displays and changes in susceptibility to smoking in secondary school students

Abstract

BackgroundTobacco companies in most European Union countries have to rely heavily on point-of-sale (PoS) displays to promote their products, and many countries are considering prohibiting this medium of marketing. To assess whether exposure to PoS displays affects smoking susceptibility (absence of a decision not to smoke) or uptake among young people, we surveyed PoS exposure and smoking behaviour, prospectively over 1 year, in a cohort of secondary school students in Nottinghamshire. The study was undertaken immediately before legislation prohibiting PoS displays in large retailers was implemented in England, in April, 2012. MethodsWe used data obtained from a cohort survey in eight schools undertaken in March, 2011, and March, 2012. In 2012, we received questionnaires from more than 4000 children (response rate ∼64%) and individually identified and linked responses from 2300 children for matched analysis. The questionnaire obtained data for demographic and socioeconomic variables, smoking status (never vs ever) and susceptibility to smoking among never smokers, rebelliousness, academic performance, smoking among family members and friends, exposure to and awareness of PoS displays, and number of brands of tobacco products recognised at PoS. We defined susceptibility to smoking using three questions on intentions to try smoking or to smoke within the next year and whether a respondent would smoke if one of their best friends were to offer a cigarette. We used multivariate logistic regression to examine the association between exposure variables and changes in susceptibility to smoking between 2011 and 2012. Likelihood ratio tests were used to identify which of the variables should be included in the final model. To avoid possible bias, we selected a sample of schools that are representative of the population. Questionnaires were filled in during school activities to avoid external influences. We also made efforts to guarantee respondents that the information provided would not be available to anyone outside the research group. We used multivariate models to adjust for a range of variables that are probably confounders and thus aimed to avoid any overestimates of the effect. FindingsIn 2011, of 2090 children who were never smokers, 1589 were non-susceptible and 501 susceptible. Of the non-susceptible, 313 became susceptible in 2012, and 253 changed their status from being non-susceptible or susceptible never smokers to smokers. After adjustment for potential confounding, PoS exposure was independently associated with an increased risk of becoming susceptible to smoking (odds ratio [OR] 1·65, 95% CI 1·16–2·35). However, there was no significant association between frequency of noticing PoS displays (most or every time vs sometimes or less) and smoking uptake (OR 1·55, 95% CI 0·84–2·85 for non-susceptible never smokers and 0·81, 0·43–1·51 for susceptible never smokers). Additionally, a higher number of cigarette brands recognised at the PoS seemed related to a higher risk of becoming susceptible to smoking, but not a smoker. Although we found some evidence at univarite level suggesting that more frequent exposure to PoS displays is related to higher risk of becoming a smoker, the effect was not significant at multivariate level. InterpretationExposure to PoS displays increased the likelihood that young people became susceptible to smoking, but was not a predictor of progression from susceptibility to uptake of smoking. Although we used only self-reported data and the number of children whose susceptibility or smoking status changed during the 1 year of the study was small, these findings suggest that PoS displays may cause young people to become likely to experiment with smoking, and hence in some cases progress to becoming a smoker. Removal of PoS displays is therefore likely to reduce smoking among young people. FundingThis study was supported by funding from the Cancer Research UK, Department of Health, Nottingham City PCT, and UK Centre for Tobacco Control Studies, with core funding from the British Heart Foundation, Cancer Research UK, Economic and Social Research Council, Medical Research Council, and the Department of Health under the auspices of the UK Clinical Research Collaboration.