A “bottom up” typology of parent feeding practices relates to Head Start child BMI and diet

We examined how parent practices for feeding preschool children relate to child BMI and diet in 4 year old Head Start (HS) children (n= 704). Feeding practices were derived from parents in structured focus groups. Parents grouped the parent-derived feeding practices into five clusters (teachable moments, practical methods, supervision/contingencies, junk foods/sweets restriction, and enhancing healthy foods availability) based on perceived similarity and rated the effectiveness of each practice with their child. Parents also rated the frequency of using parenting practices (authoritarian, authoritative, and permissive) from the Caregiver Feeding Styles Questionnaire. Parent reported child non-Head Start diet on 3 separate days. Energy-adjusted servings of sugar-sweetened beverages related positively (p< 0.05) while lean meats/eggs/beans (p< 0.05) and refined grains (p< 0.05) related negatively to child BMI z-score. Descending quartiles of sugar-sweetened beverages and increasing quartiles of lean meats/eggs/beans and refined grains were summed to create a composite diet score that related to child BMI z-score (p< 0.01). Factor analysis of the five parent-derived clusters and the three questionnaire-derived sets demonstrated three factors. Factor 1 was moderate on the supervision/contingency cluster and high for the other four parent-derived clusters. Factor 2 was high on authoritarian, authoritative, and permissive sets. Factor 3 was high on the supervision/contingency and moderate on authoritarian practices, practical methods, and junk foods/sweets restriction. Cluster analysis of the three factors demonstrated four parent feeding types. Type 1 was negative for factor 1, high for factor 2, and moderate for factor 3. Type 2 was low on factors 1 and 2 and negative for factor 3. Type 3 was low on 1, negative for factor 2 and moderate for factor 3. Type 4 was high on factor 1 and moderate on factors 2 and 3. Type 2 was associated with higher child BMI Z-score, lower composite diet score, and lower intake of fruit and non-fried vegetables and higher intake of fried vegetables, chips and cake/desserts than type 4. Types 1 and 3 were intermediate. A typology developed from parenting feeding practices defines children with adverse diets and BMI and may permit targeting interventions to limit increases in Head Start child BMI.