Overwhelming or overblown? Making sense of evidence for breastfeeding

Examine, a free weekly newsletter covering science with a sceptical, evidence-based eye, is sent every Tuesday. You’re reading an excerpt – sign up to get the whole newsletter in your inbox.

Back when having a baby still felt like a distant dream for Evelyne McMillan, she told friends there was absolutely no way she was going to breastfeed.

“Why would you want to put your body through all that pain,” she says, “when you can simply feed your baby with formula?”

Evelyne McMillan: “My mind has shifted.”Credit: Justin McManus

She is now 32 weeks pregnant and her prenatal team is encouraging her to reconsider. Like many pregnant women, McMillan finds herself struggling to decide. “My mind has shifted,” she says.

There are many good reasons a mother might choose to breastfeed or use formula or mix the two. But the health advice should be fairly straightforward: exclusive breastfeeding is recommended by the World Health Organisation for the first six months of a child’s life, and breastfeeding should continue at least until they are aged two – a position endorsed by Australia’s top health research organisation and our paediatrics college.

There is a wealth of published evidence showing benefits. Systematic reviews of these published studies show an IQ boost of about 2.6 points, and a reduction in the risk of diabetes by 35 per cent and obesity by 13 per cent. We also see important protections against mortality and infectious disease.

Australian experts agree. Dr Melinda Boss, a human lactation researcher at the University of Western Australia, says there are “irrefutable health outcomes both for mothers and their babies”.

Professor Valerie Verhasselt, head of immunology and breastfeeding at the Telethon Kids Institute, described it as a “mega-plus”.

“If we can breastfeed, we shouldn’t avoid it,” she says.

Case closed?

There are some scientists who remain … sceptical. Among them are author and economist Emily Oster; University of Wollongong epidemiologist Dr Gideon Meyerowitz-Katz, who argued last year that the “benefits of breastfeeding may be oversold”; and child nutrition researcher Professor Lisa Smithers, based at the same institution.

“I was actually quite frustrated with the way breastfeeding was spoken about by very senior breastfeeding advocates: that it’s going to do all these things, but when you look at all the evidence, bring in these different study designs, the evidence is not that strong for all outcomes,” Smithers says.

It turns out the strength of the evidence for breastfeeding depends on how you think about evidence in general.

Let’s start by thinking about why breastmilk might be better than cow’s milk-based formula. Simple: one is designed for babies, the other for bovines.

“Breastmilk is full of a plethora of hormones and bioactive compounds that protect the baby,” says Professor Donna Geddes, director of the University of Western Australia’s Centre for Human Lactation Research.

There is a wealth of published evidence showing benefits from breastfeeding. Credit: Joe Armao

Then there are those systematic reviews. They show “unequivocal differences in babies that are breastfed,” says Melinda Boss.

Here’s where the problems start.

Breastfeeding is what scientists call “strongly socially patterned”. In Australia, older mothers with higher levels of income and education and living in wealthier areas are more likely to breastfeed. Do those factors also affect child development? You betcha.

Plus mothers with higher IQs are more likely to breastfeed – and there’s a correlation between a mother’s IQ and her baby’s IQ, as Oster points out.

Is the breastfeeding making the baby smarter, or are smarter babies more likely to be breastfed? A study that simply compares breastfed and formula-fed kids can’t tell us with certainty.

Different types of evidence for breastfeeding

• Observational studies: compare children who breastfeed with those who do not. Critics say they cannot conclusively show causation.
• Randomised controlled trials: gold-standard scientific evidence.
• Cross-country comparisons: observational studies that compare countries, theoretically weeding out any confounding.
• Sibling studies: studies that compare siblings, who theoretically experience the same environment.

That’s one of the reasons scientists consider observational studies only mid-tier evidence – and why Oster and Meyerowitz-Katz remain sceptical.

What they want to see is a randomised controlled trial, in which you take a group of comparable women – similar IQ or education, for example – and encourage half to breastfeed.

Remarkably, we have only one such trial: PROBIT, a huge study done in 1996 in Belarus.

PROBIT found breastfeeding reduced childhood infections and increased IQ by nearly six points on average. But … the IQ finding was not statistically significant. And when the kids were tested at age 16, any benefit for overall IQ had vanished.

Hmm. But PROBIT’s design meant it measured only the effect of more breastfeeding versus almost none, not the effect of exclusive breastfeeding for six months – so it underestimates the effects, says Boss.

Then there are sibling studies, which test one breastfed child against their non-breastfed sibling. Such studies are usually really good at separating out the effects of something like breastfeeding – as everything else between siblings is equal. Such studies find little difference between breast and formula.

One last type of research to consider is cross-country comparisons. In Brazil, everyone breastfeeds, while in the Philippines wealthy women turn to formula. Pull data from both together and you should theoretically weed out any social patterning effects. Such studies find increases in IQ of between three and six points.

Oster and Meyerowitz-Katz have a simple case: the best evidence comes from PROBIT, and it’s not conclusive.

Sure. But as Meyerowitz-Katz told me, the vast bulk of the evidence points towards breast being best, even if it’s not conclusive. He’s just a quality-over-quantity guy.

Verhasselt and Boss lean on the dose-response effect: as you increase the amount a child is breastfed, their health outcomes improve. Plus there’s the mechanical evidence (babies are not cows), and data from small brain-scanning studies showing increased brain development. Plus it is much, much better for the planet.

Smithers opts for evidence triangulation: there’s lots of observational evidence showing breastfeeding raises IQ, and we have evidence from cross-country comparisons, and the PROBIT data is suggestive.

“I’d always err on the side of caution and say breastfeeding has benefits. You just have to weigh them all up,” she says.

“There are many health benefits, it allows you to bond, and it’s free,” Evelyne McMillan says of breastfeeding. Credit: Justin McManus

Where does this leave prospective parents? I like Evelyne McMillan’s answer.

“There are many health benefits, it allows you to bond, and it’s free,” she says. “I want to give it a go – and you won’t know until you try.”

Enjoyed Examine, our free weekly newsletter covering science with a sceptical, evidence-based eye? Sign up to get the whole newsletter in your inbox.