Aspects of confounding and effect modification in the assessment of occupational cancer risk.

In occupational health epidemiology, the confounding effects of general risk factors such as smoking and alcohol abuse with regard to cancer and other disorders seem to be rather limited. It is desirable to control such risk factors whenever possible, however, but risk ratios of about two or more can rarely be explained by suggesting uncontrolled confounding. However, various general risk factors might modify the effect of an industrial exposure, a phenomenon that sometimes tends to confuse the discussion of etiologic relationships between exposure and outcome.