A Recent Article in the Leading Breastfeeding Trade Journal Misrepresents My Work
Response to Subramani et al's "Ethical Issues in Breastfeeding and Lactation Interventions: A Scoping Review"
Subramani et al’s recent article “Ethical Issues in Breastfeeding and Lactation Interventions: A Scoping Review” (J Hum Lact 2023 Dec 12) reports incomplete and erroneous results of the described review. A PubMed search for the terms “ethical” and breastfeeding,” as the authors report doing, should have returned two of my publications; only one was cited. In addition, the authors incorrectly characterize my cited paper as not following an established research methodology and having insufficient discussion of quality (Table 1). These mischaracterizations reflect apparent miscomprehension of the substance of my peer-reviewed articles on breastfeeding, as well as of broad-reaching issues of method and quality assessment. They also lead Subramani et al to ignore the most important point in current scientific literature on breastfeeding and ethics: Insufficient milk intake in breastfed neonates is common, causes widespread preventable harm, and may substantially contribute to morbidity and mortality in previously healthy neonates worldwide.
The Breastfeeding Myth
Subramani et al recognized six themes in scientific literature on breastfeeding and ethics:
(1) respect for maternal autonomy and experiences, (2) counseling and informed consent process, (3) evidence and effectiveness of breastfeeding interventions, (4) ethical issues related to health communication, (5) ethical acceptability of financial incentives, and (6) children’s right to breastmilk.
These themes work within the “breast is best” mythology to promote it. By contrast, my work demolishes this myth using logic and evidence. In short, the breastfeeding myth is a modern, Western creation. Based on historical and cultural ignorance and bad science, this mistaken set of beliefs causes common and preventable harm to infants due to the widespread insufficiency of breastmilk alone for meeting infants’ basic needs for food and water, especially in the first days of life.
My first breastfeeding article, “Breastfeeding Insufficiencies: Common and Preventable Harm to Neonates” (Cureus 13(10): e18478) synthesizes historical and scientific sources to critique existing infant feeding science as insufficiently evidence-based, and the historically anomalous infant feeding paradigm it promotes as a tragic modern mistake. Among focal sources is PROBIT, the famous breastfeeding trial that wasn’t. My point was that, because studies including this supposedly gold-standard one did not account for potential confounds such as maternal and infant health, breastfeeding insufficiencies, and accidental starvation, the evidence they report could be interpreted as showing preventable harms rather than benefits from exclusive breastfeeding. (See also these additional posts and my invited May 2023 University of Kent talk, slides 3-10, for more details.)
This causal inferences critique shows bad science at the heart of current infant feeding guidance, practices, and beliefs. It is peculiar to have presented an argument, using top-level methods criticisms, for why the central claim of modern infant feeding science (that breastfeeding has myriad proven benefits) is insufficiently evidence-based — and to be mischaracterized in response as lacking in method and inadequately discussing quality. My thesis is precisely that current infant feeding science misses the thinking about causality that underpins scientific quality.
Current evidence does not establish causal benefits of breastfeeding. It does, however, suggest substantial risks of insufficient milk intake associated with current breastfeeding norms, especially the modern Western emphasis on exclusivity. Chief among these risks are preventable (re)hospitalizations for jaundice/hyperbilirubinemia, hypernatremia/dehydration, and hypoglycemia, all of which may risk permanent neurodevelopmental harm or death. Furthermore, there are good reasons to suspect these risks “disproportionately affect many subgroups along some usual lines of structural power disparity, with racial and geographic outcome differences overlapping socio-economic and international developmental ones,” as my next article explored in detail…
More Support for Harm Prevention
My second breastfeeding article, “Neonatal Jaundice and Autism: Precautionary Principle Invocation Overdue” (Cureus, 14(2): e22512), synthesized epidemiological evidence from and pertaining to the three PubMed-indexed meta-analyses on neonatal jaundice and autism risk. It reported:
Amin et al., Jenabi et al., and Kujabi et al. consistently found that jaundice in near and full-term neonates may risk substantial autism increase [1-3]. Their overlapping confidence intervals based on included studies’ pooled ORs estimated possible risk increases up to 67%, 68%, and 76%, respectively.
This article included in-depth discussion of relevant ethical issues, chiefly (1) persistent racial and geographic disparities as they relate to jaundice risk, subgroup effects, and data availability, and (2) the importance of harm prevention.
The case for preventing possible jaundice-associated harms is strong. Insufficient milk intake associated with breastfeeding is the most frequent cause of jaundice in previously healthy near- and full-term neonates. Jaundice associated with insufficient milk intake due to breastfeeding insufficiencies is fully preventable and trivially treatable with early, adequate, and often supplemental milk. This means:
Feeding neonates adequately may play an important role in preventing autism and other neurodevelopmental disorders including attention deficit hyperactivity disorder, cerebral palsy, epilepsy, hearing impairment, learning disorders, and mood disorders. Precautionary principle invocation is overdue.
Additional Ethical Issues
This article also included in-depth discussion of quality, including identifying several mistakes in Kujabi et al 2021:
that, if corrected, would support different conclusions. Its authors misinterpreted statistical significance test results to claim no jaundice-autism association when their analysis showed a quite sizeable possible effect. They also misused the funnel-plot test for publication bias, claiming without merit that its results indicated a high risk of publication bias. Their claim that ‘there is no evidence to suggest jaundice should be treated more aggressively to prevent autism’ was based on these mistakes, incorrect temporal sequencing, and apparent lack of consideration of evidence on jaundice severity, etiology, and different prevention, screening, and treatment approaches. Such mistaken conclusions are likely to contribute to practices that endanger neonates.
In addition, it criticized Kujabi et al’s reliance on “risk of bias as a binary quality metric [leaving] open questions about both bias and quality,” and raised ethical concerns about Kujabi et al’s and Pediatric Research’s failures to respond to the totality of these concerns, possibly contravening “relevant publishing ethics [258] and retraction guidelines [259], which emphasize the importance of protecting the integrity of the scientific record.”
Listen to Mothers, Protect Infants
So far, this response has shown that Subramani et al’s review omitted research that its search strategy should have included (Wilde 2022), and misrepresented research it did include (Wilde 2021). These errors of omission and comprehension then led the authors to the following conclusion:
This review illustrated that, while a wide range of ethical arguments were examined, the emphasis has been primarily on accounting for mothers' experiences and lactating persons' choices, as well as achieving public health objectives relating to infant nutrition in breastfeeding interventions. To effectively and ethically implement breastfeeding and lactation interventions, we must consider the social, economic, and cultural contexts in which they occur. One key learning identified [sic] was that women's experiences were missing in these interventions and, in response, we suggest moving beyond the dichotomous approach of individual health versus population health.
This misses the main empirical and ethical point of my work: Meeting newborns’ medical needs for sufficient food and water to prevent harm is the first goal of infant feeding. Infant feeding practices that frequently, medically underfeed infants risk common, preventable, and in some cases permanent harm including neurodevelopmental disability and death. These practices violate the Hippocratic oath to “first do no harm.” Current norms risk this harm in exchange for no established benefit. The breastfeeding myth is a pseudoscientific edifice of insufficiently evidence-based beliefs driving these wildly unethical practices.
Literature that says otherwise ignores mothers’ experiences — of common breastfeeding insufficiencies that cause accidental starvation and resultant harm. It is not enough to note that the literature on breastfeeding and ethics suggests we should listen to mothers’ experiences as a matter of ethics, without actually listening to mother’s experiences as a matter of ethics.
While my causal inferences critique is original within infant feeding science, it highlights the importance of heeding many other critics of the breastfeeding myth who have long raised similar harm prevention concerns (e.g., University of Pittsburgh School of Medicine Professor of Pediatrics and Director of Pediatric Nephrology, Dialysis, and Kidney Transplant Programs Michael Moritz and Fed Is Best co-founder, American Board-Certified Emergency Physician, and Newborn Brain Injury and Breastfeeding Complications Investigator Christie del Castillo-Hegyi). On this score, it is ironic that Subramani et al emphasize that “women’s experiences were missing” in breastfeeding promotion practices — while themselves erasing the voices of many mothers, like Castillo-Hegyi and myself, whose children were preventably harmed by the breastfeeding myth.
My work’s primary emphasis is on preventing harm to infants. It is scientifically and ethically unacceptable to omit this point from a review of ethical issues in breastfeeding science. Indeed, miscategorizing my point as one of “women’s experiences” does a disservice to the children whose health is put at risk by current infant feeding practices. It also does a disservice to the families who love them.
There is not a single mention in Subramani et al of the common and preventable harm to neonates that results from current infant feeding practices. Disciples of the breastfeeding myth must stop spinning valid, existential concerns about children’s safety as a few disgruntled moms whining about hurt feelings.