Breastfeeding and IQ: Evaluation of the socio‐environmental confounders

Numerous papers (1–6), including a recent metaanalysis (7), have contributed to the growing literature investigating whether breastfeeding can enhance child cognitive development, as assessed by IQ. Virtually all the Ž ndings to date concur in showing an association between better performance on IQ tests in children who were breastfed over those who were bottlefed. This intellectual advantage has been demonstrated in several cultures for full terms and more strongly for preterm and low birthweight infants. The IQ advantage for full terms is small, roughly about 3–4 points, but it is consistent across studies. The controversy in this literature is not about whether there is an enhanced IQ score among the breastfed children, but whether this difference re ects a direct nutritional advantage or a difference in socioenvironmental factors that are generally more optimal among women who breastfeed for an extended period of time. It is well established that women who breastfeed also often provide a more enriched and cognitively stimulating environment for the child, which could be responsible for the enhanced IQ scores. Most studies, including that of Rao et al. and our own (8), have shown that breastfeeding is associated with higher socioeconomic status and education. Thus, it is critical to determine whether the observed IQ advantage persists after control for the socio-environmental factors that may be the true cause of enhanced intellectual ability among breastfed children. The new study by Rao et al. (9) contributes to the literature by comparing the effects of exclusive breastfeeding on cognitive and motor performance in fullterm, small-for-gestational-age (SGA) and appropriatefor-gestational-age (AGA) children and by examining whether there is a minimum duration of breastfeeding (threshold) necessary for the breastfeeding advantage to become apparent. The authors distinguish between the impact of short (μ12 wk) and longer exclusive breastfeeding (>12 wk). This is an important distinction, because many mothers of SGA infants provide supplemental foods signiŽ cantly earlier in the Ž rst year than mothers of AGA infants if their infants are still small after several weeks of breastfeeding. Rao et al. detected no statistically signiŽ cant differences between breastfed and bottlefed infants on the Bayley Scales of Infant Development (10) administered at 13 mo. However, at 5 y, SGA children who had been breastfed for >12 wk had a 4–5 point higher total IQ score on the Norwegian version of the Wechsler Preschool and Primary Scales of Intelligence (WPSSI) (11), after adjusting for confounders. This difference appears to re ect an advantage on Performance IQ, since there was no signiŽ cant difference in relation to Verbal IQ. No apparent beneŽ t on IQ was detected for the SGA children who were breastfed for only 12 wk. Duration of breastfeeding did not impact on growth, suggesting that the Performance IQ advantage was not mediated by size. The question, therefore, is to what extent and how deŽ nitively the Performance IQ advantage can be attributed to nutrition. The nutritional beneŽ ts of breastfeeding have been demonstrated for preterm infants in a clinical trial (2). Rao et al. extend this Ž nding by showing that there is also a beneŽ t for full-term SGA infants if they are breastfed >12 wk. As they and others have suggested, certain nutrients in breast milk may enhance intellectual development or may be especially in uential, given metabolic differences between AGA and very low birthweight SGA infants (12, 13). Evidence from clinical trials with formula-fed preterm infants has demonstrated beneŽ cial effects on visual recognition memory and visual acuity from dietary supplementation with docosahexaenoic acid (DHA), a long-chain polyunsaturated fatty acid that is an important nutrient in breast milk (14). As in most studies, the Rao et al. data show a signiŽ cant positive association of breastfeeding with social class and education. Most studies have therefore included these variables as potential confounders and have reported a statistical advantage for breastfeeding after control for these two distal socio-environmental factors. In our research, the association between breastfeeding and IQ (b = 0.21, p < 0.001) also persisted after inclusion of social class and maternal education (3, 4). We found, however, that this association was substantially weakened (from b = 0.12, p < 0.025, to b = 0.06, n.s.) when we included maternal verbal IQ, assessed on the Peabody Picture Vocabulary Test-Revised (PPVT-R) (15), and quality of parenting, assessed on the Home Observation for Measurement of the Environment (HOME) (16). We therefore concluded that these socio-environmental in uences were largely responsible for the enhanced IQ scores seen in