ABSTRACT
Diet is now believed to be crucial in modulating risks for cancer at selected sites. In humans, evidence for this idea has been obtained mostly by correlation studies and case-control studies conducted in many countries. Much stronger evidence was obtained by a census population based large scale cohort study conducted in Japan, which provided direct estimates of the risk for developing cancer with different frequencies of consumption of various foods (Hirayama, 1990a). A cohort study, or a population prospective study, is believed to be one of the best methods for confirming existing hypotheses as well as for generating new hypotheses regarding any key issue of health including diet and cancer. The subjective bias in obtaining dietary information is far less than in case-control studies, as the interview is carried out prior to the start of follow-up and therefore in most cases prior to onset of the disease. In addition, the study provides precious information on the relationship between diet and cancer and other diseases, covering the entire spectrum of mortality. The data obtained may also serve as an example of the usefulness of such studies in exploring the association between diet and cancer.
