Breastfeeding and the canonization of false facts

true and false choice

“Breast is best.”

Except that it’s not.

How did an ideological claim, never proven with scientific evidence, get accepted as a “fact”?

Nissen and colleagues explain the canonization of false facts:

Science is a process of collective knowledge creation in which researchers use experimental, theoretical and observational approaches to develop a naturalistic understanding of the world. In the development of a scientific field, certain claims stand out as both significant and stable in the face of further experimentation. Once a claim reaches this stage of widespread acceptance as true, it has transitioned from claim to fact. This transition, which we call canonization, is often indicated by some or all of the following: a canonized fact can be taken for granted rather than treated as an open hypothesis in the subsequent primary literature; tests that do no more than to confirm previously canonized facts are seldom considered publication-worthy; and canonized facts begin to appear in review papers and textbooks without the company of alternative hypotheses.

A claim doesn’t have to be true to be canonized.

Consider:

Breastfeeding professionals are like the surgeons who continued doing radical mastectomies after lumpectomy was shown to produce the same outcome.

Breast cancer is best treated with radical mastectomy.
Acid causes ulcers.
Episiotomies prevent vaginal tears.

Even though these claims were considered medical “facts” for decades, they were never true. It took years of near heroic work (often while enduring widespread scorn) for doubters to overturn conventional wisdom.

It’s only a matter of time before “breast is best” is added to the list.

The process is under way as we speak as hospitals grapple with tens of thousands of newborn readmissions each year at the cost of hundreds of millions of dollars. The process is furthered by the growing body of papers detailing harms of aggressive breastfeeding promotion including in hospital suffocation of newborns in mothers’ hospital beds as well as skull-fracturing falls from those beds in the wake of closing well baby nurseries. But what will almost certainly seal the de-canonization of “breast is best” is that almost none of the predicted benefits of breastfeeding have come to pass. With the exception of premature infants, breastfeeding has never been shown to save lives, prevent severe disease, or reduce healthcare costs.

If the claims were never true, why was it so hard to convince the believers that they were never true?

White hat bias

The current era of breastfeeding promotion reflected disgust with Nestle’s unethical marketing of formula in Africa. Early breastfeeding promotion efforts in the 1980’s did NOT assert that breast was best, merely that breast was better than formula prepared with contaminated water. The International Code of Marketing of Breastmilk Substitutes was designed to restrain and punish formula companies.

In a very real sense, breastfeeding promotion had its origins in an ideologically motivated boycott of formula manufacturers. That anti-corporatist zeal persists to this day but the demonization of formula manufacturers has transmuted into the demonization of formula itself.

The professionalization of breastfeeding support

In the mid-1980’s, La Leche League realized that it could monetize the advice and support it had been providing for free for decades. The lactation professional was born. While there is certainly nothing wrong with women earning money to promote breastfeeding, it created a group whose economic health was directly tied to the spread of its promotion efforts. It is hardly surprising then that the moralization of breastfeeding followed swiftly upon its monetization. Although breastfeeding professionals have been scathing in their criticism of allowing formula companies to market within hospitals, they have not ended marketing efforts, merely replaced them with their own product: breastfeeding.

Publication bias

As Nissen et al. explain:

Publication bias arises when the probability that a scientific study is published is not independent of its results…

Publication bias is pervasive. Authors have systematic biases regarding which results they consider worth writing up… Journals similarly have biases about which results are worth publishing.

What kinds of results are most valued? Findings of statistically significant differences between groups or treatments tend to be viewed as more worthy of submission and publication than those of non-significant differences. Correlations between variables are often considered more interesting than the absence of correlations. Tests that reject null hypotheses are commonly seen as more noteworthy than tests that fail to do so. Results that are interesting in any of these ways can be described as “positive”.

Publication bias in breastfeeding research means that it is extremely difficult to publish a study that doesn’t conclude breast is best. This bias is exacerbated by the existence of journals whose purpose is to promote breastfeeding. Those journals — the primary sources for scientific information about breastfeeding — rarely publish any studies that dare to question the canonization of false facts about breastfeeding.

Sloppy research methods

As I have noted many times, most breastfeeding research is appallingly bad. It is weak, conflicting and riddled with confounding variables. The central claims are based on the massive extrapolation of small studies as well as unvalidated mathematical models, both of which assume causation whenever correlation is observed. Almost none of it has been consistently reproduced by follow up investigations.

That doesn’t mean that breastfeeding professionals are lying. They truly believe that breast is best, just like generations of surgeons who believed radical mastectomy is best and like generations of internists who believed that acid caused ulcers. There is no shame in canonizing facts that turn out to be false. The only shame is refusing to consider the possibility that you are wrong when new data don’t confirm your beliefs.

That’s where breastfeeding professionals are now. They are like those surgeons who continued doing radical mastectomies after lumpectomy was shown to produce the same outcome. They may believe in their own righteousness but the result is unnecessary suffering.

“Breast is best” is a false fact. Let’s stop pretending otherwise.

  • Allison

    I’m not a doctor – I have developmental dyspraxia, I’m not someone you want coming anywhere near a scalpel. But I am a scientist. Specifically, a chemist.

    I often say in the lab that science always plays by its own rules – the trouble is that scientists like to pretend that we know what all of those rules are. We don’t, and we likely never will. It takes a lot of research, a lot of data, and a lot of time for a fact to be elevated from ‘what we think we know’ to ‘what we know’. There’s a reason why, centuries after Sir Isaac Newton’s death, universal gravitation is still a theory, not a law.

    Recently in the lab, I noticed that we were having trouble with an experiment intended to detect the presence of chlorine compounds. We had recently switched to a new type of joint grease for getting a good seal on our glassware, and I hypothesized that either the new grease had a chlorine based contaminant or one of the reagents (known to be sensitive to air) had gone bad. So I ordered a new bottle of reagent, and ran six experiments side by side, using both the old and new reagent, and also using both the grease and a non-reactive Teflon seal. And… I ran two blanks, that had no sample, no chlorine standard, and no grease, using both the old and the new reagent. That blank was key – I got a positive result off of my negative control – the problem was not the reagent, nor was it the grease. Both of my probable hypotheses were incorrect. Further experimentation is required, but the problem is most likely a side-reaction occurring with carbon dioxide in the air. The science was absolutely being consistent with its own rules – but the scientist didn’t know what all of the rules were.

    And sometimes, SCIENCE DOES THAT. Sometimes, what we think we know is WRONG. A GOOD scientist will change their hypothesis to fit the data, rather than changing the data to fit their hypothesis.

  • fiftyfifty1

    “tests that do no more than to confirm previously canonized facts are seldom considered publication-worthy”

    Which is interesting in the case of breastfeeding because researchers continue to churn out crappy studies that say that Breast is Best and they continue to get published. I think the reason is that breastfeeding has failed to be canonized in one area: among those actually doing the breastfeeding. Sure initiation rates have gone way up. Sure mothers give lip service to the party line. But myriad women find out, one by one, that breast is NOT best in their own cases. They may feel guilt about it but most mothers, when exclusive breastfeeding is failing, switch to what gives them better results.