Bidirectional Associations between Parental Feeding Practices and Child Eating Behaviors in a Chinese Sample

Abstract

Background: Child eating behaviors (CEBs) and parental feeding practices (PFPs) play critical roles in childhood obesity. However, the bidirectional relationships between CEBs and PFPs remain equivocal. This longitudinal study aimed to explore their bidirectional relationships. Methods: A convenience sample of 870 parents with preschoolers was recruited in this longitudinal study (Shanghai, China). Three non-responsive feeding practices (NFPs), three responsive feeding practices (RFPs), five CEBs, and covariates were collected using validated questionnaires at baseline and the 6-month follow-up. Cross-lagged analyses using structural equation modeling (SEM) were performed to examine their bidirectional relationships. Results: Eight hundred and fifty-three parents completed questionnaires, with a response rate of 98%. The mean age of their children at baseline was 4.39 years (standard deviation = 0.72 years). Eighteen out of sixty longitudinal cross-lagged paths were statistically significant. Parental encouragement of healthy eating and content-restricted feeding were found to be bidirectionally associated with child food fussiness. Four parent-driven associations and one child-driven association were identified between RFPs and CEBs. For example, monitoring was negatively associated with children’s unhealthy eating habits (β = −0.066, standard error (SE) = 0.025, p < 0.01). Eight child-driven associations and one parent-driven association were observed between NFPs and CEBs. For example, higher child satiety responsiveness predicted a higher pressure to eat (β = 0.057, SE = 0.029, p < 0.01) and the use of food as a reward (β = 0.083, SE = 0.031, p < 0.01). Conclusions: There were bidirectional, parent-driven, and child-driven associations. Parents should be encouraged to adopt RFPs to shape CEBs. Increasing parents’ understanding of CEBs and providing them with reasonable coping strategies would help optimize PFPs.