An Infant Diet Score Based on Health Records Is Associated With BMI: A Nationwide Mother-Child Cohort Study in...

Nationwide health records provide innovative research opportunities. The aim of this study was to describe infant feeding in the first year of life amongst all infants in Iceland born January 2009 to June 2015 (N = 30,623), explore the feasibility of creating a composite infant diet score (IDS) for the period 0-12 months in a subset of infants with complete records from routine nutrition monitoring (n = 12,848), and examine its associations with BMI-for-age z-score (BMIz) at 12 and 18 months, maternal and birth characteristics. Recorded dietary variables included breastfeeding (> 90% at 2-3 weeks); exclusive breastfeeding (13% at 6 months); introduction of cow's milk (26% at 12 months); introduction of porridges, fruit/vegetables, meat, and fish (> 50% received ≥ one of those at 5 months); vitamin D supplements (95% at 12 months). The newly constructed IDS consisted of: duration of (i) exclusive and (ii) any breastfeeding; age of (iii) cow's milk and (iv) semi-solids/solids introduction; (v) an estimate of food variety; (vi) vitamin D supplement use. We used multiple linear and logistic regression adjusted for relevant factors to examine associations between infant feeding, birth and maternal factors, and BMIz > +2 (WHO Growth Standards, indicative of overweight/obesity). Compared with IDS quintile 5 (alignment with nutrition recommendations) IDS quintiles 1 and 2 had higher odds of overweight/obesity, at 12 months (aOR: 1.80; 95% CI: 1.20-2.74; aOR: 1.73; 95% CI: 1.16-2.64) and 18 months (aOR: 1.46; 95% CI: 1.04-2.04; aOR: 1.69; 95% CI: 1.22-2.34). Our novel IDS was associated with overweight/obesity and maternal and birth characteristics, supporting its value as a measure of infant nutrition quality.