Date of Award

Summer 5-2019

Degree Name

Master of Science (MS)



Second Advisor



Background: Past studies developed two- and three-part models to estimate the usual intake of episodically consumed foods in adults but not in children. The goal of this study is to evaluate if the three-part method is successful, by incorporating the dietary screener and two 24-hour recalls, in estimating the usual intake of fruits and vegetables consumed by boys and girls at baseline in TX Sprouts. Methods: Secondary data analysis of two 24-hour recalls and a dietary screener from TX Sprouts. Two approaches were used. First, the three-part model estimated the distribution of the usual intake of fruits and vegetables in children using the program Intake_epis_food() in Rstudio. The three parts to the model included: if a fruit or vegetable was consumed, the number of servings consumed given consumption of a fruit or vegetable, and energy intake as reported by the 24-hour recalls. Five models were estimated for usual intake in boys and girls separately: (i) fruits only, (ii) vegetables including white potatoes and starchy vegetables, (iii) vegetables excluding white potatoes and starchy vegetables, (iv) fruits and vegetables including white potatoes and starchy vegetables, and (v) fruits and vegetables excluding white potatoes and starchy vegetables. Second, the average of two 24 hour recalls was estimated for each participant and a multiple linear regression was used for the usual intake of the average fruits and vegetables, in boys and girls separately. The models controlled for the average energy from two 24-hour recalls, age, and the number of servings of fruits and vegetables as reported in the dietary screener. Results: There was a total of 737 children (335 boys, 402 girls) from TX Sprouts with complete dietary recall and screener data at baseline. The three-part models reported point estimates for each parameter in the model, samples from posterior distributions as well as from the distribution of usual intake and usual energy intake, and summary statistics of usual intake, usual energy intake, and the energy adjusted usual intake. Discussion: This investigation found that the usual intake of fruits and vegetables can be successfully estimated in children using the three-part method with a dietary screener. Additionally, we found that fruits and vegetables are not consumed episodically in children of TX Sprouts. Past studies have also found frequent intake of fruits and vegetables among children; however, the number of servings per day consumed do not meet dietary recommendations.