Abstract
Background: This study examined the associations between duration of breast feeding, early infant growth, and body mass index (BMI) at 30 and 35 years, in a birth cohort studied to age 35. Methods: Data were gathered on duration of exclusive and non-exclusive breast feeding (months), early growth (kg; 0-9 months), and BMI at ages 30 and 35 from the Christchurch Health and Development Study. The Christchurch Health and Development Study is a study of a birth cohort of 1265 children, born in Christchurch in 1977. Results: Population-averaged generalised estimating regression models showed statistically significant associations between: duration of breast feeding and mean BMI; and early growth and mean BMI. After adjustment for perinatal, family, and social background factors, statistically significant associations were found between: longer duration of breast feeding and lower adult BMI (B = -0.424 [95% confidence interval (CI) -0.708, -0.140]); and increasing early growth and higher adult BMI (B = 0.393 [95% CI 0.080, 0.707]). When breast feeding and infant growth were entered into the regression model and adjusted for covariates, breast feeding was no longer statistically significantly associated with BMI (B = -0.250 [95% CI -0.553, 0.054]), while early growth remained statistically significantly associated with BMI (B = 0.355 [95% CI 0.039, 0.671]). A test for mediation showed that the association between breast feeding and BMI was mediated by early growth (P = 0.01). Conclusions: The association between longer duration of breast feeding and later lower BMI scores in adulthood was mediated by lower early growth. Breast feeding may be included as one component of multicompartment programmes targeted at early growth and later obesity.