A study in six European cities

Aims: This study investigates the link between personal income and smoking among adolescents, and aims to answer the following questions: (i) to what extent is personal income related to smoking, independent of family socio-economic status (SES) and (ii) does the association between personal income and smoking apply to different subpopulations? Design: Cross-sectional study. Setting: Six cities from European countries (Amersfoort, the Netherlands; Coimbra, Portugal; Hannover, Germany; Latina, Italy; Namur, Belgium; Tampere, Finland) in 2013. Participants: A school-based sample of 10794 adolescents aged 14-17 years. Measurements: We modelled smoking experimentation, weekly smoking, daily smoking and (among daily smokers) smoking intensity as function of personal income, adjusting for age, sex, family SES, parental smoking and country. We tested interactions between personal income and covariates. Stratification analyses were performed for the variables for which interactions were significant. Findings: Adolescents in the highest income quintile were more likely to be smoking experimenters [odds ratio (OR) = 1.87; P < 0.01], weekly smokers (OR = 3.51; P < 0.01) and daily smokers (OR = 4.55; P < 0.01) than those in the lowest quintile. They also consumed more cigarettes per month (β = 0.79; P < 0.01). Adjusting for family SES did not modify the significance of relationships, and increased the magnitude of the association for daily smoking. None of the interactions between covariates and personal income was significant for smoking measures. For the intensity of smoking, the interaction was significant for SES. The stratified analysis showed a non-significant association between smoking intensity and personal income among the oldest adolescents and those with the lowest SES background, while significant among younger and higher SES backgrounds. Conclusion: In the Netherlands, Portugal, Germany, Italy, Belgium and Finland, adolescents' personal income is related positively to smoking behaviours independent of family socio-economic status (SES). However, among low socio-economic status adolescent daily smokers, the association between the intensity of smoking and personal income is weaker.publishe