The effect of breast-feeding on cognitive performance in a birth cohort

Breastfeeding is one of the most important health determinants in infancy and childhood: physical growth and mother-child bonding in this period are two factors that program cognitive ability in later life. The main aim of this study was to examine the long-term effects of breastfeeding on subsequent cognitive performance. A cohort of 5362 births registered over a period of one week in March, 1946 was selected by stratified sampling, based on the subject's social class. We collected data for ages 2, 8, 11, 15, 26, and 46 years, and these data were analyzed in 2004-2005 at the school of public health, Tehran. Linear structural equation models were fitted to the data by LISREL 8.12 software in order to find path models. Descriptive findings showed that children weighing ≥3500 grams at birth acquired the ability to walk, talk, stand, and sit earlier than others. Also, subgroups within the cohort showed different patterns of change in their cognitive scores. Analysis of variance revealed that after adjustment for sex, social class, age of talking, and age of walking breastfed children still had higher means of verbal and nonverbal scores at ages 8 and 15 compared to never-breastfed children. Path analysis indicated that breastfeeding and birth weight could program later-life cognitive performance directly or indirectly through the ages of 2, 8, 11, 15, 26, and 43 years. For instance, path coefficients from breastfeeding to ages of talking and walking, to 8-15 verbal scores, to 26-year cognitive scores, and then to memory and visual scores at age 43 were estimated -for females and males, respectively- and form left o the right as follows: -0.12 (se=0.02), -0.05 (se=0.02), 0.71 (se=.05), 0.10 (se=0.04) for women; -0.01 (se=0.02), -0.07 (se=0.01), 0.75 (se=0.06), 0.13 (se=0.06) for men. There are significant relationships between some early life determinants and later cognitive performance; we conclude that in longitudinal studies of later life abilities, early-life nutrition and birth weight should be incorporated in the path analysis of aging.