A Lack of Evidence for Baby-Friendly Hospitals

Seven or eight years ago, Beth Israel Deaconess Medical Center in Boston faced a choice about whether to pursue an innocent sounding, but controversial, designation: “Baby-Friendly.”

While a warm attitude toward infants might sound like a given, Baby-Friendly is an official accreditation that hospitals can earn by following 10 steps to promote breastfeeding. These steps include avoiding formula and bottles, encouraging skin-to-skin contact between moms and babies, and having infants “room in” full time with their mothers rather than visit nurseries. The World Health Organization and Unicef outlined these steps, and detailed guidelines for achieving them, in 1991. That same year, the two organizations launched their Baby-Friendly Hospital Initiative (BFHI) to implement those practices. They hoped to increase breastfeeding rates, and infant health, around the world.

In the United States, the nonprofit Baby-Friendly USA grants hospitals the Baby-Friendly designation. And by the organization’s own account, these efforts have been wildly successful. “Few social change initiatives have matched the scale, complexity, and success of the Baby-Friendly Hospital Initiative,” reads a page on its website. From the 1990s to 2015, the percentage of U.S. babies “ever” breastfed increased roughly 30 percentage points, for which Baby-Friendly USA claims some credit. “After decades of commercial interests pulling us in the opposite direction,” the website states, “most expectant mothers now have many options to ensure their infant’s birth occurs in an evidence-based environment.”

That evidence base, though, is wobbly. Globally, research shows that following the 10 steps can increase breastfeeding. But the picture is murkier when researchers zoom in on wealthy countries. “I think using global data and applying it to specific high-income contexts is troublesome,” says Vicky Fallon, a perinatal psychologist at the University of Liverpool in the United Kingdom.

From the 1990s to 2015, the percentage of U.S. babies “ever” breastfed increased roughly 30 percentage points.

Further, Fallon and some other researchers contend that a strict, one-size-fits-all set of inflexible steps is inevitably unfriendly to mothers, and even dangerous to babies. The WHO and Unicef addressed some of these concerns in 2018 with updated guidelines. Baby-Friendly USA, though, didn’t immediately adopt the new international guidelines and says that it is still working on its own updates. (The executive director of Baby-Friendly USA, Trish McEnroe, did not respond to multiple interview requests.)

Ultimately, Beth Israel Deaconess Medical Center decided to forgo accreditation. As part of the decision-making process, the hospital had formed focus groups. In their conversations, patients and nurses expressed “a real discomfort” with the Baby-Friendly program’s limitations on the ability of mothers to choose having their newborns placed in the hospital’s nursery if they desired, says acting chief of the department of obstetrics/gynecology Toni Golen. Some mothers, after all — particularly those recovering from caesarean sections or long, difficult labors —  might desire a brief respite to recover. But in some cases under a Baby-Friendly designation, that’s not an option — and that concern, Golen said, played a key role in the hospital’s decision.

More than 600 facilities in the United States, and more than 20,000 worldwide, have made the opposite choice. As U.S. hospitals continue to march toward designation, and thousands of American women give birth each year in Baby-Friendly facilities, researchers are asking whether that’s a good idea.

In the mid-20th century, formula boomed while breastfeeding plummeted. Another pendulum swing in the 1970s brought breastfeeding back into fashion in the United States, but the percentage of mothers nursing their babies stayed pretty flat from the early 1980s to mid-1990s.

Worldwide, breastfeeding rates in the 1990s were also not as high as the WHO and others judged optimal for infant health, which led to the launch of the Baby-Friendly Hospital Initiative. Shortly after, the most well-known Baby-Friendly study took place in Belarus. This trial included more than 16,000 mothers — all of whom planned to breastfeed — and 31 facilities, randomly assigned to follow the Baby-Friendly guidelines or not. It was a success: Infants in the Baby-Friendly group were more likely to be breastfeeding three, six, and even 12 months later.

But the standard of health care in that time and place bore little resemblance to the United States today, says Carrie Patnode, an investigator at Kaiser Permanente Center for Health Research in Portland, Oregon. Belarusian hospitals offered no support for breastfeeding. The study authors chose the location, in fact, so there would be a stark contrast between Baby-Friendly care and business as usual. Under these circumstances, Patnode says, it’s not surprising that any initiative would boost breastfeeding.

In 2016, Patnode and coauthors reviewed data on BFHI and other breastfeeding initiatives for the U.S. Preventive Services Task Force. They looked at 52 high-quality, controlled studies in developed countries. But since they specifically wanted to know what might work in the United States, they left out the Belarus study. “We really felt like it wasn’t applicable,” Patnode says. Based on more relevant studies, the authors concluded that Baby-Friendly designation had no proven effect on breastfeeding.

“I think people were surprised,” Patnode says. “There’s been a long-held belief that the Baby-Friendly Hospital Initiative and other policies have evidence to support them.”

Research coming out of other highly developed countries paints a similar picture. A 2013 study of women who gave birth in Queensland, Australia, where practices such as rooming-in and limiting formula are common, found that while some of these practices boosted breastfeeding, the Baby-Friendly designation in and of itself did not. In fact, women who delivered at Baby-Friendly hospitals had significantly lower odds of breastfeeding one month later.

In 2019, Fallon and coauthors published a review looking at the United Kingdom. They found some evidence that Baby-Friendly designation boosts breastfeeding in a baby’s first week of life, but no longer.

Meanwhile, other studies have concluded that the BFHI in the United States is ineffective or lacks good evidence, and that trials in other countries aren’t very relevant. One study found that Baby-Friendly designation increased breastfeeding at four weeks by 4.5 percentage points among mothers with lower education — an improvement the authors called “meaningful” — but not among the population overall. Valerie Flaherman, a pediatrician at the University of California, San Francisco, School of Medicine, says that without more evidence, caution makes sense.

“Looking at best practices in the United States alone,” Flaherman and her coauthor wrote in a 2017 editorial response in the Journal of the American Medical Association, “the BFHI does not appear to be beneficial and may cause harm.”

In the three decades since the BFHI launched, society has learned more about feeding babies and keeping them safe. The Baby-Friendly guidelines haven’t always kept up.

For example, the original 10 steps say breastfeeding infants should have no pacifiers at all. Conventional wisdom holds that pacifiers interfere with breastfeeding and can cause “nipple confusion.” But research has been mixed. A 2016 Cochrane review concluded that among moms who planned to breastfeed, introducing a pacifier at birth or two weeks later had no effect on breastfeeding. Additionally, studies have shown that pacifiers lower the risk of sudden infant death syndrome.

The 10 steps also say not to give formula “unless medically indicated” — and how this guidance is delivered can create fear, says a nurse at Boston’s Massachusetts General Hospital (MGH) who asked not to be identified for risk of losing her job. MGH has been Baby-Friendly since 2015. The nurse says new mothers leave the hospital “terrified to supplement.”

Studies have shown that when babies need extra food in their first days, carefully supplementing with formula doesn’t harm later breastfeeding. But a fear of formula can increase the likelihood of readmission, Flaherman and a coauthor wrote in 2016. Many mothers experience a delay before their bodies make enough milk. If they don’t supplement in the meantime, their babies are more likely to end up dehydrated or back in the hospital.

New mothers leave the hospital “terrified to supplement.”

The BFHI recommendations on skin-to-skin contact and rooming-in have also stirred controversy. Evidence shows that skin-to-skin contact between moms and babies right after birth does help with breastfeeding, among other benefits. Rooming-in has benefits too, such as helping parents learn their baby’s hunger signs, says Enrique Gomez-Pomar, a neonatologist at St. Bernard Regional Health Center in Jonesboro, Arkansas, who reviewed evidence for each of the 10 steps in 2018. But, he says, there isn’t strong evidence that rooming-in helps breastfeeding.

Emphasizing rooming-in and round-the-clock cuddling with babies might lead to dangerous situations, though. A 2016 American Academy of Pediatrics report warned that unobserved skin-to-skin contact and rooming-in carry potential dangers including suffocation, falls, and sudden collapse — a very rare event in which an infant’s heart or breathing stops in the first days of life. A study of 26 sudden collapse cases in Sweden found that 15 occurred during skin-to-skin contact. A 2019 paper reported a cluster of three newborn falls at one hospital after it started the Baby-Friendly designation process. All three falls happened when mothers fell asleep holding their babies; one infant had a skull fracture that caused later seizures.

Joel Bass, chair of pediatrics at Newton-Wellesley Hospital in Massachusetts, led a 2018 study finding that while overall infant death has dropped in recent years with safer sleep practices, early newborn deaths have stayed flat. He and his coauthors suggested that skin-to-skin care might be a factor. But Melissa Bartick, an assistant professor of medicine at Harvard Medical School and founder of the Massachusetts Baby-Friendly Collaborative, strongly disagrees that Baby-Friendly practices are dangerous. “That was a very loose and unproven association,” she says. “I think that we debunked that pretty firmly with our latest study” — a 2019 publication that found deaths of U.S. infants in their first week of life had actually decreased in recent years. However, Bass and coauthors raised several methodological concerns about that paper, and Bass calls some of the data Bartick used “preposterous.” (He and Bartick have sparred frequently in the literature.)

Based on the evidence, Gomez-Pomar thinks skin-to-skin care is only safe when done cautiously, with a nurse checking on the infant every 10 to 15 minutes for the first two hours after birth. “Yes, you can do it, but you need somebody to be there. You need somebody to look at the baby,” he says.

In light of the current Covid-19 pandemic, the U.S. Centers for Disease Control and Prevention has developed guidance recommending that newborns be separated from mothers with confirmed or suspected infections. For its part, Baby-Friendly USA has issued statements referring facilities to this guidance while continuing to promote breastfeeding. “We understand that deviations in practice may be required to respond to Covid-19 concerns,” the organization wrote in a March 17 statement, “and we ask simply that you clearly document the circumstances in these unique situations.”

Some hospital administrators do point to specific benefits of the BFHI accreditation process. Lauren Hanley, an OB/GYN and co-chair of BFHI for Massachusetts General Hospital, says that the process “elevated our quality of care even further.” For example, hospital staff received specific training in how to help mothers and babies learn to feed. “Previously, when the nurse was not confident in their skills, they would have to call a lactation consultant to help,” Hanley says, but now the nurses can support mothers themselves.

But not all women have felt supported at MGH and other Baby-Friendly hospitals. The nonprofit Fed Is Best has collected anecdotes from families who say Baby-Friendly policies led to tragic outcomes for their babies, including near-starvation and even death. (Some of these stories were described in a recent New York Times article.) 

The MGH nurse also gave birth at the hospital. She says that once mothers are alone with their babies, trying to room-in and exclusively breastfeed, there isn’t enough support to make these demands feasible. If she hadn’t been able to rely on her own knowledge and experience, she says, “I can’t imagine how I would have survived in that room with as little help as I had.” 

Another Boston-area mother who delivered at MGH in 2016, and who asked not to be identified because of her employer’s ties to the hospital, went into labor confused about whether a hospital nursery existed at all. When she asked her doctor about it beforehand, “She was honestly very cagey about the whole thing,” the mother says.

After a long delivery that ended in a C-section, the mother was exhausted and her baby wouldn’t stop crying. “I just kept ringing the nurses and having them come in, and I’m like, is there somewhere that my baby can go?” she says. No one offered her nursery care. (When asked about this specific incident and the hospital’s nursery policy, Hanley stressed that MGH still has a nursery, as state law requires. But the hospital’s “overarching goal is not to separate mothers and babies unless” there’s a medical reason for doing so.)

In the early-morning hours of the third day, the mother estimates, a nurse weighed the baby and seemed concerned. “[She] actually said to me, ‘I think what’s happening is that your baby is starving.’ And she said, ‘I’m not gonna put this in your record, and don’t tell anybody that I’m the nurse who did this, but we need to give this baby formula.’” The mother was shocked by how nervous the nurse seemed to be.

“The health care professional is often under the gun,” says Bass. Baby-Friendly USA guidelines require most criteria to be met by 80 percent of mothers, he explains. Nurses may feel pressure from their supervisors to maintain those percentages. Bass calls the guidelines “extremely rigid and extremely intrusive,” and says they can prevent both parents and health care providers from doing what they think is best for babies.

But Bartick says the negative Baby-Friendly experiences are outliers. “Most moms really appreciate the experience,” she says. “There’s this general perception going on that moms hate it! And that’s really not true.”

Currently, there is a lack of data on mothers’ satisfaction with Baby-Friendly hospitals. There is some qualitative research looking at mothers’ experiences more broadly, though, and Fallon reviewed five such papers in her U.K. study. She and her coauthors concluded that Baby-Friendly settings “may promote unrealistic expectations of breastfeeding, not meet women’s individual needs, and foster negative emotional experiences.”

Some women had positive experiences at Baby-Friendly hospitals, Fallon says. But these were mostly mothers who met their breastfeeding goals. When mothers had trouble breastfeeding, Baby-Friendly messaging made them feel guilty.

“We need to think about maternal wellbeing on an equal footing to infant health,” Fallon says. Studies haven’t shown that going Baby-Friendly increases breastfeeding in the U.S. or U.K. But even if did, she says, “if in turn it has a negative effect on emotional wellbeing in mothers, is that a trade-off that we want to engage with?”

The WHO addressed many of these concerns in April 2018 with updates to the BFHI guidelines. Instead of forbidding pacifiers, the 10 steps now say hospital staff should “counsel mothers” on their use. Instead of recommending skin-to-skin contact throughout the hospital stay, the new guidelines stress skin-to-skin in the first hours of life — when research shows it’s helpful — and only with “sensible vigilance and safety precautions.” The new guidelines acknowledge that some breastfed babies will need supplemental feeding.

Although the guidelines still recommend rooming-in around the clock, “they’re not as crazy about it,” Gomez-Pomar says. He says the 2018 guidelines are a change for the best. 

Bass agrees: “They really did a great job on their revision.” 

Baby-Friendly USA, though, didn’t immediately adopt the new international guidelines. In December 2019, the group published a set of “interim” guidelines that allow more flexibility in rooming-in, among other tweaks. The group “recognized that some of WHO’s new BFHI requirements had more flexibility than the ones that are currently in use,” says a statement on its website. “We anticipate releasing the completely revised [guidelines]… sometime in the second quarter of 2020.”

“It’s very difficult to admit that you did something wrong, or that your recommendations were wrong,” Gomez-Pomar says. “I think that’s the reason why they took so long to actually accept the new guidelines.”

“It’s very difficult to admit that you did something wrong, or that your recommendations were wrong.”

Another reason may be that the new international guidelines seem to undermine the whole reason for Baby-Friendly USA’s existence. “The traditional Baby-Friendly model was largely organized around the naming of Baby-Friendly facilities,” the WHO and Unicef write. However, designation is “only one of a number of useful options.” Individual hospitals may find more effective ways to best care for patients and support breastfeeding.

At Beth Israel Deaconess Medical Center, Golen stresses that the decision to forgo a Baby-Friendly designation wasn’t a dismissal of breastfeeding’s value. Just the opposite, she says: the decision “was a redoubling of our commitment to the importance of breastfeeding.” The researchers who question the evidence for the 10 steps share the same goal as Baby-Friendly USA, which is to encourage breastfeeding and keep mothers and babies healthy. They just disagree about the best way.

Bass says hospitals should still get designated if they think it will help them serve their particular population. But blanket recommendations that hospitals go Baby-Friendly, such as from the U.S. surgeon general or the United Kingdom’s NHS, don’t make sense, argues Bass.

“I don’t think that having a plaque on the wall works,” Gomez-Pomar agrees. He says evidence has already shown the two best ways to increase breastfeeding: prenatal education and postnatal support. In other words, preparing mothers and families for the hard work of feeding babies, and helping them do it after they leave the hospital. 

“Breastfeeding makes sense,” he says. “But Baby-Friendly is not the answer.”

Elizabeth Preston is a freelance writer whose work can be found in New Scientist, Discover, Quanta, The Atlantic, and STAT News, among other publications.