Abstract
OBJECTIVES: To examine associations between parental feeding behaviors and preschool children's willingness to try new foods (WTNF) and body mass index (BMI). METHODS: This study used baseline data collected from parents and children in childcare centers. Parents completed the Comprehensive Feeding Practices Questionnaire (CFPQ). We measured children's height and weight, calculated BMI z-scores (BMIz) and percentiles (%ile) using CDC criteria, and assessed WTNF by offering 6 novel and 3 familiar foods. Confirmatory factor analysis was conducted to examine the goodness of fit of the 8-subscales (child control, emotion regulation, involvement, modeling, monitoring, pressure to eat, restriction for health, and restriction for weight control) from the CFPQ. Logistic regressions was employed to examine the association between parents’ CFPQ scores and children's WTNF (dichotomized as high: tried ≥3 novel foods vs. low: tried <3 novel foods), and linear regressions to examine the association between parents’ CFPQ scores and child BMI, adjusting for child sex, age, race, full-time childcare center attendance [≥40 hours/week (hrs/wk)], and household income. RESULTS: The sample included 407 children (aged 48.2 ± 0.4 months) and their primary caregiver from 51 childcare centers. Majority of children were Caucasian (64%) and spent at least 40 hrs/wk at the childcare center (69%); 24% had BMI ≥85(th) %ile with mean BMIz of 0.36 (SD = 1.05). After dropping 3-items from the CFPQ due to low factor loadings (<0.4, n = 2) or theory (n = 1), the original 8-subscale structure demonstrated good model fit (RMSEA = 0.05, CFI = 0.97, TLI = 0.97, and SRMR = 0.07). Children of parents with elevated scores on the restriction for weight subscale had 1.53 (95% CI: 1.08–2.16) greater odds of high WTNF and 0.46 (95% CI: 0.30–0.63) increase in BMIz, compared to children of parents with low subscale scores. CONCLUSIONS: Children of parents who restrict foods in an attempt to control their child's weight are more likely to try new foods in a childcare setting without parents present, and are likely to be heavier. Longitudinal research is needed to determine the direction of the associations and how they relate to maternal and child characteristics (e.g., perceived or real child weight and eating behaviors). FUNDING SOURCES: National Institute of Diabetes and Digestive and Kidney Diseases.