Reducing inequalities in lung cancer incidence through smoking policies
Redirect to publisher's version
(publisher's version.url.txt, 47 bytes)
Introduction: Lower social class has higher lung cancer incidence, largely attributable to higher smoking prevalence among the lower social classes. We assessed the magnitude and time dimension of potential impact of targeted interventions on smoking on socioeconomic inequalities in lung cancer. Methods: Using population dynamic modelling, we projected lung cancer incidence up to 2050 in lowest and highest socioeconomic groups under two intervention scenarios (annual 10% increase in cigarette prices and health advertisement) and compared this to a scenario of no intervention. For the analysis we retrieved smoking prevalence data from the General Household Survey of England and Wales between 1980 and 2006 and cancer incidence data from the national cancer registry. Results: By 2050, the model projected that lung cancer incidence inequality would almost double (Incidence Rate Ratio (IRR) = 4.2 in 2050 vs. 2.5 in 2005) in men and slightly decrease (IRR = 2.4 in 2050 vs. 2.7 in 2005) in women compared to what was observed in 2005. If annual increase in cigarette price targeting the lowest socioeconomic group was implemented, socioeconomic inequality in lung cancer incidence in 2050 might be largely reduced (IRR = 1.5 and 1.4 among men and women, respectively). If in addition to annual price increase (targeted to the lowest socioeconomic group) health advertisement was implemented and successfully reduced smoking prevalence in the highest socioeconomic group, the lung cancer gap between the socioeconomic groups would be reduced by 78% and 58% in men and women by 2050. Conclusion: Even under the best scenarios, inequality in lung cancer was not fully eliminated within 45 years period. Though the process is lengthy, rigorous interventions may reduce the expected widening of the future inequalities in lung cancer. Modelling exercise such as ours relies heavily on the quality of the input data and the assumptions, thus caution is needed in interpretation of our findings and should consider all the assumptions taken in the analysis.