Exploring aspects of parental control over feeding: influences on children’s eating behaviour and weight.

Abstract

The current research was developed from a review of the literature on the influence of parental control on children's eating and weight, and in particular the research examining parental restriction and poor self-regulation of energy. The impetus for the research was the confusion in the literature about the positive and negative aspects of control and the need to explore the influence of parental control at earlier periods of child development in socio economically diverse samples. The aim of the first study was to investigate the factor structure and psychometric properties of the Child Feeding Questionnaire (CFQ) in an Australian sample of preschool children. Consistent with previous research, the factor structure was problematic with respect to the Restriction subscale. The two items that ask about the use of rewards showed poor cohesion with other items on the Restriction subscale, raising concerns about the conceptualisation of restriction as measured by the CFQ. The second study sought to determine whether currently available measures of restriction were appropriate for assessing parent behaviours with toddlers, and to explore the strategies that mothers use to control their toddlers' intake of energy-dense snack foods. A qualitative methodology was employed and interviews were conducted with 22 mothers who had toddlers aged between 18 and 24 months. A key finding was a lack of variability in responses to the CFQ Restriction subscale in contrast to the range of individual differences in parents' reported use of control. A thematic analysis provided the basis for the development of a questionnaire to quantify the key concepts that were identified. Exploratory factor analysis was then used to refine these concepts resulting in a five factor structure that has been named the Toddler Feeding Questionnaire (TFQ). The TFQ incorporates a broad range of measures of parental control over feeding, including how much parents allow access to energy-dense foods (Allow Access), the rules associated with managing intake of these foods (Rules), and flexibility in the way rules are applied (Flexibility). It also includes a measure of parental self-efficacy beliefs (Self-efficacy), which may influence the use of feeding practices, and a measure of toddlers' attraction to energy-dense foods (Child's Attraction), a characteristic of the child that may influence both parent feeding practices and energy intake. The factor structure was examined in a sample of mothers of toddlers, and replicated in a second sample of mothers of preschool children, who responded retrospectively about their feeding practices. Subsequent analyses were conducted to examine the internal consistency, test-retest reliability, convergent, discriminant and concurrent validity of the TFQ factors to provide a preliminary validation of the instrument. To further examine the validity of the TFQ, feeding practices of normal weight, overweight and obese parents were compared. The relationship between the TFQ factors (based on retrospective reports) and BMI z scores in preschool children were also examined. In the preschool sample only, normal weight parents had significantly more rules compared to obese parents and were less flexible compared to overweight parents, and obese parents allowed their children more access to snack foods compared with overweight and normal weight parents. The TFQ factors were not associated with BMI z scores in preschool children. The final study investigated the association between feeding practices, including the TFQ factors and CFQ Restriction, and self-regulation of energy intake in toddlers. The main aim was to determine if the TFQ factor Allow Access would moderate the relationship between CFQ Restriction and ad libitum intake of snack foods, using an age appropriate adaptation of the Free Access Procedure developed by Fisher and Birch (1999a). The findings confirmed a moderating influence of Allow Access. Restriction was a significant predictor of intake only when Allow Access was high. The implications of this finding for the development of parental feeding advice and the conceptualisation of parental control are discussed.