In a recent study of urban air pollution, a Utah county with a steel mill was compared with a county without a steel mill. The result was that 38% of respiratory cancer deaths could be attributed to the air pollution emanating from the mill. Rates for smoking in this previous study were not adjusted, but assumed rats were similar in both counties. We used smoking information obtained from an ongoing radon and lung cancer case-control study to adjust for smoking, and no difference was found in incidence rates of respiratory cancer in the county with the steel mill, compared with the other urban counties and the rural counties among male and female nonsmokers and male smokers. There was a slight excess of lung cancer among female smokers in the county with the steel mill when compared with the other urban counties (rate ratio [RR] = 1.3, 95% confidence interval [95% CI] = 1.0-1.6), but there was no effect in nonsmoking women. We conclude that the findings of the previous study can be explained by differences in smoking rates between the county with the steel mill and the other counties.