Diet qualities: healthy and unhealthy aspects of diet quality in preschool children1–3

Abstract

Diet quality indexes combine the healthy and unhealthy aspects of diet within a single construct, but few studies have evaluated their association. Emerging evidence suggests that predictors differ for the more and less healthy components of children's diets. Our objectives were to determine whether preschool-aged children's frequency of eating healthy foods was inversely related to their intake of unhealthy foods and to determine whether this differed by household income, maternal education, or child race-ethnicity. We analyzed data from a representative sample of 8900 US children (mean age: 52.5 mo) who were born in 2001 and participated in the Early Childhood Longitudinal Study-Birth Cohort. Primary caregivers reported the frequency with which children consumed fruit, vegetables, milk, juice, sugar-sweetened beverages (SSBs), fast food, sweets, and salty snacks in the past week. Response options ranged from none to ≥4 times/d. We created healthy (fruit, vegetables, milk) and unhealthy (SSBs, fast food, sweets, salty snacks) diet scores. Healthy diet behaviors were defined as ≥2 daily servings of fruit, vegetables, and milk. The prevalence of consuming fruit, vegetables, and milk ≥2 times/d (i.e., having 3 healthy diet behaviors) was 18.5%, and a similar proportion (17.6%) of children had none of these healthy behaviors. Contrary to our hypotheses, children with more healthy diet behaviors did not have lower unhealthy diet scores. The intake of healthy foods was not inversely associated with unhealthy foods overall or within any subgroup. Overall, the Spearman rank correlation between healthy and unhealthy diet scores was positive (r = 0.09). From the lowest to the highest strata of household income, these correlations were 0.12, 0.14, 0.14, 0.05, and 0.00, respectively. No evidence was found in US preschool-aged children of an inverse association between eating healthy and unhealthy foods. The implications of combining healthy and unhealthy aspects of diet quality within an overall index should be considered by researchers.