Background
In countries at early stages of the tobacco epidemic, relative risks of lung cancer due to smoking could be under-estimated, resulting in unduly low assessment of disease burden arising from smoking.
Aim
To examine the effect of smoking and smoking cessation on lung cancer mortality in Guangzhou, China.
Methods
A population-based cohort (21,645 women and 8270 men) aged 50+ years enrolled from 2003 to 2008 was followed until the end of 2012. Multivariable Cox regression was used to calculate hazard ratios (HRs).
Results
After a mean follow-up of 7.0 years (SD=1.5), 1,817 deaths (878 men) from all-causes and 192 (104 men) from lung cancer were observed. Follow up is in progress, and preliminary analysis showed that, after age and sex adjustment, compared to never smokers, current smokers showed increased risks of lung cancer mortality (HR=3.99, 95% confidence interval 2.61-6.09) with a dose-response pattern (HR for 1-39 pack-years: 3.64 (2.36-5.63); for 40+ pack-years: 4.77 (2.82-8.07); P for trend <0.001). The age adjusted HR for current smokers was 4.68 (2.77-7.91) in men and 3.52 (1.69-7.34) in women (P for sex interaction 0.44). Compared to current smokers, the age and sex adjusted HRs of lung cancer for former smokers after 1-9, 10-19 and 20+ years of abstinence were 0.57 (0.33-0.98), 0.43 (0.21-0.86) and 0.32 (0.14-0.75) (P for trend <0.01), respectively, which were still higher than that of never smokers (0.25 (0.17-0.36), (P for trend <0.01)).
Conclusions
Compared to earlier Chinese cohorts, our risk estimates of lung cancer from smoking were considerably higher and closer to those from western populations. This has very important implication on the estimation of current and future disease burden from tobacco in China and countries at early stages. The finding of rapid decline in lung cancer risk after quitting underlines the importance of promoting cessation in China.