Best make yourself a cuppa, it’s a long read!  If you’re pushed for time you can read our edited version.

Infant Feeding Support UK response to the public consultation on the draft of the document: Protection, Promotion, and Support of Breastfeeding in Facilities Providing Maternity and Newborn Services: The Revised Baby-friendly Hospital Initiative 2017[*]

Infant Feeding Support UK (IFSUK), currently under the process of registration as a UK based charity, is a group of parents and scientists who aim to communicate safe, unbiased and science-based infant feeding information to parents and carers throughout the UK.

Neither IFSUK nor I Support You, the parent-focused arm of IFSUK dedicated to communicating scientific findings on infant feeding, has any links to or income from any companies or organisations which profit from breast milk substitutes, infant food or infant feeding products including and not limited to bottles, teats, dummies, shields and breast pumps. We also do not have any contact with such companies.

Our response submission relates specifically to the infant feeding environment within the UK.

(Quotes from the document are shown in italics, our response is written underneath.)


Section 1.1: Breastfeeding Matters

The risk of dying in the first 28 days of life is 41% higher for newborns who initiated breastfeeding 2–23 h after birth, and 79% higher for those who initiated 1 day or longer after birth, compared to newborns who were put to the breast within an hour of birth.

This statement is evidenced by studies done in Ghana, India and Tanzania (BFHI reference 2). Given the much smaller burden and different causes of neonatal mortality in developed country settings such as the UK, it is unlikely that the magnitude of these effects reaches similar levels in such settings, or would have relevance to mortality in term babies. Therefore the relevance of these mortality benefits for developed countries is not clear.

Exclusive breastfeeding for 6 months provides the nutrients and energy needed for growth and development.

UK guidance from its Scientific Advisory Committee on Nutrition (SACN) specifies that exclusively breastfed babies require vitamin D supplementation from birth since the levels of this nutrient in breastmilk are insufficient for adequate nutrition1.

Improving these practices could save over 820000 lives a year.

This modelling estimate is based on data from low and middle income countries only. It is unclear that the number of lives saved through scaling up breastfeeding in developed countries will reach similar levels of impact.


Section 1.5: Revision of the Ten Steps to Successful breastfeeding and the operational guidance

Revision of the Ten Steps… (p10/11): Using the standard WHO guideline development process (32), a guidelines development group was created. Systematic literature reviews were commissioned on each of the Ten Steps.

Reference 32 in the BFHI document (WHO guideline development process) states that GRADE should be used, which would entail stating the strength of recommendations, the quality of evidence for each recommendation, the interplay between benefits and harms, and issues to do with values and preferences of stakeholders (i.e., families). It is not clear that this has been done, or at least this is not reported in the BFHI document. Nor are other GRADE items (consistency, directness, or imprecision of evidence), represented in the document. Directness is a key consideration: effects of steps such as the advice to avoid supplementation, may have very different effects in developed versus developing country settings. It is unclear why full GRADE tables are not presented for each of the revised Ten Steps.


Section 2.1.2: Immediate postnatal care

It [skin to skin] should be uninterrupted for at least 60 minutes

No evidence is cited for the requirement that at least 60 minutes are needed, as opposed to skin-to-skin for as long as the mother wishes and/or for as long as can be safely done. The guidance does not specify options for enabling skin-to-skin in such a situation as the mother is not well enough to carry out this function; i.e., whether it can be enabled with a different caregiver (e.g., father). It is good that the guidance now specifies that “sensible vigilance and safety precautions should be taken” (during skin to skin). However, it is unclear why studies such as Pejovic and Herlenius (2013)2 and Herlenius and Kuhn (2013)3, have not been cited. It may be appropriate to include clearer direction within the Ten Steps for health professional and caregiver training to increase awareness regarding the potential for Sudden Unexpected Postnatal Collapse to occur during skin to skin, and what practices should be adopted to prevent this.


Section 2.1.3: Support with breastfeeding

Responsive feeding ….puts no restrictions on the frequency or length of the infants’ feeds, and mothers are advised to breastfeed whenever the infant is hungry or as often as the infant wants. Scheduled feeding …is not recommended.

The systematic review by Fallon et al (2016)4 investigating this topic on behalf of BFHI did not find any high quality evidence comparing baby-led and scheduled breastfeeding, and concluded: “it is recommended that no changes are made to current practice guidelines without undertaking robust research”. It is unclear what level of evidence therefore has driven the BFHI recommendation that scheduled feeding is not recommended (no evidence is stated). GRADE tables for this recommendation would make clearer the evidence base for what is proposed.


Section 2.1.4: No supplements

We feel that this recommendation is problematic from a number of viewpoints and will try to go through each aspect in turn, starting with the evidence cited in the BFHI document for the recommendation.

This states: “Newborns that are fed other foods or fluids will suckle less vigorously at the breast and thus inefficiently stimulate milk production…”. No evidence is cited for this statement. As far as we are aware, the established understanding of lactation physiology suggests that once stage II lactogenesis has commenced (typically from around two days after birth), milk production is driven by a supply and demand cycle. However, until stage II lactogenesis, milk supply is hormonally driven and it is not evident that supplementation given prior to this event will interfere either with colostrum production or with the onset of stage II lactogenesis5,6.

The recommendation also states: “Supplementation with artificial milk significantly alters the intestinal microflora (47a)”. Reference 47a is not an original research study but a narrative review (not a systematic review) and only cites one article – Mackie et al, 19997 – for the claim that supplementation interferes with infant gut microbiota. Mackie et al. is itself not a research article either but again a narrative review. It cites three articles for the claim about effects of supplementation on microbiota, of which two seem to be research articles, but only one actually studies supplementation. This study is Bullen et al, 19778 and it is far from clear that it provides sufficient quality and generalisability of evidence to inform global policy on infant feeding. Bullen et al. studied 47 babies of which 13 were breastfed, 19 were formula fed and 15 were breastfed and received formula supplementation in the first week of life. It is stated that the formula milks were “reconstituted as recommended by the manufacturer”, hence, not comparable to current ready-to-use sterile formula milks such as would be used for non-breastfed newborn babies currently in developed country maternity units. The study cited8 did not look at any objective infant health outcomes. In the 40 years since the study was published, advances in microbiological techniques mean we now have a much more comprehensive understanding of the infant gut microbiome. The significance on infant health of changes in the microbiota are entirely unclear; the microbiome is influenced by multiple factors (including methodology) and there is not yet a general consensus on what a ‘normal’ microbiome is due to the individual variation in composition. It is concerning that a single, very small study using outdated methods and with little relevance to current practice (with regard to the type of formula used and its preparation) may be influencing global policy on infant feeding.

In addition, foods and liquids may contain harmful bacteria and carry a risk of disease.

This may well be true in settings where cultural practices include the provision of unsafe supplements, or where infant formula cannot be safely prepared. However, this is very unlikely to be true in developed country settings. We would suggest that BFHI guidance is clear on what supplements should not be given to breastfed babies (e.g., nutritionally inadequate, unsafe feeds such as supplementary water, unmodified animal milks, prelacteal feeds etc). At the same time guidance can ensure that setting-specific recommendations be developed on what form safe supplementation should take when it is needed.

Mothers should be discouraged from giving any food or fluids other than breast milk, unless medically indicated.

We feel this overall recommendation is problematic in developed country settings. The BFHI report cites very little evidence for this recommendation. We note that even under conditions of excellent support for breastfeeding, developed-country population samples of postnatal mothers, particularly primiparas, typically show very high rates of delayed onset of lactation9 even if they give birth in a Baby Friendly facility and receive good lactation support10. Therefore, in similar settings, the claim that “very few conditions of the infant or mother…necessitate the use of breast-milk substitutes” is unlikely to be true; a high proportion of such babies may need supplementation, if only temporarily, to avert problems associated with insufficient intake.

The recommendation to avoid supplementation until “medically indicated” (recommendation 7) entails that, in cases where the mother’s milk or colostrum is insufficient, staff wait until a baby is medically unwell before mandating supplementation. This is problematic, and two recent randomized trials not cited in the BFHI document11,12 have shown that advice to supplement babies who have experienced more than 5% weight loss does not interfere with later breastfeeding success, and rather, actually seemed to increase the likelihood that a baby is breastfed later on11. Indeed, a further study suggests that reducing the likelihood of excessive weight loss through insufficient intake reduces maternal anxiety, thus improving her breastfeeding confidence and success subsequently13. Supplementing at 5% pre-empts a medical indication yet this approach seemed to be beneficial for the infants randomized to receive supplement11.

In the UK, readmission rates in infants as a result of problems associated with insufficient feeding (e.g., emergency jaundice admissions) are rising14 even though breastfeeding rates have risen in recent years and increasing numbers of maternity units are Baby Friendly accredited. The numbers of such admissions (16,000+/yearly) are not insignificant compared with admissions due to conditions for which breastfeeding reduces relative risk, such as upper respiratory tract infection and diarrhoeal illness (20,000+).

Given that in the UK substantial proportions of mothers are likely to have problems establishing sufficient milk production, if only temporarily, we suggest that BFHI modify its guidance to be clarify that mothers wishing or needing to supplement should be supported to do so safely. Existing evidence shows that, under an environment of advocacy for strictly exclusive breastfeeding, mothers who subsequently need to use formula experience substantial problems with their emotional well-being15. The consequences of this at the severe end of the spectrum may be significant. We are aware of mothers who experience postpartum depression or anxiety, significantly impeding their ability to bond with and care for their babies, as a result of the feelings of failure that result from needing to use formula. Mothers in this situation will be acting appropriately, ensuring they fully provide for the baby’s needs, whilst the emotional consequences come about from her experience of being “trained” to believe that this approach is suboptimal and damaging to the baby’s health. Ideal guidance for UK-like settings would ensure that healthcare professionals and families receive consistent messaging, being clear that full nutritional adequacy for the baby is essential at all times and that where mother’s milk is insufficient, safe supplementation can and should be provided to meet those needs.


Section 2.1.6: Care at discharge

It is surprising that the guidance does not include specific recommendations regarding the prevention of adverse outcomes as a result of insufficient milk intake (e.g., excessive weight loss, jaundice, hypernatremia). For example, the evidence relating to approaches such as regular weighing16,17 or transcutaneous bilirubin measurements18 could be summarised to provide clear recommendations as to the efforts that postnatal care providers should take to detect newborns with complications of insufficient milk intake as early as possible.


Section 2.2.4: Monitoring and data-management systems

This section states “Facilities should aim for at least 90% early initiation and exclusive breastfeeding”. However, in the UK, the most recent Infant Feeding Survey showed that 81% of mothers started breastfeeding (with 69% exclusively breastfeeding at birth). Specifying that facilities must achieve 90% exclusive breastfeeding therefore entails changing the wishes of what a fairly large proportion of mothers in the UK want to do. No evidence is presented in the document to support this target of 90%, or its applicability to all settings globally. In settings where substantial proportions of mothers have prior breast surgery, for example, or other contraindications (e.g., a high HIV prevalence in areas where avoidance of breastfeeding is the national policy for HIV+ve mothers), this target may be physically impossible and therefore have adverse consequences.  In addition, in circumstances where exclusive breastfeeding is hindered by problems such as a high prevalence of delayed onset of lactation, attempts to achieve a facility target of 90% may well create substantial problems with readmission of undernourished infants.

Breastfeeding rates alone cannot tell us about the success of the BFHI as a whole. In order to do this it is essential that we consider the physical and mental wellbeing of families delivering their babies at BFHI Accredited hospitals.  To this end, data collected should include outcomes associated with infant feeding including satisfactory weight gain, readmissions for hypoglycaemia, hypernatremia and hyperbilirubinemia, maternal mental health relating to their infant feeding experience, and maternal physical health e.g., infections and other morbidities associated with breastfeeding.


Section 3.2: Policies and professional standards of care

At a minimum, standards for family medicine, obstetrics, paediatrics, neonatology and nursing should lay out these steps as basics of care for all newborns.

It may be valuable to also include midwifery and dietetics (and, in the UK, health visitors – child public health nurses) within these cadres.


Section 3.4: External Assessment

Initially, the external assessment should review documentation on the two sentinel outcome indicators.

This refers to the items listed in Table 1: the percentage of term infants who were put to the breast within one hour of birth – >90%; the percentage of infants who received only breast milk (either from their own mother or from a human milk bank) throughout their stay at the facility – >90%

As highlighted above, the validity of these indicators and their appropriateness when taking into account maternal wishes and the nature of the postnatal population (medical contraindications, etc) is unclear. Essentially, this indicator could result in maternity staff pressuring mothers to breastfeed when this is not their intention (as highlighted, UK data suggests 81% of mothers started breastfeeding at birth and 69% did so exclusively). It is not evident how maternity units will support mothers and babies in their individual choices and needs when this unit-level target is set. It may be more appropriate to ensure that, for mothers in a facility who have stated they wish to breastfeed, a high proportion of such mothers are helped to put the baby to the breast within one hour of birth, and are given detailed guidance and support in doing this (essentially step 3, recommendations 3-11).


Section 3.7: National monitoring

At present, national monitoring of infant feeding trends in the UK has been decommissioned with the withdrawal of the Infant Feeding Survey19. We support reinstatement of this research survey and commend BFHI in including a specification that national authorities undertake routine monitoring in relation to infant feeding. However, we would also suggest that routine monitoring include assessment of objective infant health outcomes related to adequacy and safety of feeding, such as data related to the proportion of babies readmitted to neonatal units with problems related to feeding (e.g., jaundice, hypernatremia, excessive weight loss and hypoglycaemia), and frequency of admissions related to preventable infant infections (principally diarrhoea, upper respiratory tract infections etc).


Key issues missing from this document  

  • As mentioned above, it is surprising that the guidance does not include specific recommendations regarding the prevention of adverse outcomes as a result of insufficient milk intake (g., excessive weight loss, jaundice, hypernatremia).
  • As mentioned above, research carried out in the UK into the experiences of formula feeding women show that a worryingly high percentage of mother’s experienced negative emotions as a result of their decision to use formula15. These negative emotions included feelings such as guilt, dissatisfaction and stigma. We would like to say far clearer guidance on how care givers can reconcile statements such as “Mothers should be discouraged from giving any food or fluids other than breast milk, unless medically indicated” with those such as “Families must receive quality and unbiased information about infant feeding. Facilities providing maternity and newborn services have a responsibility to promote breastfeeding, but they must also respect the mothers’ preferences and provide support for mothers to successfully feed their newborn in the way they choose.”
  • While this document presents research into the effectiveness of the BFHI in a very positive light, it fails to acknowledge less positive and critical research or the limitations of research presented. For example, the BFHI document states that “In Belarus, a group-randomized trial undertaken at the end of the 1990s increased the rate of exclusive breastfeeding at 3 months to 43% in hospitals that implemented the Ten Steps to Successful Breastfeeding, compared to only 6% in the hospitals that did not receive the intervention. While this may well be the case, it should be noted that this cannot be used as evidence that the BFHI itself actually works because it compared no breastfeeding support to breastfeeding support provided in the form of BFHI Ten Steps. In order to establish the effectiveness of the BFHI it would be necessary to compare that to other methods of breastfeeding support. The JAMA-USPSTF report20 found that “Unfortunately, there is now emerging evidence that full compliance with the 10 steps of the initiative may inadvertently be promoting potentially hazardous practices and/or having counterproductive outcomes”. Furthermore, in Ireland, studies have found no difference in breastfeeding rates according to BFHI status, and that the observed increases in breastfeeding rates are more likely to be dues to the characteristics of mothers21. Indeed, the Irish government has recently paused the funding for the BFHI Initiative.

We would strongly urge that these criticisms are addressed as a means of presenting a far more robust case for the implementation of the BFHI.


If you have any additional comments please add those here:              

  • We would like to commend the reconsideration of pacifier use in this review and ask that this recommendation also extends to use of bottles where supplementation needed. There is currently no evidence that artificial teats interfere with breastfeeding and a Cochrane systematic review showed no advantage to cup feeding22. Therefore, when supplementation is indicated supplemental feeds can be given by bottle.
  • We were unable to fully appraise all the literature cited in the BFHI document due to the short consultation time.



  1. The Scientific Advisory Committee on Nutrition (SACN) recommendations on vitamin D. (2016).
  2. Pejovic NJ and Herlenius E. (2013). Unexpected collapse of healthy newborn infants: risk factors, supervision and hypothermia. Acta Paediatrica, 102(7):680-688.
  3. Herlenius E and Kuhn P. (2013). Sudden unexpected postnatal collapse of newborn infants: a review of cases, definitions, risks, and preventive measures. Translational Stroke Research, 4(2):236-247.
  4. Fallon A, Van der Putten D, Dring C, Moylett EH, Fealy G and Devane D. (2016). Baby-led compared with scheduled (or mixed) breastfeeding for successful breastfeeding. Cochrane Database of Systematic Reviews, Issue 9. Art. No.: CD009067.
  5. Neville MC and Morton J. (2001). Physiology and Endocrine Changes Underlying Human Lactogenesis II. Journal of Nutrition, 131: 3005S–3008S.
  6. Kulski JK, Hartmann PE, Martin JD and Smith M. (1978) Effects of Bromocriptine Mesylate on the Composition of the Mammary Secretion in Non-Breast-Feeding Women. Obstetrics and Gynaecology, 52(1).
  7. Mackie RI, Sghir A and Gaskins HR. (1999). Developmental microbial ecology of the neonatal gastrointestinal tract. American Journal of Clinical Nutrition, 69(5):1035S-1045S.
  8. Bullen CL, Tearle PV and Stewart MG. (1977). The effect of “humanized” milks and supplemental breast feeding on the faecal flora of infants. Journal of Medical Microbiology, 10:403–13.
  9. Dewey KG, Nommsen-Rivers LA, Heinig MJ and Cohen RJ. (2003). Risk factors for suboptimal infant breastfeeding behavior, delayed onset of lactation, and excess neonatal weight loss. Pediatrics, 112(3 Pt 1):607-19.
  10. Nommsen-Rivers LA, Chantry CJ, Peerson JM, Cohen RJ and Dewey KG. (2010). Delayed onset of lactogenesis among first-time mothers is related to maternal obesity and factors associated with ineffective breastfeeding. American Journal of Clinical Nutrition, 92(3):574-84.
  11. Flaherman VJ, Aby J, Burgos AE, Lee KA, Cabana MD and Newman TB. (2013). Effect of early limited formula on duration and exclusivity of breastfeeding in at-risk infants: an RCT. Pediatrics, 131(6):1059-65.
  12. Straňák Z, Feyereislova S, Černá M, Kollárová J and Feyereisl J. (2016). Limited Amount of Formula May Facilitate Breastfeeding: Randomized, Controlled Trial to Compare Standard Clinical Practice versus Limited Supplemental Feeding. PLoS One, 26;11(2):e0150053.
  13. Flaherman VJ, Beiler JS, Cabana MD and Paul IM. (2016). Relationship of newborn weight loss to milk supply concern and anxiety: the impact on breastfeeding duration. Maternal and Child Nutrition, 12(3):463-72.
  14. Focus on: Emergency hospital care for children and young people. What has changed in the past 10 years? (2017). Health Foundation and Nuffield Trust. ISBN: 978-1-910953-28-0.
  15. Fallon V, Komninou S, Bennett KM, Halford JCG and Harrold, JA. (2017). The emotional and practical experiences of formula-feeding mothers. Maternal and Child Nutrition, 13:e12392.
  16. Boer S, Unal S, van Wouwe JP and van Dommelen P. (2016). Evidence Based Weighing Policy during the First Week to Prevent Neonatal Hypernatremic Dehydration while Breastfeeding. PLoS One, 20;11(12):e0167313.
  17. Iyer NP, Srinivasan R, Evans K, Ward L, Cheung WY and Matthes JW. (2008). Impact of an early weighing policy on neonatal hypernatraemic dehydration and breast feeding. Archives of Disease in Childhood, 93(4):297-9.
  18. Spence J, Bevan C ad Walton E. (2016). G236(P) Introduction of a transcutaneous bilirubinometer into children’s ED – a good initiative? Archives of Disease in Childhood, 101:A129-A130.
  19. Infant Feeding Survey. National Health and Social Care Centre:
  20. Bass JL, Gartley T and Kleinman R. (2016). Unintended Consequences of Current Breastfeeding Initiatives. JAMA Pediatrics, 170(10):923–924.
  21. Brick A and Nolan A. (2014). Explaining the increase in breastfeeding at hospital discharge in Ireland, 2004–2010. Irish Journal of Medical Sciences, 183: 333.
  22. Milne A. (2008). Summary of ‘Cup feeding versus other forms of supplemental enteral feeding for newborn infants unable to fully breastfeed’. Evidence-Based Child Health, 3: 665–666.

*Throughout this document the term ‘BFHI document” will be used to refer to the draft Protection, Promotion, and Support of Breastfeeding in Facilities Providing Maternity and Newborn Services: The Revised Baby-friendly Hospital Initiative 2017.