Population surveys often collect dietary intake data by using one or two 24-h recalls (24HR), mainly to minimize cost and respondent burden. However, in order to increase accuracy in estimating usual intake distributions, a larger number of 24HRs may be advisable.Objective:
The purpose of this study was to identify whether estimates of the intake and prevalence of nutrient inadequacy based on 3 d are better than those that are based on 1 d of information.Methods:
We conducted a national survey in 31 cities in Mexico using a sample that included 1073 individuals of both sexes, from birth through 79 y of age. Dietary information was collected by using 3 multiple-pass 24HRs (on nonconsecutive days). We estimated intake, adequacy, and prevalence of energy and nutrient inadequacy using information from 1 d and also from 3 d, adjusted for day-to-day variability, using PC-SIDE software. The prevalence of inadequacy was estimated by using the reference values proposed by the US Institute of Medicine.Results:
We found high prevalences of inadequacy for fiber (73-99%), iron (31-94%), calcium (2-85%), vitamin A (0.1-61%), and folates (2-80%) among various age and sex groups. Energy intake results showed that the variance in the estimated usual 3-d intake distribution was smaller than the variance of distribution estimated from a single daily intake. We observed bigger differences in prevalence of inadequacy between 1 and 3 d for several nutrients. For example, in preschool children, the prevalence of inadequacy of folate and calcium was 30% and 43%, respectively, with 1-d recalls and 3.7% and 4.6%, respectively, with 3-d recalls.Conclusions:
We conclude that the adjusting-by-variance method using 3 d of 24HR allows for a more accurate estimation of usual intake, dietary adequacy, and prevalence of inadequacy, thereby reducing the measurement error that could compromise the results and conclusions of surveys. J Nutr 2016;146:1043-50.