Mortality risks among heavy-smokers with special reference to women: a long-term follow-up of an urban population

Increased mortality risks associated with smoking are well established among men. There are very few population-based studies comprising a sufficient number of heavily smoking women, measuring the direct effect of smoking on mortality risks. Between 1974 and 1992, 8,499 women and 13,888 men attended a health screening programme including reporting of smoking habits. Individuals were followed for total mortality until 2005. All-cause, cancer, cardiovascular, lung cancer and respiratory mortality were calculated in smoking categories < 10 g per day, 10-19 g per day, and >= 20 g per day with never-smokers as a reference group and with adjustments for co-morbidities, socio-economic and marital status. For respiratory mortality and lung cancer adjustments for FEV1, socio-economic and marital status were performed. Smoking was associated with a two to almost threefold increased mortality risk among women and men. The relative risk (RR) with 95% confidence interval, (CI) for women who smoked 10-19 g per day was 2.44 (2.07-2.87), and for those who smoked 20 g per day or more the RR (95% CI) was 2.42 (2.00-2.92). Smoking was a strong risk factor for cardiovascular mortality among women, the RR (95% CI) for women who smoked 10-19 g per day was 4.52 (3.07-6.64). Ex-smoking women showed increased risks of all-cause mortality; RR (95% CI) 1.26 (1.04-1.52) cancer (excluding lung cancer); RR (95% CI) 1.42 (1.07-1.88) and lung cancer RR (95% CI) 2.71 (1.02-7.23) mortality. However, the cardiovascular; RR (95% CI) 1.18 (0.69-2.00) and respiratory; RR (95% CI) 0.79 (0.16-3.84) mortality risks were not statistically significant. This study confirms that as for men, middle-aged heavily smoking women have a two to threefold increased mortality risk. Adjustments for co-morbidity, socio-economic and marital status did not change these results.