242 Duration of Breastfeeding and Childhood Obesity in Australian Children: A Prospective Cohort Study

Abstract

Childhood obesity is increasing in prevalence and leads to morbidity and mortality in later life. Breastfeeding has been suggested to be protective, with an estimated population attributable risk from formula feeding of 15–20%1. We examined associations between breastfeeding and body mass index BMI in a prospective cohort of 2087 Australian children entering the study at 16 to 18 weeks of gestation.Methods: Characteristics of 2,979 mothers recruited from Perth antenatal clinics between 1989 and 1992 were recorded during pregnancy and 8 years later. Children's weight and length or height were measured at birth, 1, 3, 5 and 8 years; information about breastfeeding was collected at birth, 1 and 3 years. We used NCHS standards to define overweight or obesity in 1 year-olds and the standards of Cole et al2 at 3, 5 and 8 years. Analysis excluded children born before 37 weeks gestation or with congenital abnormalities. Associations between obesity and breastfeeding duration were examined in linear and logistic regression.Findings: Surveys included 1710, 1184, 1480 and 1430 children at 1, 3, 5 and 8 years, respectively. At 8 years 19.6% of girls and 15.4% of boys were overweight or obese. Maternal obesity and smoking during pregnancy were each associated with significantly shorter breastfeeding duration; tertiary education was significantly related to longer breastfeeding duration. In multivariate regression, maternal BMI, education, smoking and parity were significant simultaneous predictors of breastfeeding duration. Breastfeeding duration was inversely related to children's BMI at 1 year (coefficient −0.031,95% CL −0.042,−0.020;p<001) with adjustment for sex, birthweight, gestational age and ethnicity, and remained significant after adjustment for maternal BMI, smoking status, age, parity and education (−0.028,95%CL −0.040,−0.016;p<0.001). Breastfeeding duration was unrelated to BMI in 3 year-olds but was a significant predictor at 5 and 8 years, although not independent of maternal factors. In logistic regression, risk of overweight or obesity at 1 year was lower with breastfeeding for at least 12 months, relative to never breastfeeding, before (OR=0.48,95% CI 0.31,0.76) or after (OR=0.37,95% CI 0.30,0.77) adjustment for maternal factors; associations were not statistically significant at 3, 5, or 8 years.Conclusion: Beyond infancy, genetic factors and health-related behaviours within families may be more influential determinants of BMI than breastfeeding history.