Coffee

baked goods and therefore with sugar has constituted but a small proportion of its total intake, most fat being eaten independently of any sweetening. In contrast, the association of sugar with cream on fruit desserts and with fat in cakes, pies and similar foods results in an overwhelming proportion of the sugar that is consumed being eaten together with fat. Even when sugar is used to sweeten tea or coffee, the addition of milk or cream results in some association. It is consequently possible to dissociate the effects of fat consumption from those of sugar, but virtually impossible to dissociate the effects of sugar consumption from those of fat. This would readily help to explain the negative findings of the Keys multivariate analysis. Keys drew attention to the rising incidence of coronary heart disease in both the United Kingdom and the United States during the first half of the twentieth century when fat consumption in both countries rose considerably but sugar consumption changed very little. This does not exonerate sugar as a risk factor. Theoretically, if both sugar and animal fats were risk factors and sugar consumption held constant while fat consumption increased, the incidence of CHD would rise. Longitudinal studies of food consumption would only exonerate sugar if a society could be found in which fat intake was constant, sugar consumption rose but coronary heart disease incidence did not change. Such a society is unlikely ever to exist. The observations of Keys therefore, while not implicating sugar as a risk factor, certainly do not exclude it.107 In conclusion, excessive sugar intake is conducive to obesity and results in surges in serum insulin levels with an eventual increase in insulin resistance. The results of experimental dietary studies suggest that, particularly in the context of a coincidental liberal saturated fat intake, high sugar consumption is followed by elevation of serum cholesterol as well as triglyceride levels. The relation of high cholesterol levels to increasing incidence of coronary heart disease is well established and recent studies have implicated liberal sugar consumption with serum triglyceride elevation, insulin resistance and rising glycosylated levels as well. All of these are established risk factors for CHD. The great increase in English sugar consumption during the Georgian era has been shown to coincide with a sharp increase in consumption of animal fats and reasons for a direct linkage between the effects of these two dietary changes have been demonstrated. The evidence therefore suggests that the dramatic rise in sugar usage in eighteenth-century England could be implicated as an ancillary factor contributing to the initial emergence of angina pectoris and its increasing prevalence thereafter.