In Reply. —
Control patients were matched to case patients by age in our study, while sexual history, education, and church attendance (to help control for sexual activity) were controlled in the smoking analyses so that they also could be examined as independent risk factors in other analyses.1 Because of the risk of cervical cancer associated with these variables,1 major differences in their distribution between case and control subjects were noted. These variables, although associated, did not present, in the analysis, a problem of multicolinearity. While logistic regression models have the capability to control for confounding, they can only control to the extent that the data are accurate. Thus, biased risk estimates could result if important confounding variables were reported inaccurately, although they were controlled in the analyses. While there is the possibility that sexual history is reported inaccurately, we do not believe that case patients reported differently
Slattery ML. Exposure to Cigarette Smoke and Cervical Cancer-Reply. JAMA. 1989;262(4):499. doi:10.1001/jama.1989.03430040069019
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: