ABSTRACT
Background Evidence on the effectiveness of kangaroo mother care (KMC) is available and guidelines have been formulated. However, little is known about the programmatic implementation of KMC at national and subnational levels.
Methods A structured search of grey literature was conducted to identify reports of national or subnational level implementation of KMC to understand the population-based coverage of KMC, availability of KMC services at national and subnational levels, programmatic approaches to scale, and health systems actions that may influence KMC scale-up. The search strategy included two rounds of screening using a variety of grey literature search engines, portals, repositories, and targeted websites, as well as snowball sampling. Data from 212 documents were extracted and transferred into a database with an extensive topic list. These documents were then classified as “for in-depth analysis”, “possible nuggets”, and “not important”. Further analysis of 42 information-rich documents was conducted with NVivo software to identify recurring themes.
Findings We found information on KMC implementation on a national or subnational scale for 18 countries. Estimates for national- or subnational-level population-based coverage of KMC were available from only six countries (Colombia 63%, the Philippines 53%, Malawi 22%, Bangladesh 22%, India 13%, Viet Nam 8%). Information on the availability of KMC services was scant and fragmented, with no information on their quality. Programmatic KMC implementation is characterised by leadership from a variety of implementation partners and by different implementation trajectories. Countries for which information on KMC implementation was available shared common health system actions such as the inclusion of KMC in national policy, recognition of KMC as a priority newborn health intervention and strong advocacy by champions at all levels, dedicated investment in KMC and in some cases insurance cover for KMC, capacity building and motivation among health workers, dedicated space for KMC with equipment and supplies, support for KMC practice, and data collection and use.
Conclusion Programmatic implementation of KMC requires action in multiple health system building blocks with a focus on monitoring and evaluation of availability and quality of services, along with coverage.
Kangaroo Mother Care (KMC) is a method of care for preterm and low birthweight (LBW) infants that reduces neonatal morbidity and mortality [1]. The World Health Organization (WHO) defines KMC as continuous and prolonged skin-to-skin contact of the baby with a caregiver and support for exclusive breastmilk feeding [2]. Two other features that are widely considered to be part of the KMC package are early discharge from hospital with a reliable follow-up system in place and support by healthcare workers, the family and the community [1, 3–5].
KMC is a complex intervention [6] that depends on strengthening of health systems and on human behaviour [5, 7]. In addition, facility-based KMC forms an integral part of the care packages for small and sick newborns [8]. Where KMC was originally largely seen as an add-on intervention that was introduced vertically in a small number of higher-level health facilities [9], it gradually became more embedded in the wider focus of strengthening the care of small and sick newborns and promoting family-participatory care, especially at the level of district hospitals and other lower-level health facilities [8, 10–13].
Although there are many claims and reports about the successful implementation of facility-based KMC, little is known about the programmatic implementation of KMC and the availability of KMC services globally or at national or subnational levels. Data in the published domain provide insufficient information on population-based coverage of KMC. As KMC is not yet being captured in the health management information system (HMIS) of most countries, some form of review of the grey literature could help identify countries with some footprint of programmatic KMC implementation and reveal how health system and process factors have affected their implementation and scale-up activities. The objective of this review was to understand and synthesise information on population-based coverage of KMC, the availability of KMC services at national and subnational levels, programmatic approaches to scale, and health system components and process factors that may influence scale-up of KMC.
METHODS
The initial search question was framed very broadly: How is programmatic implementation of KMC reported in the grey literature? The structured search also endeavoured to identify packages of health interventions that had achieved high coverage of population-based KMC (>50%) as well as details of the components of such packages.
Coverage is defined as the proportion of people (preterm or LBW babies) who received a specific intervention or service (KMC) in a particular geographic or administrative region [14]. Implementation is interpreted as the extent to which a programme is delivered as intended – there are individual-level (mother–infant dyad) and programme-level (KMC services) measures of implementation [15]. Scale-up is defined as the programmatic expansion of KMC services to bring the benefits of KMC to more preterm and LBW babies over a wider geographical area more quickly and more equitably, over longer periods [16, 17].
Eligibility criteria
Finding documents in the grey literature on the programmatic implementation of KMC involved broaching uncharted territory. The initial search net was cast very wide before being narrowed down. Grey literature was interpreted as including any of the following types of documents: documents related to conferences and meetings (e.g. abstracts, presentations, posters, summaries, working group reports); government documents (e.g. policy statements, guidelines, strategic plans); project reports, reviews and briefs; assessments and evaluations; newsletters; blogs; miscellaneous slide presentations; and unpublished dissertations or theses.
The setting of inclusion and exclusion criteria was guided by what could realistically be managed in one review. Only printable electronic documents and documents with “kangaroo care” or “kangaroo mother care” in the title or abstract were included (French: “mère kangourou” or “maman kangourou”; Spanish: “madre canguro”; Portuguese: “mãe canguru”). Video and audio recordings and terms related to the individual components of KMC (skin-to-skin, breastfeeding, support, discharge and follow-up) and the care of small and sick newborns were excluded.
As the purpose of the review was not to find or grade evidence on the effectiveness of interventions to implement or scale up KMC programmes, criteria such as risk of bias and certainty assessments were not applicable.
Information sources
Grey literature was found in two ways: through internet searches and through snowball sampling by contacting individuals working in the field. Figure 1 provides an overview of the screening and data analysis processes that took place between June and December 2021.
Flow diagram of the search and analysis strategies.
Search strategy
The search strategy consisted of two phases of screening. Initially, a very broad exploratory search was conducted on the internet using search engines, portals, repositories and websites (henceforth called “databases”) until no new links to databases were found. The primary screening route is described in supplemental file 1. The databases included OpenDOAR and university websites with extensive links to other grey literature databases (Illinois Chicago Public Health; Monash; La Trobe). In addition to these searches, additional sources were specifically targeted for searching, such as the Kangaroo Care Bibliography which is regularly updated by Susan Ludington and the following websites: Fundación Canguru in Colombia which manages the documents of the International Network of Kangaroo Mother Care (INK); UNICEF; development agencies known for their involvement with kangaroo mother care implementation (e.g. Save the Children, USAID, jhpiego, JSI Research & Training Institute, Inc.); and specific programmes ( e.g. Saving Newborn Lives, Every Preemie-Scale, ACCESS, MCHIP, MCSP). Search histories with potentially relevant titles were saved as pdf files. The internet search was conducted on 101 databases (supplemental file 2) and 701 documents were identified for a second round of screening. The number of hits per database ranged from 0 to 8185.
During the first screening, another 52 documents that are not available in the open domain on the internet were added: 42 documents received from individuals and 10 documents the reviewer had at her disposal.
Of the 753 documents identified for a second screening, 499 documents were retrieved from five of the specifically targeted websites, with 323 from the INK website. When the 753 documents’ tables of contents and their “Find” functions were used, a second screening entailed a manual search for terms related to “coverage”, “implementation” and “scale-up”, which yielded 212 documents. These documents reported on research or project outcomes, other kinds of assessments or any means of describing implementation processes or experiences.
Data collection, management and extraction
For easier management, the 212 remaining documents were grouped as follows: (AB) Conference abstracts, posters, summaries and working groups (n=49); (AS) Assessments, evaluations and reviews (n=31); (R) Reports, briefs, newsletters and blogs (n=53); (SP) Slides and conference presentations (n=60); and (SC) KMC implementation and scale-up documents (n=19). A database was created in MS Excel to summarise the relevant information gleaned from the documents and identify the documents likely to be of greater interest. Supplemental file 3 contains a summary of the process followed to complete each column in the database. Regarding KMC coverage and components of KMC implementation, a classification of A-E was included to get a sense of the distribution of these important concepts in the documents.
The reviewed documents were marked in the database as follows: (1) red for further in-depth analysis (n=118); (2) yellow for possible “nuggets” to consider (n=58); and (3) white for nothing of importance (n=36). The documents marked with red and yellow were further reviewed to identify documents that were sufficiently information-rich to be included in an inductive qualitative analysis using NVivo9 software [18]. Thirty-seven (37) grey documents were uploaded onto NVivo, together with four published articles and one article supplement that had been brought to our attention during the review (supplemental file 4). We identified 70 categories (nodes) in this analysis (see supplemental file 5) and they were used to identify emerging factors that played a role in country implementation of KMC. Supplemental file 6 contains a further breakdown of the major document types identified in the analysis.
Synthesis of findings
The approach to the synthesis included extracting numbers, but also an analysis along the lines described by Spicer et al describe in their article, “Scaling-up is a craft not a science” [19]. Results were synthesised around two main themes: the numbers and the manner (the craft). Numbers are given with regard to population-based coverage of KMC and the availability of KMC services on a national or subnational level (district, region, province, state). For the numbers synthesis, we included published results that were not part of the grey literature review.
The manner in which countries approach KMC implementation and scale-up includes the “what” and the “how”: health system factors (building blocks) and contextual factors (people as catalysts for implementation and process as a pathway to implementation). These factors could also act as enablers of or barriers to successful implementation. Figure 2 is a snapshot of the interrelatedness of factors involved in the programmatic implementation of KMC.
The interrelatedness of factors involved in KMC implementation at scale.
FINDINGS
Programmatic implementation of KMC and availability of KMC services at a national or subnational level
Until around 2012, there was little movement with regard to implementing KMC on a large scale in most countries – in the grey literature, information on early scale-up was only available from Brazil [20–22], Malawi [23, 24], and South Africa [25, 26]. In 2018, 28 (32%) of the 90 countries that provided data for the Every Newborn Action Plan (ENAP) database reported having an updated national policy or guideline on KMC [27]. Project reports, conference presentations and publications suggesting that a country had implemented KMC at a national or subnational level were available from 17 ENAP countries: Bangladesh, Brazil, Ethiopia, Ghana, India, Indonesia, Iran (Islamic Republic), Malawi, Mali, Nepal, Nigeria, the Philippines, Rwanda, South Africa, Tanzania, Uganda and Viet Nam. In addition, information on KMC implementation was also available for Colombia. KMC implementation on a subnational scale was defined as implementation of KMC in at least a group of hospitals and health centres, not necessarily located in the same geographic or administrative area.
Information on KMC service availability was scarce and fragmented. Most reports did not refer to continuation of services or sustainability beyond the point of assessment or measurement. From the available documents at the time of the review we found that KMC services were reported to be available as follows: in the majority of departments of Colombia (2021) [28]; across 43% of national, regional, and provincial hospitals in the Philippines (2019) [29]; across 108 subdistrict health complexes and higher-level facilities (district hospitals and medical college hospitals) in urban and rural areas in Bangladesh (2020) [30]; in 11 districts in Malawi (2017) [31]; in 18 of the 63 provincial and three national hospitals in Viet Nam (2015) [32]; in more than 80% of the hospitals in the North West Province of South Africa (2015) [33]; in 53% of level-3 hospitals in the Islamic Republic of Iran (2012) [34]; and in 121 health facilities in Malawi (2012) [24].
We could only find one grey literature document reporting KMC scale-up in high-income countries. A cross-national survey performed in around 2018 among level-III neonatal units in Spain reported that 71% of the responding neonatal intensive care units practised kangaroo care [35].
Despite information on availability of KMC services, there were questions about the uptake and quality of services [33, 36]. In addition, there did not seem to be consensus on the minimum requirements for a KMC service at different levels of care.
Population-based coverage of kangaroo mother care: reliable estimates lacking in most countries
In 2017, 20% of the 90 countries contributing to the ENAP database reported having an indicator in their national HMIS for the number of babies <2000 g initiated on KMC in facilities with KMC services [37]. However, KMC coverage estimates for preterm or LBW newborns (<2000 g or <2500 g) are lacking in most countries, even in countries that have implemented KMC on a national or subnational scale.
The review of the 212 documents found three documents we had classified as A (quantitative information on high population-based KMC coverage, >50%), reporting population-based coverage of KMC from two countries (Colombia and the Philippines) and one district in Malawi. For Colombia, data on KMC coverage was available for the period 2015–2021. In 2015 it was estimated that in four cities more than 50% of eligible newborns had access to KMC (implying that they had received continuous KMC), with the estimates for five more cities ranging between 22% and 44% [38]. In 2018 it was estimated that five cities had a coverage of more than 80%; another 19 had a 50–80% coverage; and three had under 50% [39]. An updated document received in 2022 estimates that 63.2% of eligible babies had access to KMC in 2021 [28].
By 2019 an estimated 53% of preterm and LBW babies were receiving KMC in the Philippines [29]. Only one district in Malawi (the Thyolo district) reported a coverage of more than 50% in 2016 (64%), and this dropped to 49% in 2017 [31]. The national KMC coverage (measured as the number of expected cases of babies <2000 g initiated on any KMC) was estimated to be 22% in 2017, based on 2014–2017 data from 11 of 28 districts [31, 40]. However, data extracted from maternity reports onto the District Health Information Software 2 (DHIS2) in 2021 suggest that fewer than 10% of preterm or LBW newborns in Malawi received facility-based KMC [41].
With regard to documents classified as B (quantitative information on ANY KMC coverage), four documents on facility-based populations were found in the grey literature for the period 2015–2017. A Rwanda presentation reported facility-level KMC admission rates of LBW newborns in 10 districts in the 2016 financial year as being 55%, 68%, 61% and 46%, for the four quarters respectively [42]. From 1 January 2015 to 31 July 2017, about 800,000 newborns were treated in special newborn care units in India, of whom 50,959 (13.2%) received KMC, with variations from 0–31% between states, according to a report for a working group [43]. A conference abstract from Viet Nam indicated that in 2015, according to routine national data, 8.2% of LBW newborns received KMC; however data from seven provincial hospitals showed a 22% rate [32]. In two KMC district-hospital demonstration sites established through the Vriddhi project in India, an increase in KMC initiation from 25% to 45% to 63% was reported over three quarters of 2017 [44].
According to a report on data use in Malawi, which has KMC indicators in its DHIS2, “data completeness and timeliness remain a challenge” (p1) [40]. A brief on a multi-country evaluation in Africa referred to the importance of achieving an equilibrium so that services are not expanded too rapidly at the expense of quality care [36].
Health system factors involved in the programmatic implementation of KMC at scale: what countries have done
Because of the lack of information available on high-coverage, population-based KMC in the grey literature and on what these packages entailed, we used the health systems building blocks to organise the key factors reported to have facilitated large-scale implementation of KMC in seven countries for which sufficient information was available. Our findings are presented in Table 1. The categorisations in two published papers were combined into seven building blocks to organise the findings [11, 45]. It should be noted that in the same country, subnational entities may have different levels of implementation success, depending on the level of prioritisation of the care of small and sick newborns and various contextual, geographic, social, economic, and political factors outside the health system. The seven countries referred to are at different stages of national scale-up and some of the lessons listed in the table were learned within projects that had not been rolled out nationally at the time of reporting. If a country is not included in the table or an included country does not have a grey dot for a particular lesson, it merely means that information was not found in the documents consulted for the grey literature review.
Lessons from national KMC programmes by health system building block
Contextual factors influencing the programmatic implementation of KMC at scale: how countries have done it
Contextual factors cut across health system building blocks. Two main factors emerged: people as catalysts in implementation and process as pathways to implementation. People are key in driving programmatic KMC implementation and there is a need for appropriate capacity development. Then there are processes of KMC implementation that determine different pathways at different levels of national and subnational health systems and across countries.
People: catalysts in driving programmatic KMC implementation
Enthusiastic, committed individuals or passionate drivers (individuals, groups, agencies) are mentioned as the key to implementation success, be it at facility level, subnational or national level or at a broader regional or global level [46–51]. People play a role in leadership and governance, health system financing, the health workforce, health service delivery, and community ownership and partnership.
Various terms are used interchangeably in the grey literature to refer to individuals or groups of people as catalysts or change agents: stakeholders; role players; drivers; champions [16, 52–55]. We synthesised these under the umbrella term of “implementing partners” in Table 2, each with a number of topics to describe possible influences. The topics included people, entities, resources, support, and challenges.
Description of implementing partners
Process: initial pathways to KMC implementation and scale-up
Originally KMC was introduced at national, subnational or health-facility level in different fashions, with KMC services mostly introduced – after exposure to KMC [16, 53] – by committed and personally motivated professionals in individual health facilities. At the Istanbul KMC Acceleration Convening in 2013, De Graft-Johnson referred to a number of programmatic approaches to the introduction and expansion of KMC services [9], which are summarised in Table 3 with further explanations and examples.
Programmatic approaches to the introduction and expansion of KMC services
We constructed a crude categorisation in an attempt to understand the history of KMC implementation and scale-up over more than three decades (Figure 3). There were two main pathways of implementation and scale-up, both starting with some form of KMC practice at individual hospitals. One pathway emerged from within the public health system with donor support channelled through government structures. Brazil [20, 21] and South Africa [25, 26] are examples of this pathway. In 2000, Brazil included KMC in its Sistema Único de Saúde package of medical assistance paid for by the government [20] whereas since 1994 South Africa has provided free public health care services to mothers and children under five years of age [56]. The second pathway included an intermediate phase where an external driver provided technical support for KMC implementation, often in a small group of health facilities or in one or more districts in a country. This corresponds to the Vertical Phase-in approach mentioned in Table 3.
Broad categorisation of trajectories of KMC implementation.
External drivers serving as catalysts for KMC implementation included international non-governmental organisations, agencies and donors. Save the Children through its Saving Newborn Lives programme was one of the agencies that were active in many countries. Malawi is the country that appears to have received the most external support over more than two decades and KMC space and units have expanded rapidly there. Despite this expansion, population-based coverage is still low.
In Colombia and the Philippines a national KMC foundation acted as an external driver in the expansion of in-country KMC services. Their roles are described in detail in Table 2. The Indian KMC Foundation supports the government of the country in implementing KMC and is active in organising annual conferences to showcase research and experience in different clinical and implementation aspects of KMC [57]. A world first for KMC was the scale-up of KMC to 10 hospitals across Cameroon through a Development Impact Bond (DIB) [58]. The Fondation Kangourou Cameroun led the in-country implementation with the support of the Colombian Foundation. The model is data-driven, providing for rigorous performance management, but also allowing for flexibility in the adaptation of KMC service delivery to the local context of individual health facilities [59, 60].
Some countries adopted a more holistic and integrated programmatic approach to KMC implementation. For example, in Rwanda newborn care improvements are coordinated by the Ministry of Health in Rwanda and KMC implementation was embedded in the revitalisation of neonatal units in district hospitals. Collaboration between partners enabled the procurement of equipment, the development of national KMC guidelines and newborn protocols, coordinated training and formative supervision [61].
DISCUSSION
Exploring the synthesis of grey literature on the programmatic implementation of KMC involved a journey through uncharted territory. Despite the large volume of documents screened, sifted and analysed, few reports on population-based KMC and the national or subnational availability of KMC services were found. Health system factors involved in the programmatic implementation of KMC were identified and the lessons learnt were organised along the lines of the health system building blocks. Contextual factors were divided into people as catalysts for change and pathways initially followed in the implementation of KMC.
In the grey literature, information on efforts to programmatically scale up KMC services at national and subnational level was available for 18 low- and middle-income countries. In most of these countries there is a strong presence of development agencies with a focused maternal and newborn health agenda. With regard to uncoordinated and scattered implementation, Aliganyira et al observed “a highly inequitable distribution of KMC services” in Uganda (p6) [62]. This points to the urgent need to collect and analyse data on KMC-related indicators at subnational level to ensure equity of intervention impact in the most vulnerable populations.
Possible explanations for the dearth of data on national KMC services from higher-income countries include health system differences and the emphasis on the autonomy of clients. It appears as if the implementation of some form of KMC practice is driven by members of professional associations in the individual health facilities that incorporate KMC into their protocols for small and sick newborns without seeing a need to publicise a KMC service or collect and analyse KMC-specific data for individual patients. The launching of the International Policy Statement for Universal Use of Kangaroo Mother Care for Preterm and Low Birthweight Infants [63] in 2016 can be seen as an attempt to leverage the influence of professional associations in the acceleration of KMC implementation. A systematic review and realist synthesis of KMC scale-up in the United Kingdom identified lack of training, knowledge and confidence, and absence of clear guidelines as the main barriers. The authors propose a focus on the potential cost-effectiveness of KMC in reducing the need for incubator use [64].
Overall, the “what” and “how” of KMC implementation and scale-up could be equated to the capacity development needed to implement a national or subnational KMC programme. According to the United Nations Development Programme, capacity development is a process of change that could include various types of education, training and workshopping activities. Capacity development also refers to the strengthening of the health system to improve performance and ensure sustainability [65].
The health system factors influencing the programmatic implementation of KMC are also reported on in the published literature in terms of health system building blocks [11, 45, 66–68]. One issue emerging from the review is the challenge of establishing a continuum of care through linking the health facility with the community and changing negative community and family perceptions around preterm birth that may influence the uptake of and adherence to KMC, especially after discharge from the health facility. These issues are also reflected in a number of systematic and scoping reviews on barriers to and enablers of the adoption of KMC [5, 7, 69, 70]. Achieving intensive involvement of community health workers who pay frequent, regular home visits still seems to be a hurdle in many communities [71], although studies by Mazumder et al [72] and Mony et al [11] have respectively shown that better health outcomes and high KMC coverage are achieved where there are frequent visits by community health workers.
The context of programmatic KMC implementation is navigated by the people driving the process. This is linked to the health system building block of leadership and governance, within the health system and beyond. Of particular importance is the way successful people working as individuals and in teams can leverage change and create a potentially sustainable programme. Many of the documents in the grey literature were produced by implementing partners described in Table 2 as donors, development agencies and international non-governmental organisations. The challenges of fragmentation and “end-of-project syndrome” (i.e. end of funding) associated with this category of implementing partners are captured by Spicer et al in a publication titled: “The development sector is a graveyard of pilot projects!” [73]. They also found that innovations must be perceived as effective, require modest resource inputs, be acceptable to and incentivise frontline health workers, be acceptable to communities, and be adaptable across diverse geographical contexts [73]. These conditions could also be applied to the programmatic implementation of KMC in the design of implementation and scale-up plans and in advocacy messages.
Capacity development of the health workforce at all levels is indispensable in any KMC implementation. These levels include the public health leaders, administrators and programme managers at national and subnational level and personnel at health facility and community level. Initially, when KMC was still considered a novel intervention, stand-alone training in KMC was often the order of the day [74]. A challenge in the current environment where KMC is included in the training packages for the care of small and sick newborns is to ensure that different approaches to the organisation of KMC services and different types of knowledge and skills needed for different levels of facilities are accounted for responsibly.
The categorisation of three decades of KMC implementation and scale-up in Figure 3 is similar to the three waves of KMC implementation as described in a case study on three Asian countries [74]. KMC implementation and expansion are currently far more deeply embedded in the upgrading and scale-up of neonatal services. In a study conducted on the scale-up of maternal and newborn health innovations in Ethiopia, India and Nigeria, Spicer and colleagues concluded “that scale-up has no magic bullet solution – implementers must embrace multiple activities, and require substantial support from donors and governments in doing so” (p30) [19]. In Table 4 we use a stages-of-change framework initially developed for monitoring progress with KMC implementation [75, 76] to reconstruct a number of possible actions for scaling up a national KMC programme, with a special focus on the health system and health facilities. For each action, governments should find contextually appropriate answers and solutions [11]. Some actions pertain to KMC specifically; others should be aligned with the newborn health agenda and programme.
What is needed to scale up a national KMC programme?
Study limitations
This review illustrated how hard it is to find relevant information on an intervention that is embedded in a broader package of small and sick newborn care. Important documents may also have been missed because not all languages were accommodated. Governments and development partners in some countries seem to be more forthcoming than others regarding the documents they place on the internet and their activities naturally received more attention in the review. The reviewer’s involvement in the assessment of KMC services of many countries has influenced the analysis, but has also enabled the inclusion of documents that might otherwise not have been found. Lastly, the review did not include an in-depth analysis of how certain cross-cutting themes featured in the programmatic implementation of KMC because of the large volume of documents and uncertainty about how these themes could be identified with common search terms. Broader themes derived from the ENAP’s guiding principles are examples of such issues for future attention: human rights; integration; equity; accountability; and innovation [77]. Some of the information summarised in this review has also found its way into peer-reviewed publications, which were not covered by this study.
CONCLUSION
The synthesis of the grey literature contributes to the understanding of how programmatic implementation of KMC takes place and which aspects could be considered in the design and scale-up of such programmes. Two of the main takeaways from this review are: avoid “empty scale-up” with an “appearance of success” [78], and “balance rapid expansion of services with the need to improve quality of care” [36] for small and sick newborns. Programmatic implementation of KMC requires action in multiple health system building blocks with a focus on monitoring and evaluation of availability and quality of services along with coverage.
Data Availability
Materials are available upon reasonable request from the corresponding author.
Acknowledgments
We are grateful to individuals, governments and organisations who shared documents not available on the internet, to Karen Edmond for her valuable inputs into the conceptualisation of the review and to Nils Bergman for reviewing the original work plan.
Footnotes
Disclaimer: The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the World Health Organization or the University of Pretoria.
Funding: The review work was supported by a small grant to the University of Pretoria, South Africa, by the World Health Organization.
Authorship contributions: A-MB and SR conceptualised the review; A-MB conducted the review; SG provided intellectual support throughout the review; A-MB drafted the first manuscript and SG and SR reviewed and modified the draft in an iterative process until all authors had approved the final manuscript submitted for publication.
Competing interests: SG is a staff member of the World Health Organization and SR a consultant. AM-B declares no conflict of interest.
Ethics approval. Not applicable.
Availability of data and materials. Materials are available upon reasonable request from the corresponding author.
Additional material: Six online supplementary files.
REFERENCES
- 1.↵
- 2.↵
- 3.↵
- 4.
- 5.↵
- 6.↵
- 7.↵
- 8.↵
- 9.↵
- 10.↵
- 11.↵
- 12.
- 13.↵
- 14.↵
- 15.↵
- 16.↵
- 17.↵
- 18.↵
- 19.↵
- 20.↵
- 21.↵
- 22.↵
- 23.↵
- 24.↵
- 25.↵
- 26.↵
- 27.↵
- 28.↵
- 29.↵
- 30.↵
- 31.↵
- 32.↵
- 33.↵
- 34.↵
- 35.↵
- 36.↵
- 37.↵
- 38.↵
- 39.↵
- 40.↵
- 41.↵
- 42.↵
- 43.↵
- 44.↵
- 45.↵
- 46.↵
- 47.
- 48.
- 49.
- 50.
- 51.↵
- 52.↵
- 53.↵
- 54.
- 55.↵
- 56.↵
- 57.↵
- 58.↵
- 59.↵
- 60.↵
- 61.↵
- 62.↵
- 63.↵
- 64.↵
- 65.↵
- 66.↵
- 67.
- 68.↵
- 69.↵
- 70.↵
- 71.↵
- 72.↵
- 73.↵
- 74.↵
- 75.↵
- 76.↵
- 77.↵
- 78.↵
- 79.
- 80.
- 81.
- 82.
- 83.
- 84.
- 85.
- 86.
- 87.
- 88.
- 89.
- 90.
- 91.
- 92.
- 93.
- 94.
- 95.
- 96.
- 97.
- 98.
- 99.
- 100.
- 101.
- 102.
- 103.
- 104.
- 105.
- 106.
- 107.
- 108.
- 109.
- 110.
- 111.
- 112.
- 113.
- 114.
- 115.
- 116.
- 117.
- 118.
- 119.
- 120.
- 121.
- 122.
- 123.
- 124.
- 125.
- 126.
- 127.
- 128.
- 129.
- 130.
- 131.
- 132.
- 133.
- 134.
- 135.
- 136.
- 137.
- 138.
- 139.
- 140.
- 141.
- 142.
- 143.
- 144.
- 145.
- 146.
- 147.
- 148.
- 149.
- 150.
