A good friend of mine living in Scotland, who had a baby last year, mentioned to me how disappointed he has been with the U.K. National Health Service’s promotion of breastfeeding. Calling the alleged benefits overstated, he said, is itself an understatement.
I nodded in general agreement, acknowledging that many of the alleged benefits of breastfeeding have only been found in observational studies.
Observational studies on breastfeeding merit skepticism, because they all suffer from the same major problem: breastfed infants on average differ from formula-fed infants not just in how they are fed in infancy, but in practically every other possible way–maternal education, maternal IQ, poverty, neighborhood safety, exposure to environmental toxins, race, and type and quality of childcare. In scientific terms, breastfeeding is confounded, out the wazoo.
We cannot tell which benefits found in an observational study derive from breastfeeding rather than from the myriad other advantages linked with breastfeeding.
(The “good” observational studies attempt to control statistically for the other relative advantages of breastfed infants. Unfortunately, controlling for confounds only works well when (1) all the important potential confounds are known, and (2) when there is a fair amount of overlap between the groups being compared. Neither of which is true when it comes to breastfeeding.)
In an ideal world, we would settle this question by conducting several large randomized controlled trials (RCTs), in which new mothers would be randomly assigned to breastfeed. RCTs are the gold standard in medicine for determining whether a true cause and effect relationship exists. In practice, though, such trials are neither feasible nor ethical.
Fortunately, we have the next best thing: a handful of studies that have cleverly circumvented the problem of confounding. These fall into two categories:
- sibling studies, which compare siblings from the same families who were breastfed for different lengths of time, or who were not both breastfed.
- a large RCT of a highly successful breastfeeding intervention (PROBIT Trial).
(Is the PROBIT Trial an exception to the no-RCT rule? No. Women in the PROBIT trial were not randomly assigned to breastfed or not; they were randomly assigned to receive a breastfeeding intervention or not.)
After my friend and I spoke about his irritation with the medical organizations like the American Academy of Pediatrics (AAP) and NHS overstating the benefits of breastfeeding, I was dissatisfied with my vague sense that he was right. I wanted to know exactly which benefits had been oversold and exactly which were supported not just by observational studies but by better-designed studies.
The short answer: Nearly all the alleged long-term benefits are likely the result of confounding, not breastfeeding. Better-designed studies find only a handful of real benefits: a reduced chance of severe gastrointestinal infections and a lower risk of eczema during infancy, and perhaps a small boost in childhood IQ.
Alleged Breastfeeding Benefits According to the NHS
According to the NHS, breastfed infants are…
- less likely to suffer from vomiting or diarrhea and therefore less likely to go to hospital
- less likely to develop type 2 diabetes in later life
- less likely to become obese in later life
- less likely to suffer from heart disease in later life
- less likely to suffer from constipation
- less likely to get a chest or ear infection and therefore less likely to go to hospital
- less likely to suffer from tooth development problems
- more likely to have good communication and speech skills
- more likely to have good circulation
- less likely to suffer from wind, colic and constipation
- less likely to develop eczema or asthma
Sounds pretty impressive, right? Until you set aside the evidence from observational studies…
Breastfeeding Benefits Supported By Sibling Studies and A Randomized Controlled Trial
So, let’s start with the good news:
Breastfeeding clearly reduces the risk of severe gastrointestinal infections during an infant’s first year of life. This is true in developed nations, where there is widespread access to clean water, health services, and nutritious food, as well as in developing ones, where the reduction in illness is larger and often life-saving.
The lower risk of infections is also especially important for preterm babies, for whom breast milk substantially lowers the chances of necrotizing enterocolitis, a serious and life-threatening infection.
As for a lower risk of constipation, even though no published research I can find has directly tested this effect, I am inclined to give this one to the NHS. As a formerly breastfeeding mom, I know just how–dangerously and sometimes explosively–runny breastfeeding poops can be.
In terms of benefits that persist into childhood, breastfeeding may also provide a small boost to IQ. The evidence here is not as solid as that for infections, but more compelling than I’d expected. (I will dig further into this complex topic in a moment.)
And now the bad news:
Breastfeeding does not reduce the risk of colic. In fact, I have no idea how this alleged benefit made it onto the NHS’s list in the first place. Most of their alleged benefits are found in observational studies. But not colic. Even most observational studies do not find that breastfeeding lowers the chances of colic.
As for the remaining claims–reduced BMI, lower risk of Type 2 Diabetes, improved circulation, and lower risk of heart disease–there is just no solid evidence to support them. This doesn’t mean that breastfeeding has no effect on these outcomes, but it does mean that we have no reason to believe that it does. And, in my opinion, it does mean major medical organizations should not be stating these benefits as facts.
* * *
The second half of this post is for everyone who wants the nitty-gritty details of the PROBIT trial and the sibling studies–what they found, how they were designed, and what, if anything, were their limitations.
(Skip this section.)
The PROBIT Trial
The PROBIT trial, led by Dr. Michael Kramer of McGill University Medical School, recruited women who gave birth at specific hospitals in Belarus between 1996 and 1997, and who intended to breastfeed. Women were randomly assigned to receive an intervention to boost the rate of breastfeeding (cases) or to standard care (controls).
Why Belarus? It seems an odd choice, but Belarus presented two distinct advantages: First, by the mid-1990s, women in Belarus had the same access to clean water, health services, and nutritious food as women living in other developed nations. So the results of Kramer’s study would generalize to a Western context.
Second, hospitals in Belarus did not yet routinely promote breastfeeding after birth. Their after- birth care was similar to that of most developed nations in the 1940s and 1950s, when formula feeding was in vogue . So Kramer’s breastfeeding intervention would likely have a large impact.
By any reasonable standard, Kramer’s choice of location paid off. Breastfeeding rates were higher in the intervention group:
Rates of exclusive breastfeeding were lower than partial breastfeeding across the board, but again, much higher in the intervention group:
So how have the intervention children fared compared to the controls?
During their first 12 months of life, intervention infants had significantly fewer severe gastrointestinal infections and eczema, but not fewer ear or respiratory infections:
At age six and a half, the children from the intervention (largely breastfed) group had…
- No reduction in allergies and asthma. If anything, children who were breastfed showed a nonsignficant trends towards higher rates of allergies and asthma.
- No reduction in BMI or rates of obesity
- No difference in blood pressure
- No reduction in problem behaviors or attention disorders
- Higher IQ: +7.5 (+0.8 to +14.3) for verbal IQ, +2.9 (−3.3 to +9.1) for performance IQ, and +5.9 (−1.0 to +12.8) for full-scale IQ. (But no difference in IQ between infants exclusively breastfed for 3 months versus infants exclusively breastfed for 6 months.)
- Higher teacher ratings for reading and writing.
At age eleven and a half, the adolescents from the intervention group had…
- Lower rates of disordered eating among both girls and boys
- No reduction in BMI or body fat
- No difference in IGF-I (Insulin-Like Growth Factor–an important hormone for regulating growth and body composition.)
(Unfortunately, researchers did not re-assess the outcomes from age 6–academic performance, allergies, or asthma. Instead, researchers focused exclusively on weight and markers of metabolic health. So we cannot say whether the advantage in test scores among the intervention group persisted into adolescence. )
Summing Up the PROBIT Trial Findings To Date
The PROBIT trial suggests that breastfeeding does reduce the risk of severe gastrointestinal (GI) illness and eczema in the first year of life. The reduction in GI infections is supported by mountains of observational evidence, and this makes sound biological sense. Breastmilk contains numerous hormonal, immunological, and antimicrobial factors that aid gut maturation, help seed the growth of beneficial gut bacteria, and protect the gut from invading pathogens. However, the lower risk of infections lasts only until weaning. There is no clear long-term benefit.
What’s far more surprising, to my mind, is the apparent advantage in IQ seen at 6.5 years. The boost for intervention children was apparent in verbal, performance and overall IQ, as well as in teacher ratings of reading and writing.
The IQ and performance findings strike me as fairly compelling for a few reasons. (I can already envision a tidal wave of criticism coming at me for this statement. But please hear me out.)
First, if anything, the PROBIT trial errs on the side of understating the benefits of breastfeeding. This is because it is not randomized trial of breastfeeding, it is a randomized trial of a breastfeeding intervention. Some women in the control group breastfed their babies. Some women in the intervention group did not breastfeed their babies. This waters down the the differences between the control and intervention groups. If anything, the PROBIT trial is biased towards missing small benefits of breastfeeding.
Second, a benefit to brain development is biologically plausible. Breastmilk contains long- chain fatty acids, known to be important for retinal and neural development, as well as lactoferrin, a protein which binds to iron and may help protect the brain from iron-induced oxidative damage. Moreover, consistent with a true biological link, a large observational study demonstrated a dose-response relationship between duration of breastfeeding and childhood IQ at age 3 and age 7. The longer the children were breastfed, the higher their IQ, even controlling for maternal intelligence.
Third, one large sibling study also shows a small benefit in verbal ability among children who were breastfed, a finding which meshes well with those of the PROBIT trial.
Sibling Comparison Studies
In a sibling study, researchers compare the outcomes of siblings who were fed differently in infancy. In some cases, one sibling was breastfed while the other was not; in others, the siblings were breastfed for different lengths of time.
This allows researchers to largely avoid the problem of breastfeeding being confounded by race, socioeconomic status, maternal education, and so on.
(Of course, a family’s socioeconomic status and home environment can change from one birth to the next, but such differences between siblings are expected to be pretty minor, on average, compared to differences between families.)
If nothing else, sibling studies show how profoundly biased the observational studies are when it comes to studying breastfeeding.
The clearest illustration comes from a study of 1,773 siblings aged 4-14, led by Cynthia Colen of Ohio State University. Dr. Colen first analyzed her data as if it were from a typical observational study, comparing siblings across families. She found the usual result: Children who had been breastfed appeared to have significantly lower rates of obesity, asthma, and hyperactivity, and higher parental attachment, behavioral compliance, vocabulary, reading recognition, math ability, intelligence, and scholastic competence.
But then, she reanalyzed the same data, this time comparing siblings within families, thereby eliminating confounding by social and economic factors. And Voilà. She no longer found any apparent benefits of having been breastfed. In fact, the one remaining statistically significant effect was a higher risk of asthma among children who had been breastfed.
Another large sibling study has shown the exact same phenomenon: the apparent benefits of being breastfed vanished when the researchers compared siblings within families.
Other sibling studies find…
- Mixed results for IQ/cognitive ability:
- No difference in achievement test scores among 332 sibling pairs who differed in type of feeding, or among 545 sibling pairs who differed in how long they were breastfed.
- Higher vocabulary test scores among 2,734 adolescents who were breastfed compared to siblings who were not breastfed, but no benefits for GPA, likelihood to be held back a year, or likelihood of attending college.
- And mostly no benefit for BMI:
- Reduced BMI of 0.39 units (a differences of roughly 13-14 lbs. for a 14-year-old of average height) among one study of 118 adolescent sibling pairs.
- But no BMI benefit was seen in three larger studies, one of 267 adolescent sibling pairs, one of 1,773 children between the ages of 4 and 14, and one of 850 adolescent sibling pairs. Most notably, all three studies found a benefit when comparing across families. This benefit was no longer significant when comparing siblings within families.
The Bottom Line
The alleged long-term benefits of breastfeeding derive mainly from low- quality observational evidence. Better designed, less biased studies consistently fail to find these benefits.
Breastfeeding does appear lower the risk of gastrointestinal infections and eczema during infancy. For some–babies living in a developing nation, babies who are born very prematurely–the reduction in infections can be lifesaving.
Some limited evidence also suggests that breastfeeding may boost cognitive ability, especially verbal ability. But this boost, if real, is probably small. It’s way better to have a smart and loving mama than it is to be breastfed.
So why does the medical establishment rely on flimsy evidence to overstate the benefits of breastfeeding? To me, these poorly-supported claims feel downright condescending.
Instead of giving women the real pros and cons of breastfeeding and letting them make an informed decision about how to feed their baby, major medical organizations have made the decision for them, and then feel justified exaggerating and even fabricating the benefits to steer women towards making the “right” choice.
Worse, such claims do little to help our children. If we as a society truly want to raise children’s IQs, improve behavior, and reduce obesity, our efforts would be better spent addressing systemic problems–providing living wages, paid maternity and paternity leave, and affordable, high- quality childcare–than urging all women to breastfeed.
This is why I see the so-called “breastfeeding wars” as nothing more than a media-driven sideshow, an occasionally entertaining distraction from the real underlying problem: a lack of basic support for parents and other caregivers in our society.
Like this post? Check out some related posts on how formula-fed babies tend to sleep longer at night, and why exclusive breastfeeding for 6 months is overrated.
Belfort MB, Rifas-Shiman SL, Kleinman KP, Guthrie LB, Bellinger DC, Taveras EM, Gillman MW, Oken E. Infant feeding and childhood cognition at ages 3 and 7 years: Effects of breastfeeding duration and exclusivity. JAMA Pediatr. 2013 Sep;167(9):836-44.
Colen CG, Ramey DM. Is Breast Truly Best? Estimating the Effect of Breastfeeding on Long-term Child Wellbeing in the United States Using Sibling Comparisons. Social Science & Medicine (1982). 2014;109:55-65. doi:10.1016/j.socscimed.2014.01.027.
Clifford TJ, Campbell M, Speechley KN, Gorodzinsky F. Infant Colic: Empirical Evidence of the Absence of an Association With Source of Early Infant Nutrition. Arch Pediatr Adolesc Med. 2002;156(11):1123-1128. doi:10.1001/archpedi.156.11.1123.
Der Geoff, Batty G David, Deary Ian J. Effect of breast feeding on intelligence in children: prospective study, sibling pairs analysis, and meta-analysis BMJ 2006;333 :945
Der G, Batty G, Deary IJ. Results From the PROBIT Breastfeeding Trial May Have Been Overinterpreted. Arch Gen Psychiatry.2008;65(12):1456-1457. doi:10.1001/archpsyc.65.12.1456-b.
Duijts L, Jaddoe VW, Hofman A, Moll HA. Prolonged and exclusive breastfeeding reduces the risk of infectious diseases in infancy. Pediatrics. 2010 Jul;126(1):e18-25. doi: 10.1542/peds.2008-3256.
Evenhouse, E. and Reilly, S. (2005), Improved Estimates of the Benefits of Breastfeeding Using Sibling Comparisons to Reduce Selection Bias. Health Services Research, 40: 1781–1802. doi: 10.1111/j.1475-6773.2005.00453.
Hamosh, M. Bioactive Factors in Human Milk, Pediatric Clinics of North America, Volume 48, Issue 1, 1 February 2001, Pages 69-86, ISSN 0031-3955, http://dx.doi.org/10.1016/S0031-3955(05)70286-8.
Howie P W, Forsyth J S, Ogston S A, ClarkA, Florey C D. Protective effect of breast feeding against infection. BMJ 1990; 300:11
Kafouri S, Kramer M, Leonard G, Perron M, Pike B, Richer L, Toro R, Veillette S, Pausova Z, Paus T. Breastfeeding and brain structure in adolescence. Int J Epidemiol. 2013 Feb;42(1):150-9. doi: 10.1093/ije/dys172. Epub 2012 Nov 21.
Kramer M, S, Breastfeeding and Allergy: The Evidence. Ann Nutr Metab 2011;59(suppl 1):20-26
Kramer MS, Aboud F, Mironova E, et al. Breastfeeding and Child Cognitive Development: New Evidence From a Large Randomized Trial. Arch Gen Psychiatry.2008;65(5):578-584. doi:10.1001/archpsyc.65.5.578.
Kramer MS, Chalmers B, Hodnett ED, et al. Promotion of Breastfeeding Intervention Trial (PROBIT): A Randomized Trial in the Republic of Belarus. JAMA. 2001;285(4):413-420. doi:10.1001/jama.285.4.413.
Kramer MS, Fombonne E, Matush L, Bogdanovich N, Dahhou M, Platt RW. LONG-TERM BEHAVIORAL CONSEQUENCES OF INFANT FEEDING: The Limits of Observational Studies. Paediatric and perinatal epidemiology. 2011;25(6):500-506. doi:10.1111/j.1365-3016.2011.01211.x.
Kramer MS, Matush L, Bogdanovich N, Aboud F, Mazer B, Fombonne E, Collet JP, Hodnett E, Mironova E, Igumnov S, Chalmers B, Dahhou M, Platt RW. Health and development outcomes in 6.5-y-old children breastfed exclusively for 3 or 6 mo. Am J Clin Nutr. 2009 Oct;90(4):1070-4. doi: 10.3945/ajcn.2009.28021. Epub 2009 Aug 26.
Kramer MS, Matush L, Vanilovich I, et al. Effect of prolonged and exclusive breast feeding on risk of allergy and asthma: cluster randomised trial. BMJ : British Medical Journal. 2007;335(7624):815. doi:10.1136/bmj.39304.464016.AE.
Kramer M, S, Vanilovich I, Matush L, Bogdanovich N, Zhang X, Shishko G, Muller-Bolla M, Platt R, W, The Effect of Prolonged and Exclusive Breast-Feeding on Dental Caries in Early School-Age Children. Caries Res 2007;41:484-488
Kramer MS, Matush L, Vanilovich I, Platt RW, Bogdanovich N, Sevkovskaya Z, Dzikovich I, Shishko G, Collet JP, Martin RM, Smith GD, Gillman MW, Chalmers B, Hodnett E, Shapiro S. A randomized breast-feeding promotion intervention did not reduce child obesity in Belarus. J Nutr. 2009 Feb;139(2):417S-21S. doi: 10.3945/jn.108.097675
Martin RM, Patel R, Kramer MS, et al. Effects of Promoting Longer-term and Exclusive Breastfeeding on Adiposity and Insulin-like Growth Factor-I at Age 11.5 Years: A Randomized Trial. JAMA. 2013;309(10):1005-1013. doi:10.1001/jama.2013.167.
Metzger, M. W. and McDade, T. W. (2010), Breastfeeding as obesity prevention in the United States: A sibling difference model. Am. J. Hum. Biol., 22: 291–296. doi: 10.1002/ajhb.20982
Nelson MC, Gordon-Larsen P, Adair LS. Are adolescents who were breast-fed less likely to be overweight? Analyses of sibling pairs to reduce confounding. Epidemiology. 2005 Mar;16(2):247-53.
Skugarevsky O, Wade KH, Richmond RC, et al. Effects of promoting longer-term and exclusive breastfeeding on childhood eating attitudes: a cluster-randomized trial. International Journal of Epidemiology. 2014;43(4):1263-1271. doi:10.1093/ije/dyu072.
Stanley Ip, Mei Chung, Gowri Raman, Thomas A. Trikalinos, and Joseph Lau. A Summary of the Agency for Healthcare Research and Quality’s Evidence Report on Breastfeeding in Developed Countries. Breastfeeding Medicine. October 2009, 4(s1): S-17-S-30. doi:10.1089/bfm.2009.0050.