Child body mass index, genotype and parenting in the prediction of restrictive feeding

Summary Background Restrictive feeding is implicated in pediatric obesity, and caregivers increase controlling feeding practices on the basis of higher child weight status. However, few studies have examined how child genetic and parenting characteristics together impact restrictive feeding. Objectives We examined whether child body mass index (BMI) status predicts caregiver use of restrictive feeding and if this association is moderated by (i) caregiver strategies to manage their children's distress and (ii) child variations in the catechol‐ O ‐methyltransferase (COMT) gene (Val 158 Met, rs4680). Methods Participants included 126 Caucasian children (50% girls) and their caregivers who were participating in a larger study in the USA. Caregivers reported on their feeding practices and responses to child distress when children were 2.5–3.5 years of age. Child anthropometric measurements were also obtained. Restrictive feeding was assessed again 1–1.5 years later. Genomic DNA was obtained from saliva samples, and COMT‐rs4680 was genotyped using TaqMan® methodology. Results Child BMI percentile predicted subsequent caregiver restrictive feeding for children who were Met/Met and who had caregivers reporting higher use of negative responses to child distress. For Val carriers, BMI percentile predicted restrictive feeding when caregivers were below the mean on these responses. Conclusions Caregivers are at risk for use of restrictive feeding practices when their children are at higher BMI percentiles, and this association increases when caregivers use more ineffective stress regulation practices and their children are homozygous for the Met allele. Prevention programmes might focus on parenting behaviours that foster emotion regulation and consider variation in child responses to parenting.