Mother holding and breastfeeding a child. “Untitled” by StockSnap Licensed by Pixaby

Reprinted from Council on Contemporary Families Brief Reports published on May 31, 2023

Recent infant formula shortages in the United States have reignited debates about the importance of breastfeeding for child health and development. Major health and medical organizations unanimously recommend that mothers exclusively breastfeed their infants for six months or more, and breastfeeding has become a hallmark of ideal parenting. Children who are breastfed generally have lower rates of child obesity, fewer behavior and conduct disorders in childhood, and higher cognitive test scores compared to children who are not breastfed.

Although over 80% of infants receive some breastmilk, only 26% are exclusively breastfed for the recommended six months. This is partially because breastfeeding is time-intensive and requires a lot of support. As a result, breastfeeding is socially stratified in the United States and is most common among those with the most resources. For example, women with a college degree are twice as likely to exclusively breastfeed for six months compared to those with a high school degree.

Public health initiatives suggest that increasing breastfeeding rates among populations that are less likely to breastfeed is an important strategy for improving children’s health and development. But a growing body of research suggests that benefits of breastfeeding may have been overstated. We cannot conduct randomized controlled trials for breastfeeding, so it’s difficult to tell whether it is directly causing improvements in children’s health, or if breastfed children tend to have better outcomes because of their parent’s other socioeconomic advantages. Studies using sophisticated statistical techniques to account for these complicated dynamics find that, all else equal, the causal effect of breastfeeding is weak or nonexistent. For example, when we compare siblings in the same family who were fed differently (one sibling was breastfed and the other was not, or they were breastfed for different amounts of time), they tend to have very similar physical, behavioral, and cognitive development.

The lack of an average causal benefit of breastfeeding, however, does not mean that it confers no advantages.  Averages can obscure substantial variation, particularly for behaviors like breastfeeding that are highly influenced by social and economic factors. In such cases, asking if breast is best is too simplistic a question. Better questions to ask are: Does breastfeeding benefit all children equally? Or do some children reap greater benefits than others?

I conducted a study with my co-authors Kerri Raissian and Jiyeon Kim to answer these questions. Our study analyzed data from a longitudinal nationally representative sample to evaluate the relationship between breastfeeding and a comprehensive set of longer-term developmental outcomes for children ages 4-14. Specifically, we examined body mass index (BMI), three measures of behavioral development, and five measures of cognitive skills. We used statistical techniques to carefully compare children who were breastfed to children who were never breastfed but had very similar characteristics. We also grouped children based on their likelihood of breastfeeding and compared the effects across these different groups to assess whether everyone benefitted equally. We found that breastfeeding provides small benefits for children’s behavioral development, math scores, and academic ability, but only among those that are highly likely to breastfeed. We also found small benefits for children’s reading comprehension and vocabulary, but these effects were concentrated among those who were least likely to breastfeed.

One key takeaway from our study is that breastfeeding does not equally benefit everyone. What explains this variation?

One potential answer is that people experience the costs of breastfeeding differently. Although breastmilk is “free,” breastfeeding is not costless in American society. Breastfeeding is associated with prolonged earnings losses for women, in part due to lack of paid parental leave and time out of the workforce. This economic penalty may be more easily absorbed by women who are financially secure but is a more significant hardship for low-income or single women. Lack of economic resources may offset some of the benefits of breastfeeding for these children who are less likely to breastfeed. This could explain why we found that breastfeeding is related to better behavior, math scores, and academic ability among those who are most likely to breastfeed.

Another potential source of variation is that people experience the benefits of breastfeeding differently. If more advantaged parents are already providing an optimal home environment or parenting in ways that promote their children’s health or cognitive skills, they may not receive an additional boost from breastfeeding. For example, the American Academy of Pediatrics recommends that all parents read to their children from birth through kindergarten, and it is possible that parents who adopt this practice are also likely to follow the AAP recommendations for breastfeeding. If parents who are less likely to breastfeed do not engage in these early literacy activities to the same extent, the benefits of breastfeeding could be amplified among this group. This might explain why we found that breastfeeding is linked with better reading comprehension and vocabulary among children who are least likely to breastfeed. Parents may be less equipped to compensate for quantitative skills given that we did not find this pattern for math scores.

The second key takeaway is that the benefits of breastfeeding are modest. For example, among children who were least likely to breastfeed, those who breastfed for 6 months or more had about 30% of a standard deviation higher reading comprehension score than those who were never breastfed. In comparison, four extra days of mother-child shared reading time per month increases reading achievement by 32% of a standard deviation. On a broader scale, expansions in Universal Pre-K are linked with larger increases in letter-word identification (79% of a standard deviation) and spelling scores (64% of a standard deviation).

To be clear, we are not saying that breastfeeding isn’t an ideal source of nutrition for infants, or that other parenting practices or policies are direct substitutes. Rather, parents and policymakers should consider breastfeeding as one potential tool among others that may provide comparable or larger benefits for longer-term child development. This is especially useful information given the very real tradeoffs that many mothers make. It may also help to relieve some of the stigma and shame that some mothers experience when they are unable to breastfeed.

Our study provides insights for public policy. We highlight socioeconomic and racial inequality in breastfeeding within the United States. The US provides a weak safety net for parents of young children and offers no federal paid family leave. Factors such as race/ethnicity, socioeconomic status, education, income, marital status, and employment are strongly related to the likelihood that a mother will breastfeed.

Furthermore, we demonstrate that the social context not only shapes who breastfeeds in American society, but also who benefits most. Efforts to increase breastfeeding rates among populations that are least likely to breastfeed are unlikely to close disparities in child development. The benefits of breastfeeding are largely concentrated among the populations that are already likely to breastfeed. To level the playing field, rather than merely telling mothers that “breast is best,” policies should focus on reducing structural barriers and economic costs for mothers who want to breastfeed. For many mothers in the United States, breastfeeding is ideal but unattainable.

Jessica Houston Su is an Assistant Professor in the Department of Sociology at the University of North Carolina-Chapel Hill. You can follow them on Twitter: @jesshoustonsu