A Bidirectional Analysis of Feeding Practices and Eating Behaviors in Parent/Child Dyads from Low-Income and Minority Households

Abstract

Objective: To prospectively examine the bidirectional relationship between parental feeding practices (eg, instrumental feeding, encouragement to eat) and child eating behaviors (eg, food responsiveness, emotional eating) in low-income, ethnically diverse preschool children over a 3-year period.

Study design: Parent/child (age 2-4 years at baseline) pairs (n = 222 non-Hispanics; n = 312 Hispanics) participated in NET-Works (Now Everybody Together for Amazing and Healthful Kids), a randomized controlled trial carried out in community and in-home settings in urban areas of Minnesota. Data were collected at baseline and 12, 24, and 36 months. The present study is a secondary data analysis using cross-lagged models to identify bidirectional associations between parental feeding practices and child eating behaviors.

Results: Three models showed significant cross-lagged effects (P < .05): model 1, parental instrumental feeding influencing later child food responsiveness; model 2, parental emotional feeding influencing later child food responsiveness; and model 3, parental emotional feeding influencing later child eating satiety. Model 1 showed significant bidirectional temporal paths, whereas models 2 and 3 showed significant unidirectional temporal paths from parental feeding practices to child eating behaviors.

Conclusions: Parental instrumental and emotional feeding practices prospectively influence child food responsiveness and satiety. This study demonstrates causal temporality between parental feeding practices and child eating behaviors. Heath care providers may want to use findings regarding parent feeding practices as part of their anticipatory guidance during well-child visits with parents of preschoolers.

Trial registration: ClinicalTrials.gov NCT01606891.

Keywords: Hispanic/Latino; bi-directional influences; child eating behaviors; parental feeding practices; preschoolers.

Figure 1.

Paths estimated in cross-lagged models.

Figure 2.

Cross-lagged model for parent instrumental feeding with child food responsiveness. ^aBolded paths are significant at P <.05. ^bModel adjusted for child age, child sex, child BMI, child race, household maximum education attainment, household income, BMI of the primary adult respondent, and treatment group assignment.

Figure 3.

Cross-lagged model for parent emotional feeding with child food satiety. ^aBolded paths are significant at P <.05. ^bModel adjusted for child age, child sex, child BMI, child race, household maximum education attainment, household income, BMI of the primary adult respondent, and treatment group assignment.

Figure 4.

Cross-lagged model for parent emotional feeding with child food responsiveness. ^aBolded paths are significant at P <.05. ^bModel adjusted for child age, child sex, child BMI, child race, household maximum education attainment, household income, BMI of the primary adult respondent, and treatment group assignment.

Figure 5.

Cross-lagged model for parent instrumental feeding with child food responsiveness. ^aBolded paths are significant at P <.05. ^bModel adjusted for child age, child sex, child BMI, child race, household maximum education attainment, household income, BMI of the primary adult respondent, and treatment group assignment.

Figure 6.

Cross-lagged model for parent emotional feeding with child satiety response. ^aBolded paths are significant at P <.05. ^bModel adjusted for child age, child sex, child BMI, child race, household maximum education attainment, household income, BMI of the primary adult respondent, and treatment group assignment.

Figure 7.

Cross-lagged model for parent emotional feeding with child food responsiveness. ^aBolded paths are significant at P <.05. ^bModel adjusted for child age, child sex, child BMI, child race, household maximum education attainment, household income, BMI of the primary adult respondent, and treatment group assignment.

Figure 8.

Cross-lagged model for parent controlling feeding with child food responsiveness. ^aBolded paths are significant at P <.05. ^bModel adjusted for child age, child sex, child BMI, child race, household maximum education attainment, household income, BMI of the primary adult respondent, and treatment group assignment.

Figure 9.

Cross-lagged model for parent encouraging feeding with child food responsiveness. ^aBolded paths are significant at P <.05. ^bModel adjusted for child age, child sex, child BMI, child race, household maximum education attainment, household income, BMI of the primary adult respondent, and treatment group assignment.

Figure 10.

Cross-lagged model for parent controlling feeding with child food fussiness. ^aBolded paths are significant at P <.05. ^bModel adjusted for child age, child sex, child BMI, child race, household maximum education attainment, household income, BMI of the primary adult respondent, and treatment group assignment.

Figure 11.

Cross-lagged model for parent emotional feeding with child food fussiness. ^aBolded paths are significant at P <.05. ^bModel adjusted for child age, child sex, child BMI, child race, household maximum education attainment, household income, BMI of the primary adult respondent, and treatment group assignment.

Figure 12.

Cross-lagged model for parent encouraging feeding with child food fussiness. ^aBolded paths are significant at P <.05. ^bModel adjusted for child age, child sex, child BMI, child race, household maximum education attainment, household income, BMI of the primary adult respondent, and treatment group assignment.

Figure 13.

Cross-lagged model for parent instrumental feeding with child food fussiness. ^aBolded paths are significant at P <.05. ^bModel adjusted for child age, child sex, child BMI, child race, household maximum education attainment, household income, BMI of the primary adult respondent, and treatment group assignment.

Figure 14.

Cross-lagged model for parent controlling feeding with child satiety response. ^aBolded paths are significant at P <.05. ^bModel adjusted for child age, child sex, child BMI, child race, household maximum education attainment, household income, BMI of the primary adult respondent, and treatment group assignment.

Figure 15.

Cross-lagged model for parent encouraging feeding with child satiety response. ^aBolded paths are significant at P <.05. ^bModel adjusted for child age, child sex, child BMI, child race, household maximum education attainment, household income, BMI of the primary adult respondent, and treatment group assignment.

Figure 16.

Cross-lagged model for parent instrumental feeding with child satiety response. ^aBolded paths are significant at P <.05. ^bModel adjusted for child age, child sex, child BMI, child race, household maximum education attainment, household income, BMI of the primary adult respondent, and treatment group assignment.

Figure 17.

Cross-lagged model for parent controlling feeding with child food enjoyment. ^aBolded paths are significant at P <.05. ^bModel adjusted for child age, child sex, child BMI, child race, household maximum education attainment, household income, BMI of the primary adult respondent, and treatment group assignment.

Figure 18.

Cross-lagged model for parent emotional feeding with child food enjoyment. ^aBolded paths are significant at P <.05. ^bModel adjusted for child age, child sex, child BMI, child race, household maximum education attainment, household income, BMI of the primary adult respondent, and treatment group assignment.

Figure 19.

Cross-lagged model for parent encouraging feeding with child food enjoyment. ^aBolded paths are significant at P <.05. ^bModel adjusted for child age, child sex, child BMI, child race, household maximum education attainment, household income, BMI of the primary adult respondent, and treatment group assignment.

Figure 20.

Cross-lagged model for parent instrumental feeding with child food enjoyment. ^aBolded paths are significant at P <.05. ^bModel adjusted for child age, child sex, child BMI, child race, household maximum education attainment, household income, BMI of the primary adult respondent, and treatment group assignment.