TCS Daily

Is Breast Best for Tests?

By Iain Murray - May 13, 2002 12:00 AM

Mentioning breastfeeding can elicit a variety of reactions. Some people are discomforted by women breastfeeding in public; others regard it as an important feminist statement. Others are unfazed. Still others resort to sophomoric giggling. Everyone, however, is agreed that breastfeeding is beneficial to children's health.

Now comes another study purporting to show a link between breastfeeding and intelligence. Parents who want their babies to do well at school will no doubt be influenced by this research, reported positively by major newspapers (such as The Washington Post). But placing too much reliance on IQ gain as the main reason to breastfeed would be a mistake.

The study in question, by Danish researchers at Copenhagen University Hospital and Americans at the Kinsey Institute, was certainly well designed. It featured large samples of respondents and controlled for a good number of potentially confounding factors. The parents' social status and education, single motherhood, mother's health (height, age, weight gain during pregnancy, and smoking habits), number of previous pregnancies, time in the womb, birth weight, birth length and measures of pregnancy and delivery complications were all taken into account. When the possible distortions caused by these aspects were discounted, the researchers found that the duration of breastfeeding was associated with significantly higher IQ scores. A baby breastfed for less than a month, for example, had an average IQ score of 99 in young adulthood, while a baby breastfed for 7-9 months had a score of 106. The evidence seems pretty conclusive at first glance.

IQ, however, is not that useful a measure. The basic concept behind it (mental age) has fallen into disrepute, and what is left is a narrow measure of a very specific set of skills. Psychologists generally agree now that intelligence, as measured by IQ tests, is merely one factor in deciding whether a person is gifted or retarded. Intelligence has also been shown to be malleable. Two identical twins raised separately, for instance, may score very differently on IQ tests. Although about 70 percent of intelligence depends on heritability, much depends on factors beyond infancy. The researchers did not control for these factors beyond the controls already mentioned for parental education and socioeconomic status.

Furthermore, IQ is distributed normally, i.e. its distribution follows a bell curve pattern. About 70 percent of people have an IQ between the ranges of 85-115. The difference between an IQ of 99 and one of 106 is simply not that great. While the difference might be statistically significant, in practice it means little. A likely IQ of 116 for breastfed babies would be something to get excited about, but this small deviation is unlikely to have much practical difference in test scores and future achievement.

A third point we must bear in mind is that the researchers admit that they did not control for everything that might affect IQ. They hypothesize three possible reasons for the effect they found. Only one of those -- the possible effects of fatty acids in breast milk on the developing brain -- has any direct connection to the act of breastfeeding. A second explanation, that the mother and child develop a special, beneficial bond during breastfeeding, is partially connected to breastfeeding but does not have a connection to the milk. The third suggestion, that breastfeeding is in itself a by-product of the amount of beneficial time and attention devoted by the mother to the baby would imply that breastfeeding itself is irrelevant. The researchers therefore essentially admit that they have not necessarily found any causal link between breastfeeding and higher IQs.

Finally, there is some interesting evidence from a 1996 study1 that looked at the IQs of 994 men and women born between 1920 and 1930. At that time, of course, breastfeeding was in no way associated with higher socioeconomic status (wet nurses aside, probably the reverse). The researchers controlled for various parental effects but found no association between adult IQ and the method of feeding.

Breastfeeding is good for babies in many ways. Its effect on morbidity and mortality rates is demonstrably positive all over the world. It benefits the baby's immune system and contributes to healthy growth rates. It appears to be especially beneficial to premature babies and to children born with certain birth defects. It also provides significant health, weight and contraceptive benefits to the mother. There are plenty of reasons for women to breastfeed their babies. Possible marginal increases in IQ do not need to be added to the list.

1. Gale CR, Martyn CN. Breastfeeding, dummy use, and adult intelligence. Lancet 1996; 347: 1072-5.

Iain Murray is Director of Research at STATS- the Statistical Assessment Service - in Washington DC. He was bottle-fed. His daughter was breastfed for over a year.

TCS Daily Archives