A meta-analysis of six case-control studies on the relationship between reported exposure to environmental tobacco smoke (ETS) and lung cancer in nonsmoking Chinese women was performed, using the fixed-effect model. A total of 767 cases and 1193 controls from Shanghai, Guangzhou, Shenyang, Harbin, Xuanwei, and Hong Kong were reported in five of these six studies. (One study did not provide a clear description of the number of cases and controls). The Fleiss method was used to calculate the overall odds ratio (OR), the test of average degree of association (chi 2 assoc), the degree of homogeneity (chi 2 homog) and the 95% confidence interval (CI). The overall OR of lung cancer risk in nonsmoking Chinese women from exposure to ETS is 0.91 (95% CI, 0.75-1.10, chi 2homog = 4.51, P > 0.25). No statistically significant relationship was found between either the amount (cigarettes/day) or the duration (in years) of exposure to ETS and lung cancer. Bias and confounding factors are briefly discussed.