There is extensive evidence that breastfed children score higher on cognitive testing than those not breastfed, even after controlling for socioeconomic variables such as parental income, education, and occupation, but the research is still criticized for insufficient control of these confounders.

The UK Millennium Cohort Study (MCS) was designed to evaluate how much of the association between breastfeeding duration and child cognitive development is due to socio-economic factors and maternal cognitive scores. The MCS study recruited 18,818 children in the UK in 2000-2002 at approximately 9 months of age, and followed them until age 17, with cognitive evaluations at ages 3, 5, 7, 11, 14, and 17 years. All children were singletons born >37 weeks gestation.

Breastfeeding duration was grouped as ‘never BF’; <2 months; ≥ 2 and < 4 months; ≥ 4 months and < 6 months; ≥ 6 and < 12 months; ≥ 12 months.

The following confounders were measured among the mothers: social class, as measured by occupation; maternal education; gestational age at birth, maternal ethnicity; languages spoken in household; maternal partnership status; mother working outside the home; birth by cesarean; older siblings in the home; maternal age; alcohol and/or tobacco use during pregnancy; and maternal cognitive ability as measured by verbal ability testing.

Within the study population, 90% of the mothers were white, 25.9% had higher education, 47.5 had lower education, 44.2% were in the highest social class based on occupation, 57.3% were married, 64.3% never smoked, and 7.3% had heavy alcohol use in pregnancy.

It was clear that breastfed children scored higher on cognitive verbal and spatial scores, with higher scores associated with longer breastfeeding.

However, the researchers also found that children who breastfed longer were more likely to have a more educated mother from a higher social class who was non-smoking, married, older, and had higher cognitive scores.