Westat researchers coauthored a peer-reviewed article published in Nutrients examining methods used to estimate dietary intake among young children. Their study, “Evaluating Intake Estimation Methods for Young Children’s Diets,” used 3 different National Cancer Institute (NCI) methods to explore the impact of NCI’s transition from univariate and bivariate approaches to a Markov Chain Monte Carlo (MCMC) method. Using data from the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) Infant and Toddler Feeding Practices Study-2 (ITFPS-2), the article documents how methodological changes in intake estimation can affect reported intake estimates for young children.
As leaders in applying NCI methods to maternal and child dietary recalls, Westat’s Xiaoshu Zhu, PhD; Christine Borger, PhD; and Brenda Sun, MS, found that the MCMC method produced lower mean intake estimates than the univariate approach for episodically consumed foods, with mean differences of up to 37%. However, mean intake estimates for daily consumed foods differed by less than 2 percentage points.
This study is important for WIC as it looks for ways to increase children’s intake of whole grains and fish, which many U.S. children currently consume in insufficient amounts.