Improving breastfeeding support through the implementation of the baby friendly hospital and community initiatives: a scoping review protocol

Background: Improved breastfeeding practices have the potential to save the lives of over 823,000 children under 5 years old globally every year. Exclusively breastfeeding infants for the first six months would lead to the largest infant mortality reduction. The Baby-Friendly Hospital Initiative (BFHI) is a global campaign by the World Health Organization and the United Nations Children's Fund (UNICEF), which promotes best practice to support breastfeeding in maternity services. The Baby-Friendly Community Initiative (BFCI) is an extension of the BHFI’s 10 th step of the Ten Steps to Successful Breastfeeding and of the BFHI overall. Its focus is on community-based breastfeeding supports for women. There have been no known attempts to synthesise the overall body of evidence on the BFHI in recent years, and no synthesis of empirical research on the BFCI. This scoping review asks the question: what is known about the implementation of the BFHI and the BFCI globally? Methods and analysis: This scoping review will be conducted according to the Joanna Briggs Institute methodology for scoping reviews. Inclusion criteria will follow the Population, Concepts, Contexts approach. A data charting form will be developed and applied to all the included articles. Qualitative and quantitative descriptive analysis will be undertaken. The PAGER (Patterns, Advances, Gaps, Evidence for practice and Research recommendations) methodological framework will be used to analyse and report review findings. Conclusion: This review will establish gaps in current evidence which will inform areas for future research in relation to this global initiative.


Amendments from Version 1
-We have decided to chart the data according to the PAGER (Patterns, Advances, Gaps, Evidence for practice and Research recommendations) framework (Bradbury-Jones et al., 2021). This framework is a methodological framework which will enable us to analyse and report review findings.

Introduction
Globally, improved breastfeeding practices have the potential to save the lives of over 823,000 children under 5 years old every year, and save $300US billion 1 . Exclusively breastfeeding infants for the first six months of their life is known to be the best start for a baby and a more widespread adoption of exclusive breastfeeding would lead to the largest infant mortality reduction 1 . It can contribute towards meeting Sustainable Development Goals (SDG) 2 and 3 -targets on nutrition and health -as well as being linked to many other SDGs. Since 1990, the World Health Organization (WHO) recommends that all newborn babies are exclusively breastfed for the first six months of their lives, and continue to be breastfed for up to two years. Currently, just 42.2% of infants under 6 months are being exclusively breastfed and just 33 countries are on target for exclusive breastfeeding 2 . Breastfeeding rates are both supported and hindered by the social determinants of health and multi-level support is needed, including policy, health systems and services, communities and families 3 .
The Baby-Friendly Hospital Initiative (BFHI), launched by WHO and United Nations Children's Fund (UNICEF) in 1991, has been implemented globally in over 150 countries and is a pillar of the WHO/UNICEF Global Strategy for Infant and Young Child Feeding 4 . One of the nine operational targets of the Global Strategy for Infant and Young Child Feeding is to ensure that every maternity facility practices the Ten Steps to Successful Breastfeeding. Hospitals or maternity facilities that comply with the Ten Steps to Successful Breastfeeding and with the International Code of Marketing of Breastmilk Substitutes and subsequent relevant World Health Assembly resolutions (the Code) are designated as "Baby-friendly". Table 1 details the Ten Steps, which were updated and revised in 2018 5 . Although the BFHI has been widely implemented, coverage at a global level remains low. In 2017, just 10% of infants in the world were born in a facility currently designated as "Baby-friendly" 5 . In 2018, the BFHI was revised, which led to greater emphasis on scaling up to universal coverage, ensuring sustainability, and integrating the programme more fully with health-care systems 5 .
The Baby-Friendly Community Initiative (BFCI) is an extension of the BHFI's 10 th step of the Ten Steps to Successful Breastfeeding and of the BFHI overall 6 . Its focus is on communitybased breastfeeding supports for women. Given the usual short post-partum stay in facilities, this 10 th step and associated separate initiatives are often critical to support breastfeeding mothers beyond the initial days of giving birth. The 10 th BFHI step changed from "foster the establishment of support groups Facilitate immediate and uninterrupted skin-to-skin contact and support mothers to initiate breastfeeding as soon as possible after birth 5 Support mothers to initiate and maintain breastfeeding and manage common difficulties.

6
Do not provide breastfed newborns any food or fluids other than breast milk, unless medically indicated 7 Enable mothers and their infants to remain together and to practice rooming-in 24 hours a day 8 Support mothers to recognise and respond to their infants' cues for feeding

9
Counsel mothers on the use and risks of feeding bottles, teats and pacifiers 10 Coordinate discharge so that parents and their infants have timely access to ongoing support and care and refer mothers to them on discharge from hospital" in the 1989 version to "coordinate discharge so that parents and their infants have timely access to ongoing support and care" in the revised version in 2018 7 . It has been suggested that this change in step 10 signals a shift in increased responsibility of facilities in planning and facilitating community supports for mothers 7 . While the BFHI was adopted in 152 countries, it appears that the BFCI has been adopted in a smaller number of countries, namely low-and middle-income countries (LMICs), including Kenya, Cambodia, Gambia 6 and High Income Countries (HICs) such as Italy 8 and the UK 9 .
Between 2012 and 2017, almost 80% of live births occurred with the assistance of skilled health personnel globally 10 . However, the estimated coverage of births attended by skilled health personnel during this period shows significant inequality between WHO regions. Just 59% of the births in the sub-Saharan Africa Region (during the period 2012-2017) were attended by skilled health personal, where maternal mortality is highest 10 . In other WHO regions, between 68% to 99% of all births were attended by skilled health personnel 10 . This underlines the importance of supports at the community level for breastfeeding, and it points towards the need for international interventions that promote breastfeeding to be mindful of the need to improve equity of access to breastfeeding supports.
Study rationale, aims and objectives Scoping reviews are useful when a body of literature has not yet been comprehensively reviewed, or 'exhibits a large, complex or heterogeneous nature, not amenable to a more precise systematic review' 11 . Scoping reviews map the range of evidence, and also identify gaps in the knowledge base, clarify concepts, and document research that inform and address practice 12 . A pilot search as part of the initial stage of this review (see Extended data) found that the majority of articles in this area have been published since 2012. There have been no known attempts to synthesise the overall body of evidence on the BFHI in recent years, and no synthesis of empirical research on the BFCI.

Methods and analysis
This scoping review will be conducted according to the Joanna Briggs Institute (JBI) methodology for scoping reviews 13 . We will use the framework for scoping reviews developed by Arksey & O'Malley 14 as the foundation, updated and advanced by Levac et al. 15 and progressed further by new guidance from the JBI 11,13,16 .
According to this framework, there are six different stages, including: 1) identifying the research question; 2) identifying relevant articles; 3) study selection; 4) charting the data; 5) collating, summarising and reporting results; and 6) consulting with stakeholders. The scoping review will also adhere to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) to ensure rigour in reporting. The review is registered with the Open Science Framework, DOI: https://doi.org/10.17605/OSF. IO/27R3M.

Stage 1: identifying the research question
The research aims and objectives for this scoping review were developed iteratively through discussions between the research team and were informed by the pilot search of the literature. The proposed review is situated within a wider research project, which is evaluating the implementation of evidence-based policy on infant feeding in Malawi 17 , focused on exclusive breastfeeding for the first six months. In Malawi, the BFHI has been a well-known vehicle for the improvement of exclusive breastfeeding promotion in hospitals and health facilities since 1993, and as recent as 2018, several externally funded initiatives have been implemented to revive the BFHI, provide training for healthcare staff and accreditation for baby-friendly hospitals 6 .  16 will be followed.
Step one is a limited search for peer-reviewed, published papers on the PubMed and CINAHL databases (see Extended data 28 ), which has already been performed. An academic research librarian was consulted and an analysis of the words contained in the titles, abstracts and index terms generated the list of keywords detailed in Extended data. Search terms will be piloted to assess the appropriateness of databases and keywords. The second step will be conducted with the librarian which may involve refining the search terms. The third step will be examining the references of key articles that have been identified for full text review that meet the inclusion criteria. The following databases have been selected in consultation with an academic librarian: Pubmed, Embase, Web of Science, Global Health and CINAHL. Key words and index terms from the title and abstract of key articles were noted and used to inform the search strategy. The timeframe for the search will be from when the first article was published in a given database to October 2020.
Inclusion and exclusion criteria. Inclusion criteria will be guided by the Population, Concepts, Contexts approach (Table 2) 15 .

Criteria Determinants
Population Women are pregnant, postnatal period and up to 2 years post-partum

Concepts Baby-Friendly Hospital Initiative or the Baby-Friendly Community Initiative
Context Hospital or community. No country or geographic location will be excluded.
All research designs will be included: qualitative, quantitative and mixed method studies. Quantitative studies will include both experimental (e.g., randomised trials, non-randomised trials) and observational (e.g., cohort, cross-sectional) study designs. Qualitative studies will include designs such as grounded theory, ethnography, phenomenology, action research and qualitative descriptive design. In addition, all types of reviews of empirical research will be included. Grey literature will not be considered for inclusion in the review.
Draft inclusion and exclusion criteria will be tested on a sample of 15 articles to check the criteria's suitability and will be amended as necessary.

Stage 3: study selection
The screening process will consist of two phases: i) a title and abstract screening; ii) full-text screening. In stage i) all titles and abstracts will be screened by two reviewers, one reviewer will review 100% of articles, and two other reviewers will review 50% each. Screening will be undertaken through Covidence and duplicates will be removed. Where there is disagreement between reviewers as to whether an article should be included or excluded, a third reviewer will arbitrate. At full text screening stage, one reviewer will undertake a full text screening for eligibility and 100% of articles will be cross-checked by another reviewer.
Stage 4: charting the data Data will be extracted according to the JBI framework 13 . A data charting form will be developed and applied to all the included articles. Examples of information to be included in the data charting form is included in Extended data 28 . Two reviewers will independently pilot the form on a random sample of approximately five included articles. Data will be coded and entered in Microsoft Excel. In keeping with scoping review methodology, an assessment of the quality of individual articles will not be undertaken. As the quality of the evidence will not be assessed, there are limits to the extent to which conclusions can be drawn about the evidence gap.
Stage 5: collating, summarising and reporting results A 'descriptive-analytical' method will be used. As this is a scoping review, it is not anticipated that aggregation and synthesis of individual research results will be undertaken. The PAGER (Patterns, Advances, Gaps, Evidence for practice and Research recommendations) methodological framework 29 will be used to analyse and report review findings. This will firstly consist of a descriptive summary of the included findings. A patterning chart will be developed, which will then allow for identification of advances in the field of research on the BFHI/ BFCI. Gaps will then be identified from this analytical process. A framework for reporting evidence for practice will include consideration of the following stakeholders: policy makers; research commissioners and practitioners/service providers. The research recommendations will build on the identification of gaps. These final two stages of the framework will be conducted in consultation with stakeholders at Stage 6 of the review. Key reflective questions at each stage of the framework will guide the analysis 29 .
Stage 6: consulting with stakeholders Findings from the review will be prepared for stakeholders who have expertise in relation to the BFHI and the BFCI. These will include researchers, practitioners and policy makers at the global level and at WHO regional levels.

Research checklist
This scoping review protocol was drafted using the PRISMA-SCR extension checklist 28,30 .

Study status
This study has completed the scoping stage, including piloting and refining search terms. Full database searches are ready to run and de-duplication and screening will begin in December 2020.

Conclusion
The aim of this scoping review is to map and examine the evidence relating to the implementation of Baby-Friendly Hospital and Community Initiatives globally. Results will be published in a peer-reviewed journal and disseminated through conferences and/or seminars. This review will establish gaps in current evidence which will inform areas for future research in relation to this global initiative.

Data availability Underlying data
No data is associated with this article. This project contains the following extended data: -Search strategy pilot.
Data are available under the terms of the Creative Commons Zero "No rights reserved" data waiver (CC0 1.0 Public domain dedication). The proposal is relevant and well outlined. The authors should specify how the disagreements will be settled during the time of the review. This needs to be specified ahead of time.

Open Peer Review
There seems to be a leaning towards Malawi as such the authors have to be specific if this review is Malawi specific or will include other countries as well. Either way, the direction as far as the setting is concerned and boundaries for the papers that will be included should be explicit. If it covers research beyond Malawi, then other works from other countries should be cited as well and Malawi should not stand out.

Are sufficient details of the methods provided to allow replication by others? Yes
Are the datasets clearly presented in a useable and accessible format? Partly School of Nursing and Midwifery, National University of Ireland, Cork, Ireland Thank you for the opportunity to review this scoping review protocol, which seems particularly timely as the authors have clearly identified the dearth of synthesised evidence relating to the BFHI and particularly the BFCI. This scoping review using the framework by Arskey and O'Malley (2005) is clearly described and the rationale for choice of review is defended. The Scoping Review protocol as presented is clearly articulated outlining each step of the systematic review process. As reviewers we have the following suggestions which we propose would enhance the review protocol: The inclusion of the reference to Malawi in the abstract is a distraction and recommend its removal. 1.
The conclusion to the scoping review needs to be mirrored in the abstract as there seems to be a disconnect between the global perspective and the research plans articulated for Malawi.

2.
The last statement of the Study rationale section refers to empirical research, yet the inclusion criteria includes reviews. This needs to be revised.

3.
There is no need to use the abbreviations in the 'Identifying the research question' section as this had already been previously addressed.

4.
The iterative process for this scoping review is to be commended. However, as per point 2 above, the global perspective needs to be better reflected in stage 1.

5.
In Stage 2, it is to be commended that a librarian would be engaged in the search strategy as this will ensure that all relevant databases and appropriate search terms will be utilised.

6.
The authors refer to the inclusion of 'all types of review', however, we believe that inclusion of reviews should be confined to systematic reviews, metasynthesis or any that have been underpinned by a defined framework such as PRISMA.

7.
In relation to point iii) of the inclusion criteria, could greater clarity be provided by referring to the support from the 10 steps to interrogate the evidence as support in general is such a broad concept.

8.
In relation to point vii) empirical studies, it would be best to revise if the authors are including reviews.

9.
In terms of Exclusion criteria no iii), we suggest that conference proceeding also be included here.

10.
In Stage 3, we suggest us of the term papers be employed rather than studies to reflect that more that one paper may be published from a particularly study.

11.
In Stage 4, it is important to note that identifying gaps in the literature through a scoping study will not necessarily identify research gaps where the research itself is of poor quality since quality assessment does not form part of the scoping study remit. Therefore we 12.
suggest this limitation is acknowledged.
On the data extraction plan, the final point (no 11) states 'key findings' which seems non specific. The inclusion of sub divisions would aid clarity.

13.
In Stage 5, use of the Levsque framework is not clearly articulated, for example, is it the intention to extract data according to the framework? 14.
In Stage 6, following on from the above point, reference to the '…application of the accessibility framework...' is not clear.

15.
In Stage 6, there is reference to realist research which is not sufficiently contextualised or integrated with the scoping review protocol and maybe consider removing.

16.
Thanking you and wishing you every success with this work Patricia and Helen Is the rationale for, and objectives of, the study clearly described? Yes

Are sufficient details of the methods provided to allow replication by others? Partly
Are the datasets clearly presented in a useable and accessible format? Not applicable be included here.
Conference proceedings, trial registrations, study protocols and letters have now been included as exclusion criteria.