Extract

Drs. Risch and Miller state that we ( 1 ) and Bain et al. ( 2 ) criticize their earlier case–control study of lung cancer, and they purport to offer an alternative explanation for the discrepancy between the findings of case–control and cohort studies examining differences in lung cancer risk between men and women. They state (we believe incorrectly) that whereas the case–control studies have addressed the statistical interaction between smoking and sex on the risk of lung cancer, the cohort studies have examined only the main effects of sex. They conclude that current evidence indicates that female smokers are more susceptible to smoking-induced lung cancer than male smokers.

We would not characterize our comments on the Risch et al. ( 3 ) case–control study as criticism; indeed, we simply described the study's findings of higher odds ratios among women than men across smoking categories. We did note that the focus of case–control studies is the estimation of relative risks of lung cancer among smokers compared with never smokers, whereas the cohort studies, including that of Bain et al. ( 2 ) , compare actual incidence or mortality rates of lung cancer among men and women at similar levels of smoking. When the cohort studies examine rates of lung cancer among men and women with equivalent amounts smoked, they show consistently that the rates are about the same or even somewhat less for women. Thus, the studies clearly indicate that the probability of developing lung cancer is about the same in men and women when approximately equal smoking histories are considered.

You do not currently have access to this article.