Parental feeding practices in New Zealand

Background: Although various parental feeding practices have been associated with dietary intake and body weight in children, many studies are limited by sample size or lack of diversity, perhaps explaining conflicting results. Related areas of interest with limited exploration to date include how parental feeding practices relate to i) problem food behaviours in children and ii) food availability in the home. Objective: The aim of this thesis was to examine the relationships between parental feeding practices, diet, weight, home food availability, and problem food behaviours in a large, diverse sample of New Zealand children aged 4-8 years. Methods: 1093 children were recruited from general practice and secondary care clinics for a weight screening initiative (The MInT Study). Children and their parents attended a comprehensive health check including measurement of child body mass index (BMI) and completion of a questionnaire detailing measures of child dietary intake, parental attitudes of their child’s weight and diet, parental feeding practices, parental discipline practices, demographics, problem food behaviours and home food availability. A factor analysis of the Comprehensive Feeding Practices Questionnaire determined the feeding practices represented by the data. Examination of these factors in relation to demographic variables (ethnicity, socio-economic status, sex, maternal education and weight status) was undertaken using ANOVA. Correlations and regression analyses established associations between parental feeding practices and children’s dietary intake, the disciplining practices of parents, maternal BMI, problem food behaviours and home food availability using either the full sample or a subsample (overweight children only). This included exploration of some interactions. Bootstrap analysis was used to determine a linear regression model for BMI z-score. Results: The reassessment of the factor structure of the Comprehensive Feeding Practices Questionnaire determined that the original twelve factors did not reduce the data appropriately and that a five-factor structure was more suitable for this population. These five feeding practices were: healthy eating guidance, monitoring, parent pressure, restriction and child control. These feeding practices varied with demographics, for example parents used more restriction and less pressure with girls and with overweight children. Furthermore, healthy eating guidance and monitoring were associated with less dysfunctional parenting practices, more fruits and vegetables, and less sweet drinks. By contrast, child control exhibited inverse associations with these factors. Bootstrap analysis indicated that restriction (B=0.37, p<0.01), parent pressure (B=-0.19, p<0.01) and healthy eating guidance (B=-0.13, p=0.01), along with maternal BMI (B=0.03, p