Does policy adoption influence the association between socio-economic differences and smoking prevalence? A longitudinal analysis 2008-2015

Abstract: Abstract Introduction: Smoking is a major public health concern, accounting for nearly 700,000 deaths on an annual basis, globally. The socio-economic differences within smoking are well established, reflected in a social gradient that is estimated to account for half of the socio- economic differences in mortality for men aged 35-69 years in developed settings, such as the European region. Aim: The study aimed to investigate the association between socio-economic differences and smoking prevalence, and furthermore to explore the influence of policy adoption on the association between socio-economic differences and smoking prevalence. Methodology: This ecological longitudinal study adopted a quantitative approach to methodology, modelling panel data from 2008-2015. Data for units of analysis (n = 7) were extracted from public data sources including the World Bank, OECD, ESPAD and WHO’s Global Observatory. Statistical analysis was conducted using R software, for multivariable linear regression in the form of fixed effects regression models and interaction terms. Results: Smoking prevalence showed statistical significance in association to education (p=8.07x10-6)(p=7.63x10-5)(p=2.79x10-5),and employment rate (p=0.002)(p=0.003)(p=0.003), which were negatively and positively associated, respectively. In isolation, policy adoption of tobacco control strategies showed no statistical significance (p=0.696)(p=0.459) in association to smoking prevalence. However both models, which incorporated a measure for policy adoption, indicated statistical significance (p=2.80x10-9) (p=2.31x10-9), yet at a slightly decreased significance level when compared to the model that did not (p=5.99x10-10). Conclusions: All three models, with and without incorporated measures for policy adoption influenced the association between smoking prevalence and socio-economic differences, which could be deduced to omitted variable bias. Findings suggest policy regarding tobacco control needs to be stronger and more clearly regulated, with shifted focus on health education, in order to have a truly significant effect, and to overcome the socio-economic differences present within smoking. Key words: Tobacco, Tobacco control, Tobacco prevention and Socio-economic status