Assessing the impact of the diet on cardiometabolic outcomes: are multiple measurements post-intervention necessary?

The purpose of this study was to examine how using the mean of two consecutive measurements vs. one measurement post-treatment influences the sample size required to detect changes in cardiometabolic risk factors in dietary studies. For a given statistical power, using the mean of two measurements taken on consecutive days post-treatment instead of a single measurement significantly reduces the sample size required to observe changes in triglyceride, total apolipoprotein B100, and C-reactive protein concentrations in the context of a supplementation study. In the context of a controlled-feeding study, this gain is seen only in the case of change in triglyceride concentrations.