Manufactured Mommy Wars. Le Sigh.
On Thursday en route to Denver to hang out with some super awesome peeps during their Lactation Project Program Retreat, the newest study to be sensationalized in the manufactured, media-fueled mommy wars exploded across my twitter & facebook. In a FB thread, Melanie Martin was making some great points so I invited her to write a guest post for Mammals Suck. Mammals Suck systematically supports moms, advocating for them to have ALL the options, be respected for whichever options fit their lives, and that ALL the options be better; better breast-feeding support, better formula, and the best breast milk science. This post addresses how the recent study does not necessarily deliver the best breast milk science.
(in this case, ewoks might provide added value... for ONCE!)
About Melanie, in her own words: "I’m a biological anthropologist who studies variation in breastfeeding, breast milk, complementary feeding, and infant health and development. I’m also a breastfeeding mom."
About Melanie, in my words: All that stuff she just said PLUS she is brilliant and kickass. She conducts her research among the forager-horticulturalist Tsimane people of Bolivia. Most unfortunately, the Tsimane people were just devastated by horrible flooding. You can help!
Me and Melanie on a boat
coming back from the Channel Islands in 2009.
A recent study found that there was no effect of breastfeeding on several long-term health and behavioral outcomes in U.S. children (Colen and Ramey 2014). Cue the media feeding frenzy and online mommy-war finger wagging.
Outlets like the Daily Mail and Slate were quick to trumpet the study as ‘confirmation’ that the benefits of breastfeeding are overstated. Conspicuously those outlets have supported this claim before through misrepresentations or unsystematic, biased opinion pieces on breastfeeding research. In this case though, the authors of the study make essentially the same argument. Does it hold water?
In brief, authors Colen and Ramey took a huge national survey (N=8,237) of U.S. kids aged 4-14 and tested if having been breastfed during infancy made any difference in their measures on 11 outcomes: BMI, obesity, asthma, hyperactivity, parental attachment, behavioral compliance, and 5 measures of scholastic competence used to gauge intelligence and/or academic achievement. When they analyzed these outcomes across the full set of subjects, they found that kids who had been breastfed fared better in every measure except asthma (for which breastfeeding was unexpectedly associated).
But then they limited their sample to a nifty little subset: sibling pairs in which onesibling had been breastfed but the other had not (N=1,773). Siblings share way more genes and early life experiences than do unrelated subjects, so restricting the sample to only discordant sibs allowed the authors to approach breastfeeding as the main driver of potential differences.
The Kennedy kids in 1928 (JFK presidential library)
When they did that, they found no effect of breastfeeding on any outcome. And from this result, they concluded that the beneficial effect of breastfeeding on these outcomes—in both their initial analysis and all the studies preceding it—was really not an effect of breastfeeding at all, but something else. So, are the ‘benefits of breastfeeding’ really just a statistical construct?
Hardly.
First of all, while the authors’ statistical methods appear to be on the up and up, some of their methodological decisions and interpretations raise eyebrows— for an excellent systematic breakdown see Tracy Cassel’s Evolutionary Parenting post and EA Quinn's Biomarkers & Milk post. Secondly (and echoing a point Tracy & EA make), the Colen and Ramey results speak ONLY to breastfeeding-associated differences in a handful of outcomes in U.S. children—NOT to differences in those outcomes in the rest of the world, and NOT to breastfeeding-associated differences in a whole lot of other outcomes that influence public health recommendations (i.e., infant growth, development, disease and mortality risk, maternal breast and reproductive cancers). So the sweeping statements made by the authors and the media about the “overstated” benefits of breastfeeding are more than a little galling.
Finally though, there is a central conceit of this study and many that precede it that fundamentally limits take-ways— that statistically evaluating the effects of breast- vs. formula feeding on long-term, multifactorial health and behavioral outcomes is the best way to test for differences between breast-feeding and formula-feeding (spoiler alert- it's not, retrospective epidemiological studies are useful but by no means the gold standard).
Imagine this scenario: I find 1000 people who, for at least 2 weeks in 2010, drank organic fresh fruit superfood smoothies everyday for breakfast. Then I find 1000 people who did not. I weigh them today, in 2014, and I find that the two groups differed in obesity rates, but between siblings that did and did not drink smoothies, there was no difference in obesity rates. If I concluded from this result that eating healthy isn't protective against obesity, would I sound like an ass?
Yes I would.
You would probably retort that many other factors—like genetics, exercise, and everything else that the subjects have consumed in this time period—likely had as much if not more influence on their weight in 2014 than did some period of smoothie-drinking in 2010. Even if I controlled for factors like income and education (known to correlate with body mass in the U.S.), you would probably still argue that those controls don’t really capture all the other mitigating factors. And you would be right.
Colen and Ramey effectively showed that the totality of one’s childhood experiences—and not simply whether one was breastfed or not—is what really explains variation in multifactorial health and behavioral outcomes. Good for them. Also, duh.
But this study does not imply that we shouldn’t expect to find differences in breastfed vs. formula-fed infants. Human breast milk evolved over millions of years in tandem with the evolving immunological, nutritional, and cognitive needs of human infants. There are thousands of bioactive components in human milk that influence infant growth, metabolism, immune function, infectious disease risk, and physiological, emotional, and cognitive development in ways we are only beginning to understand (e.g. Ballard and Morrow 2013, Hinde 2013, Martin and Sela 2013). Or that the magnitude of effects may be much greater among vulnerable infants (e.g. in the NICU).
However, to really understand what those differences are, and if they matter, we need to devote more time, money, and energy to researching truly critical questions in lactation and breastfeeding research (Neville et al., 2012), and to studying breastfeeding-associated outcomes that are biologically informed and quantifiable.
We need to more firmly establish how all those myriad human milk components really function in the body—how, for example, specific hormonal, nutritional, and pre-and probiotic factors in milk interact with various hormonal and metabolic systems in infants to regulate growth, energy turnover, and fat storage. Once we establish what those pathways are, then we can determine if and how they might be altered by synthetic substitutes. And then we can study how alterations in those pathways interact with other risk factors to influence more complex outcomes down the road.
Look, Colen and Ramey are absolutely correct that social and economic realities in the U.S. currently make formula feeding a more desirable and/or feasible option for many U.S. mothers and that unilaterally promoting breastfeeding on the basis of its tenuous long-term ‘benefits’ both fails to help mothers who do breastfeed and stigmatizes those who don’t. But studies that persist in looking for differences in inherently muddled outcomes aren’t helping anybody either.
What if we stopped looking for small statistical effects of breastfeeding on IQ, and put more of our money and effort into researching the unique and variable aspects of human milk composition and synthesis? What if we conducted clinical studies to determine if different ratios of breast milk intake, or different durations of breastfeeding, result in observable, biologically meaningful differences in metabolic, immune, and neurological function? Wouldn’t infants benefit from formula that more precisely mimics human milk in structure and function? Wouldn’t caregivers benefit from knowing more precisely what amount of breast milk, and over what duration, gives the most bang for the buck? Couldn’t they then make truly informed and individualized decisions about how to best feed their babies within the context of their lives?
Contributed by Melanie Martin
Department of Anthropology
University of California Santa Barbara
Melanie's website and academic articles here and here!
And now check out bioanthropologist EA Quinn's post with SPECTACULAR tetris joke!
Citations:
Colen CG, Ramey DM. Is breast truly best? Estimating the effects of breastfeeding on long-term child health and wellbeing in the United States using sibling comparisons. Social Science & Medicine. 2014. doi:10.1016/j.socscimed.2014.01.027.
Ballard O, Morrow AL. Human milk composition: nutrients and bioactive factors. Pediatric clinics of North America. 2013;60(1):49–74. doi:10.1016/j.pcl.2012.10.002.
Hinde K. Lactational programming of infant behavioral phenotype. In: Clancy KBH, Hinde K, Rutherford JN, eds. Building babies: primate development in proximate and ultimate perspective. New York, NY: Springer New York; 2013:187–207. doi:10.1007/978-1-4614-4060-4.
Martin MA, Sela DA. Infant gut microbiota: developmental influences and health outcomes. In: Clancy KBH, Hinde K, Rutherford JN, eds. Building babies: primate development in proximate and ultimate perspective. New York, NY: Springer New York; 2012:233–256. doi:10.1007/978-1-
Neville MC, Anderson SM, Mcmanaman JL, et al. Lactation and Neonatal Nutrition : Defining and Refining the Critical Questions. 2012. doi:10.1007/s10911-012-9261-5.
Comments
Post a Comment