If you’ve had kids—or are considering having them—you’ve probably heard the mantra “breast is best.” A majority of doctors, along with every major medical group in the US, recommends that moms breastfeed babies exclusively for at least the first six months of their lives. And most major studies show that there are substantive physical and socio-emotional benefits to breastfeeding, for both infants and mothers (paywall).
But a study published in the August 2018 issue of Social Science & Medicine: Population Health challenges that conventional wisdom, arguing that major papers overestimate the benefits of breastfeeding because they fail to take into account the fact that US mothers who breastfeed tend to be richer and better-educated than mothers who formula-feed. The authors further argue that a mother’s socio-economic advantage, measured through her intention to breastfeed, is associated with the same health benefits for her baby as actually breastfeeding.
As a reporter who focuses on early childhood development, I was immediately intrigued by this finding. At first, I thought I might write a quick story about it. But the more I talked to experts in the field, the more I realized that evaluating the study was no simple task. For a variety of reasons, not least of which are the social pressures faced by new moms, breastfeeding is an extremely emotionally charged topic. And the debate among researchers about its benefits is more complicated than I imagined.
The socioeconomics of breastfeeding
According to the Center for Disease Control and Prevention, 83% of US children born in 2015 were breastfed. That doesn’t mean they were breastfed exclusively for six months, as many medical guidelines recommend; Many mothers may have breastfed for less time, or supplemented breast milk with formula. But it does mean that a large majority of US mothers breastfeed their children at some point. And many mothers think it is important. A survey of US women (pdf) conducted in 2000 showed that the main motivation behind their decision to breastfeed is the belief that it is better for their child’s health.
But a 2012 CDC study showed that roughly two-thirds of mothers who intend to exclusively breastfeed their babies for the first three months do not end up doing so, for various reasons—some because they weren’t taught how; some because of lactational difficulties; some because of demands involving work or school.
The researchers behind the Population Health study, Kerri Raissian and Jessica Houston Su, wanted to find out what the health outcomes were for children of moms who had intended to breastfeed, but didn’t. They looked at whether those children wound up being less healthy than their peers who were breastfed, as measured by number of ear infections, a type of lung and respiratory tract infection known as respiratory syncytial viruses, and antibiotic usage in the infant’s first year of life.
The study gathered data on 1,008 mothers from their last trimester of pregnancy through the first year of their infant’s life, taken from the Infant Feeding Practices Study II, a two-year longitudinal study conducted by the Food and Drug Administration and the Centers for Disease Control and Prevention. After controlling for maternal intent to breastfeed, the researchers found no difference in the health outcomes in the two groups. However, the study also found that mothers’ intention to breastfeed was strongly linked with infant health outcomes, irrespective of whether the child was actually breastfed. “This suggests that most physical health benefits associated with breastfeeding are likely attributable to demographic characteristics such as race and socioeconomic status, and other difficult to measure unobservable characteristics,” the authors write.
In other words, a mother’s intention to breastfeed serves as a proxy for her privilege, information about breastfeeding, and access to help. As Su told the University of Connecticut, “What we found is that intending mothers had more information about nutrition and diet; they more frequently consulted their physicians; and had better access to information related to infant health than those moms who did not intend to breastfeed.”
The researchers are quick to say that they’re not trying to overhaul decades of maternal health research, nor are they trying to discourage women from breastfeeding. “Breastfeeding is an excellent nutritional source,” Raissian told Quartz, “and if mothers want to pursue that they should. But the most likely alternative, formula, also seems to be an excellent source of nutrition, and mothers should do what’s right for them.”
Raissian and Su’s theory that breastfeeding may be serving as a proxy for the socio-economic advantages of mothers is not brand-new. We already know that breastfeeding mothers in the US are more likely to be well-educated, white, married, and have higher income than mothers who don’t breastfeed. As Raissian writes in her study, “It is possible that these sociodemographic advantages are related to both successful breastfeeding and better infant health outcomes.”
Quartz’s Corinne Purtill and Dan Kopf have calculated that the monetary value of the time spent breastfeeding in the first six months for a woman making $60,000 a year pre-tax and working 50 hours per week—based on the average number and duration of daily feeds—is $14,250. They write, “The children of well-off, well-educated mothers are far more likely to reach that six-month target—and to be breastfed at all—than babies born into less-privileged homes.”
Challenging conventional wisdom
Challenging the conventional “breast is best” wisdom is not for the faint of heart. As Raissian, who co-authored the study, says, “It’s much harder to be challenging an orthodoxy, especially one that’s becoming so entrenched.”
Indeed, critics have been quick to highlight the study’s limitations—from its methodology to its assumptions and conclusions.
Quartz spoke to four outside experts in epidemiology, statistics, and social demography. They disagreed among themselves about the study’s basic premise—that breastfeeding is a socio-economic marker, and therefore that breastfeeding studies that don’t take this into account are biased by nature. They also disagreed with one another about the conclusion that breastfeeding may not be as impactful as we have been led to believe. What they all agreed upon, however, was that the Population Health study had serious limitations.
One concern that many cited was the small sample size of about 1,000 expectant mothers. By comparison, the hallmark study of the field, the PROBIT trial, had a sample size of 17,000 mothers in Belarus. Apart from the fact that larger studies yield more reliable results, critics also point out that the 1,000 expectant mothers are not nationally representative, which makes generalizing the findings difficult. Raissian and Su acknowledge this in their study, writing that their sample of breastfeeding mothers was “perhaps still slightly more advantaged” than other nationally representative samples used in medical guidelines produced by the American Academy of Pediatrics.
Critics also point out that the study measures infant health through only three factors, all of which are self-reported by the mothers. This raises the possibility of inaccurate reporting for things like antibiotic use or ear infection prevalence.
Rebecca Goldin, a statistician and professor of mathematics at George Mason University who leads STATS, an organization dedicated to helping journalists interpret scientific studies, says that “the self-reporting is problematic,” because it fails to measure “whether there are differences among the moms in self-reporting behavior that are related to whether they actually breastfed or not.” Goldin characterizes the study as “exploratory rather than conclusive.”
Wilma Otten and Caren Lanting, both researchers in health behavior at TNO, a Dutch research organization, agree with Goldin. They criticized the choice of using mother-reported health information as a variable for infant health, especially for infections: “Infections can be unnoticed,” they explained.
Raissian says this doesn’t change her and her co-author’s interpretation of the findings, because “we are more interested in health differences [between breastfed and non-breastfed children], rather than trying to calculate the actual number of health events [like ear infections].”
That does not account for the fact that the mothers of children who were not breastfed might be more inclined to avoid reporting health problems, given the stigma attached to women who do not breastfeed. But Raissian says she isn’t worried about the risk of under-reporting: “It’s not a huge concern of mine, though I can appreciate why it might be a concern for someone,” she explained. “As the child gets older, moms become more at peace with the reality of whether they were able to breastfeed or not and so I think, if there is a reporting bias, it certainly lessens over time, and we’re looking throughout the infant’s first year of life.”
Another issue, according to critics, is that the authors of the new breastfeeding study do not fully acknowledge the protective properties of human milk on an infant’s immune system. Dozens of studies show that human milk is uniquely good for infants’ healthy development.
That conclusion isn’t necessarily incompatible with the findings of the new study, but it’s possible that Raissaian and Su would have found different outcomes if they had looked at a different set of infant health indicators. For example, Goldin notes, the PROBIT trial in Belarus considered gastrointestinal tract infections, respiratory tract infections, and eczema.
While the PROBIT trial is highly regarded, Raissian said she has doubts about its applicability to US mothers. “The Belarus study is carefully done, and I think, generalizable to Belarus; but I have very serious concerns about it being generalizable to an American context,” she said. She suggests formula in Belarus may be worse for babies because drinking water there is historically of poor quality. As Raissian explains, “formula is only as good as the drinking water that goes into it.” (Rafael Pérez-Escamilla, a professor of epidemiology and public health at the Yale School of Public Health, clarifies that, while Belarus drinking water is contaminated today, it was not when the PROBIT study was conducted, which is one of the reasons why the authors chose the country.)
Indeed, in developing countries, formula is often unaffordable, inaccessible, and poorly regulated, with potentially harmful consequences for babies’ health. As Annalisa Merrelli writes for Quartz:
When breastfeeding mothers feed their babies exclusively with formula, they quickly stop producing breast milk, making it impossible to revert back. This makes formula particularly problematic for poor mothers, who may not be able to buy sufficient amounts of the product, and end up watering it down or feeding the child smaller quantities, which then leads to malnourishment.
In this context, breastfeeding has been shown to save lives. The World Health Organization estimates that “nearly half of all diarrheal diseases and one-third of all respiratory infections in children in low- and middle-income countries could be prevented with increased rates of breastfeeding.”
To that end, a final and common criticism of the Raissian-Su paper is that it only focuses on the US, where it is well-established that wealthier and better-educated women are more likely to breastfeed, and neglects to consider the benefits of breastfeeding in the developing world.
Other researchers pointed to the fact that breastfeeding isn’t just good for infant health. In both developed and developing countries, breastfeeding has been shown to be good for moms (paywall), reducing their relative risk of breast cancer, Type 2 diabetes, and rheumatoid arthritis. That means there may be very good reasons for women to keep choosing to breastfeed, even if it turns out that Raissian and Su are correct in saying that some of the benefits attributed to breastfeeding are inflated.
Pérez-Escamilla says the evidence pointing to the benefits of breastfeeding for both mothers and children cannot be overlooked. “At the end of the day, public health policy around breastfeeding should be guided by a careful and systematic examination of the bulk of the evidence, giving more weight to studies with the most robust designs.” The Raissian-Su study, he says, “fails to acknowledge dozens if not hundreds of much more robust studies that don’t agree with their interpretation of findings.”
The push to stop shaming new moms
Otten and Lanting, the researchers at TNO, acknowledge that any research that could seem to contradict breastfeeding dogma can be fraught. “Sometimes, it seems like the proponents of breastfeeding are afraid that, if you say anything negative about breastfeeding, women are not going to be willing to try it anymore,” they said.
But science is not meant to accept wisdom. It is meant to challenge it—which means there’s room for further study.
Since ethical and logistical barriers typically inhibit breastfeeding researchers from conducting randomized control trials, the gold standard in any kind of research, another way they can get around the selection problem is by comparing siblings who were breastfed to their siblings who were not, because a mother’s advantage would be the same in that situation. As Purtill and Kopf write:
The largest of these studies (pdf), conducted by researchers at Ohio State University, did not find meaningful positive effects of breastfeeding. Using data from an annual survey of American households, the researchers examined health, behavioral, and academic outcomes of siblings between the ages of four and 14, from nearly 700 families where at least one child wasn’t breastfed and one of the others was. They looked at 11 measures of child wellbeing, and found essentially no discernible difference between the breastfed and non-breastfed.
Breastfeeding requires a huge time and financial investment that many women make because they believe they are giving their child the best possible start to life. But as Raissian and Su write in their study, women are often shamed if they choose not to, or find that they can’t, breastfeed: “The ‘breast is best’ message has been so deeply internalized that failure to meet breastfeeding recommendations makes many mothers feel inadequate, placing them at increased risk for maternal depression.”
Otten and Lanting agree: “Maybe the message should not be so strict. Not, ‘you have to breastfeed, because it’s really important,’ but maybe a bit more information, so people don’t feel guilty if they do not breastfeed.”
And so the conversations sparked by studies like Raissian and Su’s could impact the socio-emotional well-being of expectant mothers who can’t breastfeed, or who choose not to. They could also go a long way towards tackling the economic and racial dynamics of the breastfeeding debate, whereby poor women, or women of color, are shamed for not doing what wealthier, white women do for their babies. As Courtney Jung writes in The New York Times, “The effect of the moral fervor surrounding breastfeeding goes beyond mere shaming. It also reflects, and reinforces, the divisions of race and class that have long characterized American social life.”
In our conversations, Raissian takes care to clarify that the point of her study is not to discourage women from breastfeeding. Her goal, she explains, is to jumpstart a conversation about which of the benefits attributed to breastfeeding (if any) have been overstated. “We need a place where we can have a much more honest communication in understanding the true, actual benefits of breastfeeding,” she says—because “if we’re overstating it, we’re over-prescribing it.” Researchers’ ultimate focus, she adds, should be on helping society make the right investments in supporting all new mothers, whether they choose to breastfeed or not.
Read more from our series on Rewiring Childhood. This reporting is part of a series supported by a grant from the Bernard van Leer Foundation. The author’s views are not necessarily those of the Bernard van Leer Foundation.
This post has been updated with clarification about Belarus drinking water.