Associations between maternal nutrition among women with hyperglycaemia in pregnancy and childhood anthropometry: The PANDORA cohort.

OBJECTIVE: The aim of this study was to describe the associations of nutrition during pregnancy with childhood anthropometry among First Nations and non-First Nations women with hyperglycaemia in pregnancy. METHODS: 224 women (58% First Nations women) with hyperglycaemia in pregnancy (168 gestational diabetes, 56 pre-existing type 2 diabetes) were administered food-frequency questionnaires during the third trimester. Childhood anthropometry body mass index, waist and arm circumferences and skinfold thickness) was measured at median (range) follow-up of 2.7 (1.5-5.0) years of age. Multivariable linear regression assessed associations of maternal nutrition with children's anthropometry after adjusting for age, sex, diabetes type, maternal body mass index, ethnicity, socioeconomic factors, birthweight, and breastfeeding. RESULTS: On multivariable regression, greater maternal frequency of fast-food (two or more times a week vs. weekly or less) was associated with higher body mass index (ß-coefficient 0.6 kg/m2, p=0.03), waist circumference (ß-coefficient 2.2 cm, p<0.01), arm circumference (ß-coefficient 0.7 cm, p<0.01) and sum of skinfolds (ß-coefficient 4.0 mm, p<0.01) in children. There were no associations between consumption of sugar-sweetened beverages, fruit or vegetable intake and outcomes. CONCLUSIONS: Addressing factors that support women and their families to maintain healthy diets is likely essential for improving childhood anthropometric outcomes. IMPLICATIONS FOR PUBLIC HEALTH: System-level changes focusing on food environments are essential to shift dietary patterns during pregnancy.