Evidence for the impact of BFHI in the UK: an interview with lead researcher, Dr Vicky Fallon

The very start of 2019 saw the launch of the NHS Long Term Plan. Most importantly for us and our readers, the ‘Maternity and Neonatal Services‘ section of the report states its purpose to ‘accelerate action to achieve 50% reductions in stillbirth, maternal mortality, neonatal mortality and serious brain injury by 2025’. One of the ways in which they propose to do this is to ensure that maternity services become compliant with UNICEF’s Baby Friendly Initiative;

All maternity services that do not deliver an accredited, evidence-based infant feeding programme, such as the UNICEF Baby Friendly Initiative, will begin the accreditation process in 2019/20.

The aims of the long term plan are laudable and, as advocates for science-based medicine, we would hope that the recommendations provided to achieve this goal have a strong evidence base to support their implementation. Given this, we were pretty surprised to read the findings from Dr Vicky Fallon’s systematic review of the evidence for the impact of the BFHI in a UK setting.

Worryingly, while there did appear to be a positive impact on breastfeeding rates up to the first week postpartum, this is not sustained. There was evidence of a positive impact of some individual steps of the Baby Friendly Community Initiative on breastfeeding rates at 6-8 weeks postpartum. However, unrealistic expectations of breastfeeding may be leaving families with negative emotional experiences. Most shocking is that there does not appear to be any data on the wider maternal and infant health outcomes in a BFHI setting, which leads us to question how its implementation could further the aims of the Long Term Plan.

In light of these shaky results, we were delighted when the lead researcher the study, Dr Vicky Fallon, agreed to answer some of our most pressing questions on the inspiration behind the paper, the methodology and crucially, the direction in which this research may now lead.

What inspired you to undertake this research?

There were two main inspirations for the current research. The initial idea for the research came from a wider systematic review of the Baby Friendly Hospital Initiative (BFHI) conducted by Rafael Perez-Escamilla and colleagues in 2016. They examined the impact of BFI on breastfeeding and child health outcomes, both globally and specific to the US. Global evidence demonstrated that BFI steps had a positive impact on multiple child health outcomes, including breastfeeding initiation and duration up to one year postpartum. However, the US BFI evidence only demonstrated improvements in breastfeeding up to six weeks, and these were not sustained. We were interested in finding out whether the US findings extended to other countries with resource rich environments like the UK. If this was the case, this would indicate that we may need to adapt the BFI model of care to better support breastfeeding and child health outcomes in these contexts.

The second reason we wanted to conduct the review was based on a body of previous work that we (and others) have conducted in resource rich environments examining the experiences of mothers using different feeding methods. A number of these studies have found that the current approach to infant feeding promotion and support may be related to negative emotional and practical experiences in new mothers. However, they do not specify whether the populations were drawn from BFI-accredited settings. When assessing any health care intervention, it is important to not only know the effects of the intervention (in this case, whether it improves breastfeeding and subsequent health outcomes), but also how patients experience the intervention (in this case, how mothers feel about BFI care). We decided to examine all of the literature exploring women’s experience of BFI-compliant care in the UK to ensure we were getting a comprehensive picture of the intervention.

Could you explain a little bit for non-scientists, how you select the studies and other key points to consider when doing a systematic review?

Sure, we firstly start with some aims. Our aims were to a) examine the impact of BFI
implementation (hospital and community) on maternal and infant health outcomes in the UK, and b) explore the experiences and views of women receiving BFI-compliant care in the UK. We then develop pre-defined criteria to help us decide whether a study should be included in the review. We kept this really broad as we wanted to capture as much information relating to our aims as possible.

We included published studies using any methodology as long as they collected data relating to the implementation of BFI and any maternal and infant physical and mental health outcomes. Studies which explored the experiences and views of women receiving BFI compliant care were also eligible. All studies which included women who were pregnant or had children under the age of two conducted in a UK setting were considered too.

We then develop a comprehensive list of search terms based on all of these criteria and systematically apply them to hundreds of pre-defined databases to pull up as many studies as possible that are relevant. We read all of the studies and assess whether they meet our criteria at three different stages (titles, abstract, then full text). Then, we extract the data that we need from the studies according to our aims and examine the quality of each piece of research. This is done by more than one person to ensure that it is objective.

Finally, we synthesise all of the data together to give a comprehensive overview of all of the available literature relating to our aims. Systematic reviews are really useful because they provide us with much more information than one study alone which allows us to draw more meaningful conclusions. Also, because the steps involved in them are so “systematic” and we objectively examine study quality, it means that any risk of bias in the findings is minimised.

What surprised you about this research?

I was really surprised at how little evidence there was examining UK-BFI. The programme
was introduced to the UK almost 30 years ago, it is a nationally recognised mark of quality of care, and the NHS long term plan aims to roll this out in every maternity unit by 2019/2020 to support breastfeeding and ultimately enable better health outcomes for mothers and infants. Yet there is currently no UK data available relating to wider physical maternal or infant health outcomes. There is also no UK evidence examining the effects of the intervention on breastfeeding outcomes up to six months despite the initiative being underpinned by the WHO/UNICEF recommendation of exclusive breastfeeding to six months. The evidence relating to shorter-term breastfeeding outcomes is limited and varies a lot in terms of quality too.

It is not enough to apply global evidence for BFI to a specific country and assume that it works there because it works elsewhere in the world. Barriers and determinants of breastfeeding behaviour differ according to income setting which shows that we need tailored breastfeeding promotion and support. Furthermore, breastfeeding rates differ by country income which indicates that we need to tailor breastfeeding initiatives to specific patterns recorded in each country.

What didn’t surprise you?

The evidence that explored women’s experience of BFI-compliant care was not a massive surprise. We found support is highly influential to women’s experiences of BFI-compliant care, but current provision may promote unrealistic expectations of breastfeeding, not meet women’s individual needs, and foster negative emotional experiences. We also found that women want regular, personal, and practical infant feeding support. This resonates with work that we have done previously in the UK and with other research in resource rich environments. This is the first time that the evidence base has been explored specifically in BFI accredited settings and collectively the findings indicate that the current approach to BFI needs to be adapted in these settings.

Where has this paper taken you in terms of new projects?

There were a number of limitations in the existing evidence base that we would like to overcome in future work in order to provide a comprehensive picture of the impact of BFI implementation in the UK. First, we would like to examine the impact of BFI on wider maternal and infant health outcomes as this has not yet been done. Second, we would like to examine the impact of BFI on breastfeeding outcomes up to six months of age in line with current feeding recommendations. Third, we don’t know which components of the model work well and which may need to be adapted so examining the effects of individual steps of BFI in future studies is necessary. Finally, it is currently unclear whether the qualitative findings are purely specific to BFI practices or UK maternity care in general so it is essential that we conduct research directly comparing BFI vs non-BFI care in relation to maternal and infant health outcomes.

Team feed is super thankful to Dr Fallon for taking the time to answer our questions; it’s great to hear the most up to date research directly from the scientists at the coal face and we appreciate the time she took to share her thoughts with us. It’s clear the evidence base for BFI in the UK is severely lacking and this needs urgently addressed, particularly in light of the plans to rollout out across the whole of the UK.

Here at Feed we feel that prior to full implementation of BFI across maternity units in the UK (at substantial cost to the UK taxpayer), the initiate needs to undergo rigorous analysis to ensure that it delivers on the quality, science based, compassionate infant feeding education and support that all families in the UK deserve.  Join us in calling on your MP to push for a review of the evidence underpinning the BFI because policy and healthcare commitments that touch our lives need to be proven to make a positive change. If we’re serious about investing in families and babies, let’s prove it by making sure whatever we do works. It’s a fundamental.