Epidemiologic studies of diet and cancer

Correlational studies comparing national diets with mortality rates of colon and breast cancer indicate very strong associations, on the order of correlation coefficients = +0.8, between intake of meat or animal fat and rates of both these cancers. However, both case-control and prospective cohort studies have found little positive association between these dietary factors and risk of breast cancer. This lack of a positive association, at least in the largest prospective study, cannot be explained by insufficient variability in diet in the study population or by imprecision in the dietary questionnaire, although it is possible that the latent period between dietary intake and diagnosis of breast cancer is so long that an effect of fat intake could not be seen. However, the normal risk of breast cancer in Seventh-Day Adventists and vegetarian nuns, the small increase in breast cancer rates in Japan since World War II despite a 2.5-fold increase in fat intake, and the lack of correlation between fat intake and risk of breast cancer mortality among rural Chinese populations all suggest that a major association between meat and animal fat intake and risk of breast cancer is unlikely or weak. In contrast, the striking international correlations between intake of meat and animal fat and colon cancer rates are generally supported by case-control and cohort studies. In particular, associations between intake of red meat and colon cancer have been seen in many, although not all, of these studies.(ABSTRACT TRUNCATED AT 250 WORDS)