The challenge of reforming breastfeeding research

Reform concept. Wooden letters on the office desk, informative and communication background

Most breastfeeding scientists and much of the public often consider epidemiologic associations of breastfeeding to represent causal effects that can inform public health policy and guidelines. However, the emerging picture of breastfeeding research is difficult to reconcile with good scientific principles. The field needs radical reform.

If that paragraph seems familiar to readers of scientific journals it is because it is a paraphrase of the lead paragraph of John Ioannidis’ new paper The Challenge of Reforming Nutritional Epidemiological Research.

Most nutrition epidemiological research is nonsense and breastfeeding epidemiological research is a subset of nutrition research.

Ioannidis writes:
[pullquote align=”right” cite=”” link=”” color=”” class=”” size=””]We should ignore hasty statements of causal inference and advocacy to public policy made by breastfeeding researchers.[/pullquote]

In recent updated meta-analyses of prospective cohort studies, almost all foods revealed statistically significant associations with mortality risk…

Assuming the meta-analyzed evidence from cohort studies represents life span–long causal associations, for a baseline life expectancy of 80 years, eating 12 hazelnuts daily (1 oz) would prolong life by 12 years (ie, 1 year per hazelnut), drinking 3 cups of coffee daily would achieve a similar gain of 12 extra years, and eating a single mandarin orange daily (80 g) would add 5 years of life. Conversely, consuming 1 egg daily would reduce life expectancy by 6 years, and eating 2 slices of bacon (30 g) daily would shorten life by a decade, an effect worse than smoking. Could these results possibly be true? Authors often use causal language when reporting the findings from these studies (eg, “optimal consumption of risk-decreasing foods results in a 56% reduction of all-cause mortality”). Burden-of-disease studies and guidelines endorse these estimates. Even when authors add caveats, results are still often presented by the media as causal.

Breastfeeding studies are exactly the same. The breastfeeding equivalent of claiming 12 hazelnuts a day would prolong life by 12 years is claiming that breastfeeding could save 800,000 lives per year. That’s equally nonsensical. It’s based on a mathematical model that has never been validated. Even when formula companies engaged in their unethical campaign to promote formula in Africa, the actual death toll at the peak year of formula promotion was 65,000. That represents 65,000 preventable tragedies, but nowhere near what we might expect if breastfeeding researchers’ claims were true.

Since then according to Paul Gertler whose research established the 65,000 peak death toll:

…[T]he annual death toll has dropped to about 25,000, driven by improved access to clean water in the Southern Hemisphere.

That’s just 3% of the number claimed by breastfeeding researchers.

How do good people end up making such bad claims?

These implausible estimates of benefits or risks associated with diet probably reflect almost exclusively the magnitude of the cumulative biases in this type of research, with extensive residual confounding and selective reporting. Almost all nutritional variables are correlated with one another; thus, if one variable is causally related to health outcomes, many other variables will also yield significant associations in large enough data sets. With more research involving big data, almost all nutritional variables will be associated with almost all outcomes. Moreover, given the complicated associations of eating behaviors and patterns with many time-varying social and behavioral factors that also affect health, no currently available cohort includes sufficient information to address confounding in nutritional associations.


…[T]he literature is shaped by investigators who report nonprespecified results that are possible to analyze in very different ways. Consequently, meta-analyses become weighted averages of expert opinions.

That’s precisely what has happened with breastfeeding research. Researchers continue to make absurd claims about the benefits of breastfeeding based on extrapolations from small studies with multiple confounding variable; this despite the fact that they can find NO EVIDENCE in real life that their claims are true. There is no evidence that breastfeeding rates are correlated in any way with infant mortality and there’s no evidence that increasing breastfeeding rates leads to corresponding declines in infant deaths.

Okay, so perhaps nutrition and breastfeeding researchers have exaggerated various risks and benefits, but what’s the harm?

Nutritional research may have adversely affected the public perception of science. Resources for some of these studies could have been better spent on unambiguous, directly manageable threats to health such as smoking, lack of exercise, air pollution, or climate change. Moreover, the perpetuated nutritional epidemiologic model probably also harms public health nutrition. Unfounded beliefs that justify eating more food, provided “quality food” is consumed, confuse the public and detract from the agenda of preventing and treating obesity.

The harm is even larger in breastfeeding research because breastfeeding has risks as well as benefits. Aggressive efforts to increase breastfeeding rates have led to tens of thousands of neonatal hospital readmissions for dehydration and jaundice, some of which have culminated in infant brain injuries and deaths.

To the extent that nutritional epidemiological research has revealed anything at all, it has demonstrated that there is no such thing as a nutrition “silver bullet.” There is no food, herb or supplement or combination of food, herbs and supplements that magically assures health. That goes for breastfeeding, too. Despite what breastfeeding researchers claim, breastfeeding is NOT a silver bullet for infant health and in many circumstances formula feeding may actually be healthier.

Ioannidis concludes:

Reform has long been due. Data from existing cohorts should become available for reanalysis by independent investigators. Their results should be presented in their totality for all nutritional factors measured, with standardized methods and standardized exploration of the sensitivity of conclusions to model and analysis choices. Readers and guideline developers may ignore hasty statements of causal inference and advocacy to public policy made by past nutritional epidemiology articles. Such statements should be avoided in the future.

The same goes for breastfeeding. Readers and guideline developers should IGNORE hasty statements of causal inference and advocacy to public policy made by past breastfeeding epidemiology articles. And breastfeeding researchers should avoide such claims in the future. They are not true and they may even be harmful.