초록 close

Lung cancer death rates are increasingly used as an index of cumulative tobacco smoke exposure (smoke load). That index lacks direct validation and calibration in Koreans. This study calibrates that index against non-stomach non-lung (NSNL) cancer death age-adjusted rates in Korean males from 1985 to 2002. We assessed temporal smoke load/NSNL cancer rate associations and smoking-attributable fractions (SAF) of 2002 cancer death rates in Korean males. This study uses lung cancer death rates as an exposure index, linear regression of the lung/NSNL cancer death rates, the formula: SAF=(1- (rate in the unexposed)/(rate in the exposed)), and cohort-derived stomach cancer death SAFs. Lung cancer death rates predicted 93.0% of the variance in NSNL cancer death rates from 1985-2002 (p<0.0005). The findings suggest that the SAF of the all sites cancer death in Korean males was 54% in 2002 (sensitivity range 45~62%). Smoking may cause most premature cancer deaths and nearly all temporal NSNL cancer death rate disparities in Korean men. NSNL and lung cancer death rate trends appear to both be largely unaffected by other exposures or treatments. Urgent implementation of the Framework Convention on Tobacco Control in Korea appears merited. (Cancer Prev Res 11, 176-182, 2006)