INSIGHT Study Parenting Intervention to Prevent Childhood Obesity Improves Patterns of Dietary Exposures in Infants

Composition of the infant diet changes rapidly throughout the first year of life, and caregiver decisions on when to introduce foods and what to feed an infant vary. Early dietary patterns have been associated with differing risk for obesity as well as food preferences and intake later in childhood. The objectives of this analysis were to 1) identify patterns of milk and complementary feeding among 40 week old infants participating in the INSIGHT study, an ongoing randomized, controlled trial designed to prevent childhood obesity, and 2) determine whether infants in the parenting intervention group consumed a wider variety of age-appropriate, nutrient dense foods than control infants. Primiparous mother-newborn dyads (n=291) were randomized to parenting intervention or a safety control group 2 weeks after birth. The parenting curriculum, built around a framework of responsive parenting in the domains of feeding, sleep, emotional/social regulation, and active social play, or a child safety control curriculum, was delivered at nurse home visits at 3, 16, 28, and 40 weeks after birth. Feeding guidance included messaging on responsive feeding, delayed introduction of solids, repeated exposure to novel foods, and age-appropriate foods and portion sizes. Infant dietary intake was assessed by food frequency questionnaire at 40 weeks. Latent class analysis was used to identify patterns of dietary exposure encompassing both milk feeding type (formula vs. breast) and solid foods at 40 weeks. Dietary pattern class membership at 40 weeks was used to predict child weight outcomes measured at 1 year. Similar to previous work, five distinct patterns of dietary exposures were identified: 1) “Breastfed, Fruits and Vegetables (BFV)”, 2) “Breastfed, Low Variety (BLV)”, 3) “Formula, Fruits and Vegetables (FFV)”, 4) “Formula, Low Variety (FLV)”, and 5) “Formula, High Energy Density (FHED)”. Class membership probabilities differed between the control and parenting groups (p=0.002). A greater percentage of infants in the parenting group than control were predicted to be in the BFV and FFV classes (60.5% vs 38.5%, p=0.0004), classes characterized by higher fruit and vegetable intake and low intake of high energy-dense foods. Among formula fed infants, a greater percentage of infants in the parenting group than control were predicted to be in the FFV class rather than the FLV or FHED classes (47.5% vs 19.0% of formula fed infants, p=0.0001). At 1 year, children in the FLV class had significantly higher average weight-for-length percentile than those in the BFV class (p=0.04). These results suggest that the parenting intervention improved patterns of dietary intake in 40 week-old infants, particularly those being fed formula, by increasing fruit and vegetable intake, which may reduce later risk of obesity. Support or Funding Information NIDDK R01DK088244 and The Childhood Obesity Training Grant #2011-67001-30117 from USDA