Seminar on Atherosclerosis

T HE correlation between the amount of dietary fat, the concentration of serum cholesterol and the incidence of ischemic heart diseaset is widely accepted as a cause and effect relationship by nutritionists, public health authorities, biochemists, practising physicians and by the public itself. Evidence favoring this correlation has been presented in a large number of epidemiologic studies, among which those of Keys et al. [2,3] figure prominently. The validity of the conclusions drawn from these data has been questioned on numerous occasions, most recently by Yudkin [4], Yerushalmy and Hilleboe [fl and Mann [6]. Their warning bears repetition: a direct correlation, no matter how strong, cannot be used as proof of cause and effect. Each of these reports has emphasized the weaknesses inherent in the basic data, in the mortality statistics, in the food consumption data and in the statistical treatment applied to the data. Their evaluations do not indicate that dietary fat has nothing to do with the incidence of ischemic heart disease; they do emphasize that conclusive proof of a specific association is still lacking. Indeed, the correlations between ischemic heart disease and the intake of animal protein or sugar, the number of television sets or automobile licenses, are said to be stronger than those with total fat, animal fat, vegetable fat, butter fat or margarine. Other defects in the postulate that ischemic heart disease is caused by eating too much fat (or too