Contribution of lifetime smoking habit in France and Northern Ireland to country and socioeconomic differentials in mortality and cardiovascular incidence: the PRIME Study

Background This study examines the contribution of lifetime smoking habit to the socioeconomic gradient in all-cause and smoking-related mortality and in cardiovascular incidence in two countries. Methods 10 600 men aged 50-59 years were examined in 1991-4 in centres in Northern Ireland and France and followed annually for 10 years. Deaths and cardiovascular events were documented. Current smoking habit, lifetime smoking (pack-years) and other health behaviours were evaluated at baseline. As socio-occupational coding schemes differ between the countries seven proxy socioeconomic indicators were used. Results Lifetime smoking habit showed marked associations with most socioeconomic indicators in both countries, but lifetime smoking was more than 10 pack-years greater overall in Northern Ireland and smoking patterns differed. Total mortality was 49% higher in Northern Ireland than in France, and smoking-related mortality and cardiovascular incidence were 93% and 92% higher, respectively. Both lifetime smoking and fibrinogen contributed independently to these differentials, but together explained only 42% of the difference in total mortality between countries, adjusted for both biological and lifestyle confounders. Socioeconomic gradients were steeper for total and smoking-related mortality than for cardiovascular incidence. Residual contributions of lifetime smoking habit ranged from 6% to 34% for the seven proxy indicators of socioeconomic position for total and smoking-related mortality. Socioeconomic gradients in cardiovascular incidence were minimal following adjustment for confounders. Conclusion In Northern Ireland and France lifetime smoking appeared to explain a significant part of the gradients in total and smoking-related mortality between socioeconomic groups, but the contribution of smoking was generally small for cardiovascular incidence.