The roles of smoking and cooking emissions in lung cancer risk among Chinese women in Hong Kong.

BACKGROUND We conducted this case-control study to evaluate smoking effect on lung cancer conditional on the level of exposure to cooking emissions and to explore whether there is a joint effect of these two risk factors. SUBJECTS AND METHODS We selected 279 newly diagnosed primary lung cancer cases and 322 community controls from Hong Kong females, frequency matched by age group, and collected relevant data. We applied logistic regression to estimate lung cancer risk related to smoking and cooking fume exposure, expressed as total cooking dish-years, while adjusting for various potential confounding factors. RESULTS Current smoking was associated with four-fold increased risk, and ex-smoking with two-fold risk, which was not much influenced by cooking dish-years. No increased risk was observed in environmental tobacco smoking. Increasing intakes of yellow/orange vegetables and multivitamins were significant protective factors in all models. In the analysis of joint effect, the combination of smoking and cooking dish-years tended to have a greater risk than exposure to cooking fumes alone. There was a dose-response gradient with total dish-years in nonsmokers, but not in smokers. Smoking was more strongly associated with nonadenocarcinoma, whereas exposure to cooking fumes appeared to be related to both adenocarcinoma and nonadenocarcinoma. CONCLUSION We confirmed the important roles of smoking and cooking emissions in lung cancer risk among the women. These two major risk factors appeared to act independently.