Nutrition and cancer-an international journal vol:22 issue:1 pages:85-98
Highly significant correlations exist between total cancer mortality and age expressed by a log total cancer mortality-log age equation (mean r(2) 0.991 in men and 0.996 in women) or by a second-order polynomial equation including age and age(2) (mean r(2) 0.999 in men and 0.998 in women). In all countries considered (n = 32), the second-order term of age is negative, indicating a decrease in the rate of rise of log cancer mortality at older age. This could be explained by a lesser accuracy of the diagnosis of cancer at older age, by selective survival of subjects resistant to cancer, by a cohort effect, or by a decrease in the rate of growth of cancer al older age. The decrease in the rate of rise of cancer mortality, after 65 years of age occurs in all countries and applies to nearly all cancers except breast cancer in women after 75 years of age. A high cancer mortality in a country is characterized by a low intercept and a steep slope of the log mortality-log age equation. These parameters are influenced by dietary fat intake in men and women, with saturated fat increasing total cancer mortality and the ratio of polyunsaturated to saturated fat and the ratio of unsaturated to saturated fat decreasing it. The data on dietary fat were obtained from the Food and Agriculture Organization of the United Nations (FAO) food balance sheets (n = 32) and from dietary sunveys (n = 21). Both vary in the same direction, but only the dietary data from the FAO correlate significantly with cancer mortality. This finding points toward a relationship between the level of dietary fat intake and total cancer mortality at the population level.