RESEARCH IN BIOSTATISTICAL METHODOLOGY AND MATHEMATICAL MODELING

点击翻译按钮获取中文摘要

Different statistical methods to adjust the effect of macronutrient intake for total energy intake are currently being used to analyze epidemiologic studies of diet and disease. This research examines the statistical properties of these methods. Published during the past year were: 1) a paper which interprets the regression coefficients of three alternative regression models showing that four different effects of interest (related to either adding calories to the diet or substituting sources of calories in the diet) are estimable by each model and derives the standard errors of these estimates; and 2) a paper which examines the behavior of these methods when the study subjects are categorized into a small number of groups according to their nutrient intake. When the true macronutrient intakes and their inter- relationships are known without error, this investigation shows that Willett's "residual" method is more powerful than the "standard" method (both measuring the effect of calorie substitution) and very similar to the "density" method. Submitted for publication is a paper which examines the ability of currently used energy adjustment regression methods to disentangle the effects of total energy from its component macronutient-specific parts. We determined that, based solely on results from a diet-cancer epidemiologic study, concluding the existence of a specific macronutrient effect as opposed to a generic energy effect is not possible without making additional assumptions.