Day Variation in Intake Implications for Design and Interpretation

Range within which true mean intake will fall 95% of time (+ % of observed intake) 

In theory, one might adjust the assessment of observed intake to take into account this noise in the data. As discussed below, this is feasible for population data however, at the level of a particular individual, it seems impractical. The only feasible approach to assessing the nutrient intake of individuals is to increase the number of intake records, preferably on days that are independent of one another and which appropriately represent the days of the week. It may not be feasible to obtain fully reliable information about a particular individuars usual intake (Table V). But at least one should recognize the extent of this unreliability in diet records and take appropriate precautions in interpreting nutrient intake data. 

---