We need better birth and postnatal care for our new mothers and babies, not separation in the name of risk.  

A response to the recent JAMA article ‘Unintended Consequences of Current Breastfeeding Initiatives’


A recent article published in the Journal of the American Medical Association, entitled ‘Unintended consequences of current breastfeeding initiatives’ proposed that following the Unicef Baby Friendly Ten Steps to Successful Breastfeeding initiative1 may be having unintended negative consequences for infant health and even mortality2.

This headline is naturally alarming and the abstract and first page of the article are being shared widely on social media as evidence against the Ten Steps, and even as far as breastfeeding in general. But what is the evidence behind the article? What is the actual level of risk being discussed? And is the focus on Ten Steps really fair?

Reading the article in context

There are a number of points that need to be considered when reading this article

  1. It is based on the views of three medical doctors, not any evaluation of the Baby Friendly Initiative. This is not to say in any way these views are wrong, or that these individuals are not qualified to publish the paper, but that that this paper does not directly relate to a specific research study. It is designed to stimulate discussion and research, not immediately cast aside all Ten Steps policy.
  1. The authors, and a number of the underpinning references are based on data from the USA, which is unlikely to directly be relevant to different practices in UK settings
  1. Suggesting that the Ten Steps is where the issue lies is looking too far down the birth process. Mothers and babies have evolved to care for their babies safely in the way that the 10 steps recommends. Skin to skin, keeping mother and baby together, and breastfeeding are normal processes that have evolved to protect mother and baby. However modern medical care and a lack of postnatal care may impede this process.

The background to the Ten Steps

The Ten Steps is a Unicef Baby Friendly Initiative policy based on evidence supporting normal bonding processes between mother and baby in the early hours and days. It encourages hospitals to take steps that are known to promote successful breastfeeding such as skin-to-skin, keeping mother and baby together and not using pacifiers. The World Health Organisation have published an 111 page report on the evidence behind the Ten Steps, clearly showing its evidence base3.

Three of the Ten Steps are specifically critiqued in the JAMA article, namely Skin to skin contact, keeping mother and baby together after birth and not using pacifiers. These behaviors are perceived to increase risk for mother and baby and the authors argue that they should not be so widely promoted.

However, each of these steps has an evidence base behind why it promotes successful breastfeeding, and wider infant health and wellbeing. For example, babies who have skin to skin have better temperatures, steadier heart rates and breathing and better oxygen levels. They are more likely to breastfeed as are naturally close to the breast4. Babies who are kept close to their mothers are more likely to be breastfed, likely because they are fed more regularly and responsively. Separating mother and baby increases the risk of difficulties5. Pacifiers can increase the risk of breastfeeding difficulties and also other infections6. Therefore any critique of the Ten Steps must look at overall benefit and risk to mother and baby.

What issues do the authors raise?

The authors of the JAMA paper raise three main issues, notably skin to skin, keeping mother and baby together and not promoting pacifier use. They note that:

‘It is important to be certain that the basis for the recommendations has been documented in reproducible scientific studies and that the benefits of the practices recommended outweigh the risks. 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.’

  1. Sudden Unexpected Postnatal Collapse (SUPC)

The authors cite research around babies who collapse unexpectedly (e.g. not because of illness or prematurity) in the first few days of life, suggesting that skin-to-skin promotion by the Ten Steps increases this risk. There are a few issues to bear in mind here.

Firstly, thankfully SUPC is rare. Estimates from two studies in Germany, cited in the paper the authors refer to suggest a risk of around 3 in 100,000 babies. This is obviously clearly very traumatic for those who have a baby who collapses, but in terms of individual risk, the risk of not collapsing is 99.97%. Babies do not commonly collapse.

Secondly, the paper suggests that only one third of these cases happen in the first two hours of life (the key skin to skin time being around an hour after birth). The risk is therefore reduced to 99.99% of not collapsing.

Thirdly, the three cases referred to in the paper look at collapse that happens during skin to skin, but this is not skin to skin immediately post birth. One baby sadly collapsed at 100 minutes old, but the other two collapsed on the second day of life. They just happened to be having skin to skin with their mother7.

One study referred to in that paper suggests an association between skin to skin and collapse but no pathway to explain it8. However, its conclusions are not that skin to skin should be stopped, but more resources should be put in place to monitor babies:

As such contact has been proven to be beneficial and without apparent risks, this practice should be promoted. However, maternity staff should be vigilant during skin-to-skin contact, especially if the mother is alone with her neonate or other risk factors are present.’

Of course we want a world where we do not experience any SUPC. However, for the vast majority of babies, skin to skin will bring benefits without such a collapse. Can we really suggest that we do not encourage a natural process that has benefits for infant temperature, breathing and breastfeeding based on such a small risk? As the authors of the skin to skin study suggest, this is an issue with a lack of postnatal care, not skin to skin. 

  1. Risk of babies falling out of bed if kept with mother

The second issue raised is that encouraging mother and baby being kept together through rooming in may increase the risk of babies falling out of bed. As a mother who has given birth several times in the last ten years in the UK, rooming in was an entirely normal and expected practice. The idea of all babies going to the hospital nursery was abandoned some years before based on the benefits for infant wellbeing and breastfeeding seen when mothers stayed together9.

However in the USA, hospital nurseries are still current practice and there is a lot of debate over the move to encourage mother and baby staying together. Speaking to older colleagues and mothers, this debate also happened in the UK when practices were changed, as is the case when anything is changed. However, now it is an accepted practice. Fears in the USA are based around based mainly around mothers being exhausted after birth and babies being accidentally dropped.

The authors of the JAMA paper refer to a USA research paper which estimated the risk of babies falling out of bed in hospital at around 1.6 – 4.1 in every 10,000 births – again a very low individual risk10. This is obviously traumatic if it happens to your baby but should we be stopping the 99%+ of babies who do not fall out of bed being kept together with their mother as nature intended based on this risk? Again, this falls not to an issue with mother and baby being together, but a lack of postnatal care that prevents mothers from being so exhausted their baby falls. Indeed, the authors of the Canadian study conclude not that mother and baby should be separated but there should be interventions put in place including:

‘Monitoring mothers more closely, improving equipment safety, and spreading information about newborn falls within the state and throughout the hospital system’.

In other words not separating mother and baby. I searched for similar statistics in the UK but did not find them (although apparently 20,000 adults in the UK attend A&E each year after falling out of bed). However, one key study in the UK that looked at baby sleeping position in hospitals comparing babies being in bed, side car cribs and stand alone cots found that no baby experienced an ‘adverse event’. Importantly, there were also no differences in mother or infant sleep11 which challenges the notion that separating mother and baby will promote rest.

3. Not encouraging pacifier use

Thirdly, the authors critique a lack of use of pacifiers. Research has shown that giving a baby a pacifier decreases the risk of sudden infant death syndrome (SIDS). However it is not clear why. One reason proposed is that sucking on a pacifier, or being woken up by it if the baby is on its tummy, stops the baby from sleeping too deeply which might be a SIDS risk. However, pacifier use can increase the risk of breastfeeding difficulties in the first few weeks, which is why it is recommended in the Ten Steps not to give a dummy until around six weeks. Incidentally, breastfeeding also decreases SIDS risk, potentially through reduced infection, but likely because babies who breastfeed often fall asleep at the breast and wake more frequently, meaning they do not sleep so deeply. Research has not distinguished between breast and formula fed babies who do not use pacifiers and SIDS risk.

However what risk is a baby really at who doesn’t use a pacifier? The risk of a baby dying from SIDS is around 300 babies a year in the UK out of the 700,000 born, so an individual risk of around 0.04%12. The JAMA article states that around half of babies die in the first six weeks before a pacifier is recommended, so this risk is again halved. Again, clearly beyond tragic for the families involved but with a low individual risk.

It is also important to note that there are also a number of other factors associated with preventing SIDS e.g. putting the baby to sleep on their back and not smoking; a lack of pacifier does not explain all SIDS deaths. Giving all babies pacifiers from birth would sadly not prevent all SIDS but it would increase the risk of breastfeeding difficulties and certain infections. Using them is not risk free.

What is the real issue here?

Firstly, if you look at any behavior closely enough you will find a risk. Risk is part of our everyday lives. Getting in the car, walking down the street or even walking down the stairs. Indeed, over 30,000 adults die from falls every year in the USA. You can look at statistics and argue risk for home births, whilst someone else can show risks from giving birth in a hospital. It’s all about balancing individual risk against potential individual benefits and for the vast majority of babies, skin to skin, not using pacifiers and keeping mother and baby together will have the best outcomes.

The authors of JAMA raise statistical trends that are rare, but I argue that lessening these risks should not fall at the stage of the Ten Steps, but much before this. We need to do three things: enable normal birthing practices where possible, provide far more postnatal care and support new mothers better in general. Many new mothers might feel too exhausted and unsupported to care for their babies according to the Ten Steps, but that is not an issue with the Ten Steps, it is an issue with our society.

Birthing practices can make caring for a newborn more difficult and in the USA the risk of intervention during birth is high, increasing the chances of mothers feeling exhausted, medicated and unable to care for their baby immediately after the birth. The overall rate of caesarean section is 32.2% in the USA13 compared to 25% in the UK14. The epidural rate in the USA is 61%15, whilst the UK has a rate half of that 16.

Caesarean sections and epidurals increase the risk of the mother being immobilized and for longer postnatally. These interventions (albeit life saving for some), can also affect normal hormone patterns after birth, increasing stress hormones and reducing oxytocin levels17. This can make it more difficult for a mother to have her baby close to her. However we should not consider mother and baby being close together as the issue that needs solving rather the high level of birth interventions creating this situation in the first place. Promoting normal birth where possible would allow more mothers to feel able to care for their baby immediately after birth.

Secondly, you would have to be living under a rock to be unaware of staffing crises in hospitals. Midwives and doctors are rushed off their feet and having little time to sit and care for new mothers. Each of the articles used by the JAMA authors to critique the Ten Steps raises the issue of poor postnatal care as the key focal issue. Again, skin-to-skin and keeping mother and baby together are not the issue – a lack of support and care for new mothers during the postnatal period is. We need to invest in postnatal care if we genuinely want to improve the health and wellbeing of our mothers and babies.

Finally, if our new mothers are too exhausted to spend time having skin to skin, getting breastfeeding off to the best start and keeping their babies close, the answer is not to separate them to solve the issue, but to support them to care for and bond with their babies. As a society we need to start to ‘mother our mothers’ better so that they can care for their babies without these statistically rare perceived risks.

You can download a PDF on this article here We need better birth and postnatal care for our mothers not separation in the name of risk


  1. http://www.unicef.org/newsline/tenstps.htm
  1. Bass, J., Gartley, Y., & Kleinman, R. Unintended Consequences of Current Breastfeeding Initiatives. JAMA, doi:1001/jamapediatrics.2016.1529
  1. http://www.who.int/nutrition/publications/evidence_ten_step_eng.pdf
  1. Charpak, N., Gabriel Ruiz, J., Zupan, J., Cattaneo, A., Figueroa, Z., Tessier, R., … & Mokhachane, M. (2005). Kangaroo mother care: 25 years after. Acta Paediatrica, 94(5), 514-522.
  1. Bystrova, K., Ivanova, V., Edhborg, M., Matthiesen, A. S., Ransjö‐Arvidson, A. B., Mukhamedrakhimov, R., … & Widström, A. M. (2009). Early contact versus separation: effects on mother–infant interaction one year later. Birth, 36(2), 97-109.
  1. https://www.isisonline.org.uk/sleep_health/sleep_aids/dummies/
  1. Herlenius, E., & 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.
  1. Gómez, J. R. A., Elorriaga, I. A., Fernández-Llebrez, L., Fernández, A. P., Avellanal, C. U., & Sierra, C. O. (2011). Episodios aparentemente letales en las primeras dos horas de vida durante el contacto piel con piel. Incidencia y factores de riesgo. Progresos de obstetricia y ginecología, 54(2), 55-59.
  1. Crenshaw, J. (2007). Care practice# 6: no separation of mother and baby, with unlimited opportunities for breastfeeding. The Journal of perinatal education, 16(3), 39-43.
  1. Helsley, L., McDonald, J. V., & Stewart, V. T. (2010). Addressing In-Hospital. The Joint Commission Journal on Quality and Patient Safety, 36(7), 327-333.
  1. Ball, H. L., Ward-Platt, M. P., Heslop, E., Leech, S. J., & Brown, K. A. (2006). Randomised trial of infant sleep location on the postnatal ward. Archives of Disease in Childhood, 91(12), 1005-1010.
  1. http://www.lullabytrust.org.uk/document.doc?id=297
  1. http://www.cdc.gov/nchs/nvss/births.htm
  1. https://www.nct.org.uk/professional/research/maternity%20statistics/maternity-statistics-england
  1. https://www.cdc.gov/nchs/data/nvsr/nvsr59/nvsr59_05.pdf
  1. https://www.rcm.org.uk/learning-and-career/learning-and-research/ebm-articles/pain-and-epidural-use-in-normal-childbirth
  1. Brown, A., & Jordan, S. (2013). Impact of birth complications on breastfeeding duration: an internet survey. Journal of advanced nursing, 69(4), 828-839.