Bidirectional association between parental child-feeding practices and body mass index at 4 and 7 y of age

Am J Clin Nutr. 2016 Mar;103(3):861-7. doi: 10.3945/ajcn.115.120824. Epub 2016 Feb 3.


Background: Evidence of the association between parental child-feeding practices and the child's body mass index (BMI) is controversial, and bidirectional effects have been poorly studied.

Objective: We aimed to examine bidirectional associations between parental child-feeding practices and BMI at 4 and 7 y of age.

Design: This study included 3708 singleton children from the Generation XXI birth cohort with data on parental child-feeding practices and BMI at 4 and 7 y old. Feeding practices were assessed through a self-administered questionnaire by combining the Child Feeding Questionnaire and the Overt/Covert Control scale and then adapting it to Portuguese preschool children. Weight and height were measured according to standardized procedures, and age- and sex-specific BMI z scores were computed based on the WHO Growth References. Linear regression models were used to estimate the bidirectional associations between each practice and BMI z score. Crosslagged analyses were performed to compare the directions of those associations (the mean score of each practice and BMI z score at both ages were standardized to enable effect size comparisons).

Results: After adjustments, pressure to eat and overt control at 4 y of age were associated with a lower BMI z score 3 y later (β: -0.05; 95% CI: -0.08, -0.03 and β: -0.05; 95% CI: -0.09, -0.01, respectively). Regarding the opposite direction of association, a higher BMI z score at 4 y of age was significantly associated with higher levels of restriction and covert control at 7 y of age (β: 0.06; 95% CI: 0.03, 0.08 and β: 0.06; 95% CI: 0.04, 0.08, respectively) and with lower levels of pressure to eat (β: -0.17; 95% CI: -0.20, -0.15). The only bidirectional practice, pressure to eat, was more strongly influenced by the BMI z score than the reverse (βstandardized: -0.17 compared with βstandardized: -0.04; likelihood ratio test: P < 0.001).

Conclusions: We found that parents both respond to and influence the child's weight; thus, this child-parent interaction should be considered in future research.

Keywords: body mass index; children; cohort studies; feeding behaviors; longitudinal studies.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Body Mass Index*
  • Body Weight*
  • Child
  • Child, Preschool
  • Feeding Behavior*
  • Female
  • Humans
  • Male
  • Parent-Child Relations*
  • Parenting*
  • Parents*
  • Pediatric Obesity / etiology*
  • Surveys and Questionnaires