Association between tobacco control policies and smoking behaviour in Europe

BACKGROUND: Smoking is an important public health problem. The European Health Interview Survey (EHIS) provides regular data on prevalence and determinants of smoking for the populations of European countries, while the Tobacco Control Scale (TCS) quantifies the implementation of tobacco control measures. The aim is to examine the associations between tobacco control policies at country level and individual-level smoking prevalence. METHODS: Using multi-level regression modelling, individual data from EHIS 2019/2020 and country-level data from TCS 2019 were analysed. The outcome measure was self-reported current smoking (daily or occasional) at the individual level. Predictor variables were at the individual level: gender, age and education attainment level, and at country level: scores of TCS policy components ‘Price', ‘Public place bans', ‘Advertising bans', ‘Health warnings', ‘Treatment’ and ‘Illicit trade'. Microdata from the 3rd wave of EHIS for 28 European countries were used with a total of n = 66.934 individuals aged >15 years. A weighting factor was applied to consider each country proportionally according to its sample size in EHIS. RESULTS: There are large cross-national differences in current smoking prevalence ranged from 12.6% (95% confidence interval (CI) 11.9-13.3) in Sweden to 36.2% (95% CI 35.0-37.4) in Bulgaria. In the final multi-level model besides individual variables (female gender, age 60 years and over, medium or high education level), following variables at country level were significantly associated with lower prevalence of current smoking: ‘Price’ (Odds Ratio (OR)=0.96 (p < 0.05) and ‘Treatment’ OR = 0.88; (p < 0.01). For ‘Public place bans', ‘Advertising bans', ‘Health warnings’ and and ‘Illicit trade’ association were in same direction but without significance. CONCLUSIONS: Tobacco control policies help to reduce the prevalence of smoking in population, although the different measures have differing impact. KEY MESSAGES: • Our study confirms the link between the implementation of tobacco control policies and smoking prevalence using recent data. • Policy components have different effects on smoking prevalence at population level.