Changing social gradients in cigarette smoking and cessation over two decades of adult follow-up in a British birth cohort

Background We aimed to establish whether socioeconomic gradients in smoking among men and women increase with age as a result of differential uptake, quitting and smoking persistence over time. Methods A prospective British birth cohort ( all births 3 - 9 March 1958) was followed to 41 years. Analyses of smoking at 41 years by socioeconomic position of origin include 10 521 participants and for socioeconomic position at 23 years n = 9240. Results By 41 years half of the cohort had smoked regularly ( greater than or equal to1 cigarette/ day). Smoking prevalence peaked at 23 years ( 40 per cent) and subsequently declined; quitting increased between 23 years ( 10 per cent) and 41 years ( 29 per cent). Individuals from manual backgrounds were more likely to smoke and less likely to quit than those from non-manual groups, and these differences increased over the two decades during which the cohort was followed up. For social position of origin, the odds ratio for current smoking at 23 years among women was 1.28 ( 95 per cent confidence interval (CI) 1.21, 1.35), i.e. a 28 per cent greater risk of smoking per unit increment on a four-point scale from professional/ managerial to unskilled manual. The odds ratio increased to 1.45 ( 95 per cent CI 1.36, 1.56) at 41 years, trend over time p = 0.01. For men, equivalent results are 1.18 (1.11, 1.24) at 23 years and 1.33 ( 1.24, 1.42) at 41 years, trend p = 0.01. The social gradients in current smoking also increase over time for men and women using social position at 23 years. Conclusions Social gradients in smoking have become more marked across the lifecourse of this birth cohort. This implies continued socioeconomic inequalities in future health outcomes in a contemporary adult population.