This Article Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Haenszel, W. Search for Related Content PubMed PubMed Citation Articles by Haenszel, W.

Multiple Factors: Discussion of Statistical and Epidemiologic Aspects

National Cancer Institute, USPHS, Bethesda, Maryland

One can be optimistic about the contributions which the statistical-epidemiologic approach will make to future studies of multiple factors in cancer etiology. The epidemiologic method can lend itself to the study of both host characteristics and environmental factors. However, cancer epidemiology has not completely left the descriptive phase behind, and emphasis still needs to be placed on identification of factors associated with gradients in risk. As the fund of information on variations in cancer risk among population groups grows, more and more questions will be posed that that will lead to investigations of multiple factors. Epidemiologic observations should foster certain lines of experimental work, the findings from which can later be incorporated into more refined epidemiologic studies, thus continuing and sustaining the interaction between epidemiology and experimental carcinogenesis.

Above all, investigators should not be deterred by possible imperfections in the data they can collect. It is better to do what can be done today, but with an eye to the future and recognizing that current, imperfect data are to some degree self-perfecting as provisional findings are introduced into the design and pursuit of new studies.