CAN ENERGY ADJUSTMENT SEPARATE THE EFFECTS OF ENERGY FROM THOSE OF SPECIFIC MACRONUTRIENTS

Energy adjustment is used in nutritional epidemiology in an attempt to separate specific effects of macronutrients (carbohydrate, fat, and p rotein) from one another and from the generic effect of the total quan tity of energy consumed. However, models in which the risk of disease is allowed to depend simultaneously on daily total energy consumption and separate components of energy that sum to the total are not identi fiable: the specific effects of individual macronutrients and the gene ric effect of energy cannot be disentangled by multivariate analysis. The standard, residual, and partition methods exclude one or more macr onutrients from consideration, thereby allowing estimation, but the pa rameters that are estimated no longer represent specific macronutrient or generic energy effects. Therefore, an interpretation of a regressi on coefficient from these methods as a specific effect of a macronutri ent or as the generic effect of energy requires additional, almost alw ays questionable, assumptions. For example, a conclusion based on data alone that there is a specific fat effect upon the development of bre ast cancer but no specific effects of other macronutrients and no gene ric energy effect is not possible. Notwithstanding these serious probl ems, some useful etiologic inference still can be made.