Breastmilk, the microbiome and misogyny


Imagine a study that compared the mouth microbiome of men and women and found that up to 1/3 of the common mouth bacteria differed between men and women. Plausible, right, since male and female hormones are different and might exert some impact on the bacteria that live inside mouths?

Now imagine that the researchers concluded that women are inferior to men because their mouth microbiome differs.

This is not science; it’s misogyny. And that sucks!

Outrageous, right?

How do the researchers know which microbiome is better? Why do they take men as a the standard? Why do they assume that any differences between women and men are “proof” that women are inferior? Where is the evidence that the mouth microbiome has anything to tell us about the relative value of men and women?

It’s outrageous because the researchers obviously started with the assumption that men are the standard against which women ought to be judged. Furthermore, they engaged in motivated reasoning; to the extent that any differences between men and women exist, they are assumed (without any scientific evidence) to support the pre-determined conclusion that men are “better” than women.

If you understand why the above imaginary study would be junk science, you now understand why the new study, Composition and Variation of the Human Milk Microbiota Are Influenced by Maternal and Early-Life Factors, comparing breastmilk to pumped breastmilk is also junk science.

It starts with the unwarranted assumption that breastmilk taken from women who breastfeed exclusively is some sort of standard. It assumes that any differences detected in the microbiome of women who pumped are “proof” that pumped breastmilk is inferior. And it demonstrates that the conclusion to which the authors were committed a priori was that fresh breastmilk is better than pumped.

According to the MedicalXpress:

A large-scale analysis in humans reveals that indirect breastfeeding using pumped milk is associated with the depletion of oral bacteria and a higher abundance of potential pathogens compared with direct breastfeeding at the breast…

Specifically, indirect breastfeeding (defined as at least one serving of pumped milk in the preceding two weeks) was associated with a higher abundance of potential opportunistic pathogens, such as Stenotrophomonas and Pseudomonadaceae.”Increased exposure to potential pathogens in breastmilk could pose a risk of respiratory infection in the infant, potentially explaining why infants fed pumped milk are at increased risk for pediatric asthma compared to those fed exclusively at the breast,” says first author Shirin Moossavi of the University of Manitoba. To determine if this is the case, there will need to be additional research into how changes in the milk microbiota affect colonization of the infant gut microbiome, which influences health.

There is so much wrong with these claims that it’s difficult to know where to start:

Potential pathogens?

Could pose a risk?

“Indirect” breastfeeding, implying it’s not real breastfeeding?

Let’s look at the actual scientific findings.

What did the authors set out to explore?

Two main pathways have been proposed to explain the origin of milk microbiota: entero-mammary translocation of the maternal gut microbiota and retrograde inoculation by the infant’s oral microbiota. The fact that colostrum collected even before the first infant feeding already contains a microbial community supports the entero-mammary pathway, while the similarity of infant oral microbiota to breastmilk microbiota supports the retrograde pathway. It is plausible that both pathways are contributing to the bacterial content of human milk.

What did they find?

To identify potential inherent patterns in the milk microbiota, hierarchical clustering was performed on the core microbiota…

Milk Microbiota Diversity Is Associated with Mode of Breastfeeding, Method of Milk Expression, and Other Maternal Factors

Overall, milk samples had a mean ± SD richness (observed ASVs) of 147 ± 44 and diversity (inverse Simpson index) of 15.8 ± 8.7. We explored the association of α diversity with multiple maternal and infant factors. In multivariable linear regression adjusting for infant sex, mode of feeding, mode of delivery, and parity (number of older siblings), indirect breastfeeding was independently associated with lower milk bacterial richness (adjusted Math Eq–18.9, 95% CI, −27.9, −9.9, p < 0.001) and diversity (adjusted Math Eq−2.08, 95% CI, −3.91, −0.25, p < 0.05; Table 1). In a subset with data on milk expression method, pump versus manual expression was associated with significantly lower bacterial richness (Math Eq = −39.6, 95% CI, −60.5, −18.7, p < 0.001)…

In addition:

Milk Microbiota Composition Is Associated with Breastfeeding Practices, Multiparity, and Other Maternal Factors in a Sex-Specific Manner


We observed intriguing differences in milk microbiota α diversity and overall community structure according to infant sex. There was a trend in direct association of infant sex with milk microbiota in our SEM analysis, and we also found sex-dependent associations of breastfeeding mode, exclusivity, and lactation stage (higher R2 in males), as well as parity, mode of delivery, and maternal BMI (significant in females only) with milk microbiota in RDA…


Notably, however, the many factors we evaluated collectively explained less than a third of the total variation observed in milk microbiota composition, indicating that other unmeasured factors are contributing to the large inter-individual variation in milk microbiota profiles.

The study has some serious limitations:

The main limitation of our study is that milk samples were pooled from multiple feeds and were not collected aseptically.

In summary:

1. The authors found wide variation in the microbiome of pooled samples of of breastmilk that were not collected aseptically.
2. More than two thirds of the variation was unexplained.
3. Less than one third of the variation could be attributed either to the gender of the infant or a history of pumping breastmilk.
4. There is no evidence that any of these differences are clinically relevant.

So why did the authors blare their “conclusion” that pumping changes the milk microbiome? It certainly wasn’t because the findings have ANY implication for mothers or babies.

This paper, like nearly every aspect of professional lactivism from the inception of La Leche League in the 1950’s, isn’t about what is good for babies. It is about controlling women’s behavior through guilt. That’s why I have a handy rule of thumb for breastfeeding “science”:

Any paper that concludes or implies that giving up career and livelihood to stay home and breastfeed is “better” is deserving of special scrutiny.

More often than not, such papers are not science, but exercises in motivated reasoning.

Most people are unaware that La Leche League, the first and still the most influential lactivist organization was founded specifically to convince women to stay home with young children.

In the book La Leche League: At the Crossroads of Medicine, Feminism, and Religion, Jule DeJager Ward explains:

[A] central characteristic of La Leche League’s ideology is that it was born of Catholic moral discourse on family life … The League has very strong convictions about the needs of families. These convictions are the normative heart of its narrative… The League’s presentations and literature carry a strong suggestion that breast feeding is obligatory. Their message is simple: Nature intended mothers to nurse their babies; therefore, mothers ought to nurse…

In 2019 you can no longer tell women that it is “better” for babies to have a stay at home mother. It’s easy to recognize the misogyny in that claim. Therefore, lactivist tactics have switched to convincing women that breastfeeding is better for babies. They’ve spent the past 30+ years promoting an expanding list of health benefits … that have nearly all been debunked.

So the search is on for ever more arcane “benefits” and investigations of the breastmilk microbiome are part of that effort. Our understanding of microbiomes is primitive. We have literally no idea what a “normal” microbiome looks like. We have literally no idea whether variations among individuals are clinically relevant or merely incidental. That hasn’t stopped breastfeeding researchers from drawing junk science “conclusions.” And those conclusions always rest on the assumption that a mother breastfeeding directly and exclusively is the standard against which any variations should be assumed inferior.

That reflects the patriarchal belief a woman’s place is in the home and nowhere else. It also reflects the patriarchal belief that women should be judged by the function of their reproductive organs, not their intellect, talents or character. That’s not science; it’s misogyny. And that sucks!

12 Responses to “Breastmilk, the microbiome and misogyny”

  1. sheistolerable
    February 25, 2019 at 9:15 am #

    Dr Amy, first, let me say what a tremendous gift your work has been to me as a stressed out new mom. I chose to combo feed my son so I could get enough sleep to function and my husband could share the feeding relationship. I am certain I was headed for PPD otherwise and your clear communication of research was so important to my decision. Thank you.

    Obviously this study is specious for the reasons you explain, but women deserve good research that helps them choose among breastfeeding, formula feeding and bottle feeding breastmilk. Women are heavily pressured to pump for any supplemental feeds and to provide all the baby’s food when we return to work. Public policy in the US promotes pumping more so than breastfeeding-that would require long term paid leave for every mom, not crappy little pumping rooms at large firms only. From what I can tell, this is based on the assumption that any benefits of breastmilk are in the milk, not the mechanical act of feeding, and thatthey are retained despite any freezing, etc. Women go nuts pumping based on this assumption which I have never seen supported by the science. With the benefits of breastmilk over formula as trivial as you frequently point out, women absolutely deserve to know if pumping makes a difference.

  2. Bobismeisbob
    February 18, 2019 at 12:09 pm #

    One small nitpick the ““Indirect” breastfeeding, implying it’s not real breastfeeding?” seems to be a stretch to me.

    It’s a perfectly functional term for feeding using pumped breast milk and doesn’t seem to imply it’s not “real” just that it’s different from direct breastfeeding (which again they use a special term for and so you could also claim they are “implying it’s not real breastfeeding”.

  3. rational thinker
    February 15, 2019 at 5:23 pm #

    “Oh no if mom can pump adequate amounts of breastmilk then there is nothing stopping her from (gasp!) getting a job. Quick lets make a new mom feel guilty for pumping by doing a bullshit study and conclude that pumping is dangerous for babies.”

  4. Mimc
    February 15, 2019 at 5:23 pm #

    Ever since I saw this story I’ve been expecting your response. It seems the study was just as bogus as I thought it was. Thank you.

    • Sarah
      February 16, 2019 at 2:36 am #

      Microbiome has been the trendy stick to beat with over mode of birth and feeding for a few years now. They take a truth, that microbiome differs depending on mode of feeding and of birth, and then make the assumption with no evidence that some microbiomes are better than others. This is no exception.

      • rational thinker
        February 16, 2019 at 7:55 am #

        Sanctimommies have really latched onto “microbiome” cause it sounds scientific so they will appear to other laypeople that maybe the know what they are talking about even though they don’t.

        • Anna
          February 16, 2019 at 7:08 pm #


        • mabelcruet
          February 17, 2019 at 6:02 am #

          It’s being used in the same way as ‘epigenetics’. It sounds sciency and clever and to a lay person it gives the user a spurious authority.

          Hannah Dahlen has been touting her epigenetics theories for years (she calls it the EPIIC hypothesis) and produces reams of highly speculatively ‘research’ papers on it. The papers are without exception just reviews with a lot of speculation, no actual primary research or genuine results, just a load of possibles and maybes-they are actually great examples of how to spin a candy floss thin evidence base into something that sounds substantial and meaningful. But they are being used as evidence and are impacting on mothers-the practice of smearing baby’s face with maternal vaginal secretions if the baby was born by section, to mimic the exposure the baby should have got, originated with her work.

  5. fiftyfifty1
    February 15, 2019 at 4:35 pm #

    talk about P-hacking

  6. David Whitlock
    February 15, 2019 at 1:36 pm #

    I question how relevant this study actually is. It is well known that antibiotics have strong affects on the human microbiome, yet they see no effects of antibiotics at all. Maybe their study didn’t have enough power to detect those differences. In that case, what they are ‘seeing’, could well be ‘noise’ and not systematic variations that are relevant for health.

    From table S1,

    Maternal intrapartum antibiotics 139 (35.8%)

    Maternal postpartum antibiotics before 3-4 months 41 (10.6%)

    Child antibiotics before 3-4 months 11 (2.8%)

    That is a whole lot of antibiotic use, however, they also say:

    “No associations were observed between individual taxa and other factors examined, including maternal secretor status, ethnicity, atopy, parity, mode of delivery, or intrapartum antibiotics (data not shown).”

    I work in the Human Microbiome, at AOBiome. We are focused on ammonia oxidizing bacteria (which we think are ‘keystone species’, which this study did not find).

    Another thing that stood out to me, was the difference between diversity in ’emergency C-section’ and ‘non-emergency C-section” in table 1. The values all overlapped, but there was a difference between emergency and non-emergency; implying that some of the difference in microbiome may have related to the maternal physiology that caused the need for an emergency C-section (something I think is very likely, but which has not been studied).

  7. space_upstairs
    February 15, 2019 at 1:06 pm #

    I read this as I heard non-medically-indicated C-section scare-mongering on the radio. So…I had a technically elective C-section and have chosen to pump as my primary (near-exclusive, even) means of feeding my baby. The only known problem so far, and it may not even really be much of one, is that my baby eats less and gains weight slower so far than the doctor says she should, despite good height growth, normal bowel movements, the quicker food delivery of bottles, and still-chubby cheeks and legs. So I’m supposed to think there are other, untold, barely understood problems and freak out about them, because the doctor didn’t try everything to force her through my hips and I didn’t want to primarily breastfeed on tap after painful cracked nipples from doing just that for 3 days? Meh. I’d rather raise my daughter feminist, telling her when she’s old enough to understand that you have choices not just in theory, but in practice too if you’re bold enough and have a bit of luck on your side (financial security).

    • Sarah
      February 16, 2019 at 2:35 am #

      That would appear to be exactly what we’re supposed to do, yes. I’ll be sitting over here not giving a fuck, laughing at the junk science.

Leave a Reply

You must be logged in to post a comment.