Characterisation of smoking behaviour across the life course and its impact on decline in lung function and all-cause mortality: Evidence from a British birth cohort

Abstract

Objectives: To describe smoking trajectories from early adolescence into mid-life and to examine the effects of these trajectories on health and all-cause mortality. Methods: A nationally representative birth cohort study including 3387 men and women followed up since their birth in 1946 in England, Scotland and Wales. The main outcome measure is all-cause mortality by age 60 years and rate of decline in forced expiratory volume in 1 second (FEV1). Results: Eighteen per cent of the sample were categorised as lifelong smokers (smokers at all six waves at ages 20, 25, 31, 36, 43, 53 years), of whom 90% had begun smoking by age 18 years. By age 60 years, 10% of all lifelong smokers had died. They had a threefold increase in mortality rate compared with never smokers (hazard ratio (HR) 3.2, 95% confidence interval (CI) 2.1 to 4.8). For predominantly smokers (smokers for at least four of the six data collections), mortality rate remained higher than never smokers (HR 1.6, 95% CI 1.0 to 2.5). Predominantly non-smokers did not differ from those who never smoked (HR 1.3, 95% CI 0.9 to 2.0). Using the most recent smoking status available, current smokers had more than double the risk of mortality compared with never smokers (HR 2.4, 95% CI 1.6 to 3.5). Lifelong smokers and predominantly smokers had a greater rate of decline in lung function than never smokers (regression coefficients -18 ml/year, 95% CI -22 to -13; -6, 95% CI -10.3 to -1.7 respectively). For current smokers, the decline was 8.4 ml/year (95% CI -12.0 to -5.0) faster than never smokers. Conclusions: The strength and differentiation of adverse effects identified by using simplified smoking behaviours has highlighted the advantages of obtaining further information on lifelong smoking behaviour from former smokers, rather than just current smoking status.