Re: “A Critical Review of the Baby-Friendly Hospital Initiative Is in the Works” by Eidelman ( Breastfeed Med 2018;13(9):557–558)

We want to express our dissenting opinion against the editorial published in the November issue of Breastfeeding Medicine entitled “A Critical Review of the Baby-Friendly Hospital Initiative Is in the Works.” ¹

There is ample evidence that the Baby-Friendly Hospital Initiative (BFHI) is associated with increased duration and exclusivity of breastfeeding in a dose–response manner in relation to the Ten Steps.^2–8 A 2016 worldwide systematic review, specifically including the United States, also found the same dose–response effect with respect to the Ten Steps. ⁹ The PROBIT trial, a randomized controlled trial, showed that dyads in the baby-friendly-type intervention were far more likely to be exclusively breastfeeding at 3 months (6% versus 43%) and to be breastfeeding to any degree at 12 months (20% versus 11%), and that infants had better health outcomes, ¹⁰ and better cognitive development. ¹¹ A recent comprehensive review of BFHI done by the U.S. Agency for Healthcare Research and Quality ¹² confirmed similar breastfeeding outcomes. This dose–response effect shows the effect of evidence-based care bundles, in which quality initiatives are applied together to achieve best patient outcomes. ¹³

The BFHI is essential to reducing breastfeeding disparities, ¹⁴ and in the United States, its implementation is an essential part of decreasing breastfeeding disparities leading to health injustice and alarming rates of maternal and infant mortality especially among the African American and Native American communities. Recent interventions to increase the numbers of baby-friendly hospitals in African American communities in U.S. states with high levels of poverty have shown a marked increase in breastfeeding rates: Louisiana went from zero births in baby-friendly hospitals in 2007 and 15% exclusive breastfeeding rate at 3 months (2004 birth cohort), to 42% of births in 2018, with a 39% exclusive breastfeeding rate at 3 months (2015 birth cohort). ¹⁵

The editorial cites a controversial review ¹⁶ that is described as comprehensive and detailed; however, it would be categorized as a Level of Evidence 5 (lower level) by the Oxford Center for Evidence-Based Medicine (OCEBM), ¹⁷ as it is not a systematic review but just a review of cherry-picked studies garnished with the authors' own opinions and suggestions, which were not scientifically based. The editorial further states that “one would surmise the reviewed published studies were of either historical controls or comparison of facilities,” and this is used to question the BFHI's efficacy by arguing there is a lack of randomized control trials. It is worrisome, however, that the most recent systematic reviews of BFHI^9,18 were neither included in the cited review ¹⁶ nor examined in this editorial. ¹ These systematic reviews clearly contradict the quoted statement made in both articles: “there is no clear evidence to support the positive influence of the BFHI to improve breastfeeding.” ¹⁹

Moreover, the editorial implies that only randomized controlled trials (RCTs) may be considered good quality evidence. However, for some years now, this type of evidence is considered best suited to answer therapeutic effects, such as medication effectiveness. But to analyze the evidence on breastfeeding interventions as well as on other health issues, other types of evidence such as epidemiologic evidence may be better suited. As the OCEBM clearly acknowledges, placing RCTs categorically above observational studies has been widely criticized as simplistic.

There are several other flaws in this editorial that also need to be mentioned. The BFHI does not require “strict adherence to the Ten Steps” and certainly does not impose actions on mothers and families such as forcing mothers to room in with their infants when they choose not to. Rather, the BFHI asks health professionals to offer the best quality evidence care and to provide families with best and more current information, free of commercial interests to enable mothers and infants to attain their goals and make informed decisions. In the case of Step 7, BFHI literally states as follows: “to enable mothers and infants to remain together and practice rooming in.” The BFHI offers a structured evidence-based quality of care model that aims to improve the care that is offered to mothers and infants in maternity facilities all over the world. It may not be the only way, but, at present, it certainly is the best model, and as such is endorsed by many international health care organizations.^20–23 As the final arbiter of the Ten Steps of the BFHI, the World Health Organization periodically reviews the evidence for Ten Steps for safety and efficacy and revises them, most recently in 2017,^24,25 thus addressing the very concerns raised in the editorial.

As readers may be aware, an editorial is just an opinion piece, and as such carries the lowest level of evidence. However, rigorous scientific journals such as Breastfeeding Medicine should take care to publish commentaries that are based on the best available evidence. This editorial does not do so, and certainly does not match the type of rigorous systematic review, which the Academy of Breastfeeding Medicine (ABM) requires in writing our own clinical protocols.