Abstract

Traditional approaches and study designs in cancer epidemiology have not been very successful identifying and evaluating potential risks and/or protective factors associated with disease. Failure is often due to small study sample size and inadequate exposure information. In this paper, we discuss issues and approaches relevant to these two challenges.

Multicenter study design is proposed as a way to increase study size and to mitigate criticism of meta-analysis of independent studies. Individual studies from a multicenter study can be either pooled using original data, or combined by meta analysis of summarized results. A multicenter study of large cohort or case-control studies also offers an exciting opportunity to study the contribution of epigenetic events that may be associated with life-style and environmental risk factors for human health. Optimizing methods for exposure assessment and reduction of exposure misclassification represent difficult components in epidemiological studies. Biomarkers present a potentially useful approach for improving exposure estimates. An example is provided to demonstrate how biomarkers of exposures can provide valuable information in addition to traditional exposure measurements in epidemiological studies.

Finally, it is argued that risk assessment and the precautionary principle should not be viewed as conflicting paradigms but, rather, as a complementary approach according to the data available. Together they may be used for developing appropriate policies to address risks posed by exposure to carcinogens and a wide spectrum of other health hazards.