But those headline numbers don't tell the whole story. To conduct their research, the authors of the studies, published in the medical journal The Lancet, combed through the massive amount of medical literature that has amassed on the best way to feed infants -- and found it is rife with low-quality studies and potentially biased results. They had to root out many studies in order to get to their estimate of breastfeeding's true benefit.
The popular notion of breast milk as a panacea has stemmed from dozens of studies that have reported benefits ranging from a lower risk of asthma to increased IQ. But flaws in some of the studies tracking long-term health effects raise questions about the magnitude -- and at times the existence -- of some of those advantages. Many fail to properly control for factors such as the mothers' education and wealth. When those effects are accounted for, the benefits almost always shrink.
In other words, some of the advantage that studies ascribe to mothers' milk can be explained by infants' environments and parents, particularly in high income countries. For some researchers, that raises questions about whether efforts aimed at increasing breastfeeding rates in high-income countries will have the expected results.
"Sure, the children might be slightly better off because of being breastfed," said Cynthia Colen, a sociologist at Ohio State University. "But we haven’t changed anything about the environment in which they live: lead in the water, under-performing schools. I don’t think you’re really solving the problem; it's almost passing the buck."
A miracle food
Breast milk for a long time has seemed like a miracle health food, warding off dangerous infections when babies are young -- a crucial protection in countries where clean water and medical care may not be readily available. Other studies have suggested that breastfeeding has a health halo that extends later into life, preventing asthma, leukemia, obesity and diabetes. It increases intelligence and protects mothers against breast cancer.
But in recent years, the pendulum has begun to swing back -- the long-term benefits potentially there, but more modest than it once seemed and hard to tease out.
In 2014 in Social Science and Medicine, Colen of Ohio State University examined 25 years of data tracking children and adolescents in the U.S. to discern the differences between children who were breastfed and those who were not. When she analyzed the data using traditional methods, there was a clear pattern: Children aged 4 to 14 who were breastfed longer did better on 10 of 11 measures, including obesity and tests of cognitive abilities than children who were not. But when Colen more carefully constructed the analysis to compare siblings who were breastfed for different durations or not at all, the differences narrowed and were no longer statistically significant.
In the current study, researchers examined 11 studies that collectively showed a 35 percent reduction in a child's risk of type 2 diabetes. But when they ruled out the eight studies that were not high quality, the reduction in diabetes risk was no longer statistically significant. For obesity, 113 studies showed longer duration of breastfeeding was associated with a 26 percent reduction in the chance of becoming overweight or obese. When they limited the evidence to 23 high-quality studies, they found a reduction of just 13 percent.
The new study found no effect on asthma, allergies, blood pressure or cholesterol. Breastfeeding did protect against infections in infancy and modestly boosted intelligence. And it prevented 20,000 deaths from breast cancer in mothers. But despite dozens of studies, there was not conclusive evidence it protected against obesity or diabetes, although the authors argued the case was "growing."
The scientific breastfeeding literature can sometimes seem like a ping pong match. One study will show a connection; another study will tear it down.
In 2006, a study in the British Medical Journal examined the effect of breastfeeding on cognitive abilities in U.S. youth. At first glance, that study of 5,475 children and more than 3,000 mothers found that breast feeding was associated with a 4-point advantage in IQ over being bottle fed. But when the researchers took into account maternal intelligence, the effect largely evaporated, to a half-point boost.
That conflicts with a 2008 study in the Archives of General Psychiatry that found a greater than 7 point advantage for breastfed children in which women were randomly assigned to an intervention to promote breastfeeding.
Contrast that with a study published last year in PLOS ONE that examined IQ in a population of twins from age 2 to 16 and found only a small effect on IQ at age 2, only on girls.
"It does seem to be an area where the possibility of a benefit is still there, although I would say if there’s something there it’s quite likely to be small," said Geoff Der, a statistician at the University of Glasgow who led the BMJ study. He noted that there's currently a 14-fold variation in the effect size from various studies, which could have a potentially huge impact on any economic estimate of the impact of breastfeeding on productivity.
In order to get closer to the truth, the Lancet reports considered only the highest quality studies that controlled for factors such as maternal intelligence. The researchers found breastfeeding had a positive effect, of 2.6 IQ points. That difference may seem small, but the authors estimated that stepping up current breastfeeding practices to a minimum of six months could account for $302 billion in global productivity lost each year.
That finding isn't uncontroversial.
"This research, this series, I found particularly upsetting, because it is basically the same people saying the same thing, with no new information, and saying it more loudly and in more dire terms," said Joan B. Wolf, an associate professor of women's and gender studies at Texas A&M University. "The information they present offers all of the same problems we’ve been dealing with before."
Nigel Rollins of the World Health Organization, one of the leaders of the Lancet studies, wrote in an e-mail that research suggests interventions early in the lifespan can have the biggest return on investment and to overlook breastfeeding would be "folly."
Another leader, Cesar Victora, an epidemiologist at the Federal University of Pelotas in Brazil, sees breastfeeding singled out and held to a higher level of scrutiny than other research.
"I have worked in over 50 countries, and I have not seen so much opposition to BF [breastfeeding] as in the USA," Victora wrote. "Is it because many women feel wrongly guilty for not breastfeeding in a society that does everything it can to prevent mothers from BF (no paid maternity leave, free formula samples in hospitals, no BF in public, etc)? Are there other explanations?"
"The debate that shouldn't be"
That resistance has something to do with the difficult fact that breastfeeding is also about motherhood.
After the journal Lancet Global Health published the good news that breastfeeding increased intelligence last year, the editor noted a backlash on social media, noticeably from people in high income countries.
"Women should not feel bullied or emotionally blackmailed into breastfeeding by one over-zealous section of society any more than they should be made to feel ashamed for breastfeeding in public by another," Zoe Mullan wrote in an essay called "The debate that shouldn't be." "Breastmilk provided exclusively for at least 6 months is unequivocally the best nutrition a baby can receive; women and their families need respectful advice to make the choice wherever that is possible."
But Wolf points out that women must weigh other risks against breastfeeding. For instance, women may have to reduce their short-term or long-term income by taking time off work in order to breastfeed, which may negatively impact children. She argues that breastfeeding is treated differently than other risks to children's health that are deemed acceptable -- women are not trusted to balance the risks and benefits when it comes to issues of childbearing.
"We continue to conceptualize the mother as a vehicle for the baby, a vector of risk, or a vector of health," Wolf said. "Recommendations that we make and statements about women’s obligations and what is actually healthy for babies, we’d probably never make if they had to do with the father or government keeping the air clean."
The fact that breastfeeding's positive results have been tempered and diminished some as scientific understanding evolved is a good thing. It is a natural part of the way science corrects and refines its findings. But as the beneficial effects become more nuanced, to some it raises questions about where society should be putting its resources -- and how much of a payback should be expected.
"What I always come back to is why do we keep viewing mothers' time and women's time as free," Colen said. "There’s a cost to women’s time, whether they’re working or not; there’s a cost to her time when she’s pregnant and when she’s giving birth and caring for the infant after, and the entire time she’s breastfeeding. We have to recognize those costs in terms of women's time and factor those into economic analyses."