The effect of smoking on lung cancer: ethnic differences and the smoking paradox.

The objectives of this review were to determine whether the smoking paradox still exists and to summarize possible explanations for the smoking paradox. Based on published data, we compared the risk of cigarette smoking for lung cancer in Western and Asian countries. We extracted data from the relevant studies about annual tobacco consumption, lung cancer mortality rates according to smoking status from each country, and possible explanations for the smoking paradox. A significantly greater risk of lung cancer death was found among current smokers in Asian countries than among nonsmokers, with relative risks (RRs) of 4.0 to 4.6 for Koreans, 3.7 to 5.1 for Japanese, and 2.4 to 6.5 for Chinese. Although a significantly greater risk of lung cancer was present among current smokers in Asian countries, the RRs in Asian countries were much lower than those reported in Western countries (range, 9.4 to 23.2). Possible explanations for the smoking paradox included epidemiologic characteristics, such as the smoking amount, age at smoking initiation, and the use of filtered or mild tobacco. The smoking paradox definitely exists, but may be explained by major epidemiologic characteristics. Therefore, the smoking paradox should not be interpreted as indicating that tobacco is safer or less harmful for Asians.