Mega‐map of systematic reviews and evidence and gap maps on the interventions to improve child well‐being in low‐ and middle‐income countries

Abstract Background Despite a considerable reduction in child mortality, nearly six million children under the age of five die each year. Millions more are poorly nourished and in many parts of the world, the quality of education remains poor. Children are at risk from multiple violations of their rights, including child labour, early marriage, and sexual exploitation. Research plays a crucial role in helping to close the remaining gaps in child well‐being, yet the global evidence base for interventions to meet these challenges is mostly weak, scattered and often unusable by policymakers and practitioners. This mega‐map encourages the generation and use of rigorous evidence on effective ways to improve child well‐being for policy and programming. Objectives The aim of this mega‐map is to identify, map and provide an overview of the existing evidence synthesis on the interventions aimed at improving child well‐being in low‐ and middle‐income countries (LMICs). Methods Campbell evidence and gap maps (EGMs) are based on a review of existing mapping standards (Saran & White, 2018) which drew in particular of the approach developed by 3ie (Snilstveit, Vojtkova, Bhavsar, & Gaarder, 2013). As defined in the Campbell EGM guidance paper; “Mega‐map is a map of evidence synthesis, that is, systematic reviews, and does not include primary studies” (Campbell Collaboration, 2020). The mega‐map on child well‐being includes studies with participants aged 0–18 years, conducted in LMICs, and published from year 2000 onwards. The search followed strict inclusion criteria for interventions and outcomes in the domains of health, education, social work and welfare, social protection, environmental health, water supply and sanitation (WASH) and governance. Critical appraisal of included systematic reviews was conducted using “A Measurement Tool to Assess Systematic Reviews”‐AMSTAR‐2 rating scale (Shea, et al., 2017). Results We identified 333 systematic reviews and 23 EGMs. The number of studies being published has increased year‐on‐year since 2000. However, the distribution of studies across World Bank regions, intervention and outcome categories are uneven. Most systematic reviews examine interventions pertaining to traditional areas of health and education. Systematic reviews in these traditional areas are also the most funded. There is limited evidence in social work and social protection. About 69% (231) of the reviews are assessed to be of low and medium quality. There are evidence gaps with respect to key vulnerable populations, including children with disabilities and those who belong to minority groups. Conclusion Although an increasing number of systematic reviews addressing child well‐being topics are being published, some clear gaps in the evidence remain in terms of quality of reviews and some interventions and outcome areas. The clear gap is the small number of reviews focusing explicitly on either equity or programmes for disadvantaged groups and those who are discriminated against.


| Screening, data extraction and critical appraisal
Title and abstract screening and the evidence classification were undertaken by two independent reviewers, and any discrepancies were resolved by a third reviewer. The studies that passed on to full text stage were screened against the eligibility criterias by two independent reviewers and conflicts resolved by a third reviewer.
After screening, all studies were coded for a wide array of information and populated into the map. The studies were coded for bibliographic details, the interventions and outcomes and other relevant aspects such as population, region and countries. The coding was again carried out by two independent reviewers and conflicts were reconciled by a third reviewer. Critical appraisal appraisal of all included studies was performed by independent reviewers.

| Results
The mega-map includes 333 systematic reviews and 23 EGMs. A main finding from the map is that there is a substantial amount of evidence in the traditional areas such as early childhood development, health and nutrition and education. However, there is uneven distribution of evidence across region, quality of studies and intervention and outcome subcategories.
For example, though health and nutrition were the most populated areas of the map, some gaps were identified; management of severe acute malnutrition (8), agricultural intervention/biofortification (13), m-health intervention for child health (16). In education, there is limited evidence around remedial education, nonformal education and scholarships and systemic intervention. In terms of quality, about 26% (58) of the systematic reviews were identified to be of low quality for health and nutrition. This was found to be similar for early childhood development (30%) and education (22%).
Striking gaps in evidence were identified in the nontraditional areas such as social protection and social welfare, with lack of evidence in many areas such as child trafficking, substance abuse prevention, interventions for child abuse, social insurance and labour market insurance. These were not just under-researched but also under-funded. The most funded systematic review areas have been health and nutrition and early childhood development over the years.
The major funders of such reviews have been the DFID, World Health Organisation (WHO), Bill and Melinda Gates Foundation, National Institute of Health Research (NIHR) and UNICEF.
Health and learning outcomes were found to be the most studied as well. Few studies were identified assessing the impact of interventions on risk reduction, safety, equity, and economic outcomes.
UNICEF Strategic Goal 1 (Every child survives and thrives) and Goal 2 (Every child learns) are heavily populated. Goal 3 (Every child is protected from violence and exploitation) and Goal 4 (Every child lives in a clean and safe environment) are less populated and there is scant evidence in Goal area 5 (Every child has an equitable chance in life).

| Conclusion
Although an increasing number of systematic reviews addressing child well-being topics are being published, there is a need for more studies across some interventions and outcome areas such as social protection and social welfare. The clear gap is the small number of reviews focusing explicitly on either equity or programmes for disadvantaged groups and those who are discriminated against. There is a need for more studies which are explicitly focused on programmes for vulnerable children.
There are limited systematic reviews focusing on gender and equity issues in child well-being. Very few studies were identified assessing programme cost of the interventions.
It would be useful for global and country partners to work together to achieve consensus on priority areas for evidence synthesis and to adopt a coordinated approach to undertaking and updating maps and reviews across priority areas.

| The state of child well-being in LMICs
Child well-being is a multidimensional and a holistic concept which provides a contextual understanding of a child in different domains such as health, material well-being, education, conditions of housing and environment, and interpersonal relations (UNICEF, 2014). A decent level of child well-being is underpinned by the UN convention of Rights of the Child which states that "The child, by reason of his physical and mental immaturity, needs special safeguards and care, including appropriate legal protection, before as well as after birth" (UNICEF, Convention on the Rights of the Child, 1989). But many children around the world still suffer deficiencies in many dimensions of well-being.

| Aspects of shortfalls in child well-being
According to UNICEF's "State of the World's Children" report (2017), there are 2.2 billion people aged under the age of 18 and the majority of these are living in poverty. Evidence from many countries shows that children who grow up in poverty are more likely to experience poor health, fewer opportunities to access good quality education and to be low paid or unemployed in the future (UNICEF 2017). One in three children (200 million globally) fails to reach their full physical, cognitive, psychological and/or socioemotional potential due to poverty, poor health and nutrition, insufficient care, stimulation and other risk factors related to early childhood development (Grantham-McGregor, et al., 2007). It is estimated that globally, almost 385 million children are living in extreme poverty. Poverty, malnutrition, poor health, unstimulating home environments and violence against children are major risk factors, which detrimentally affect the cognitive, motor and socialemotional development of children (Tran, Luchters, & Fisher, 2017).
Children living in extremely poor households are unevenly concentrated in certain parts of the developing world. Sub-Saharan Africa has the highest rates of children living in extreme poverty at just under 49%, and South Asia has the second highest share at nearly 36% (UNICEF, Ending extreme poverty: a focus on children. Sustainable Development Goals (SDGs) in Rich Countries, 2016). In 2017, the under-five mortality rate in low-income countries was 69 deaths per 1,000 live births-around 14 times the average rate in high-income countries (five deaths per 1,000 live births). Preterm birth complications, acute respiratory infections, intrapartum-related complications, congenital anomalies and diarrhoea are the main factors continuing causes of high numbers of under-five deaths (WHO, 2019). Globally, millions of adolescents die or become sick from preventable causes such as road accidents, HIV, suicides and interpersonal violence (WHO, 2018). Further adding to the plight of children in developing countries, young girls and adolescent women are invariably subjected to various forms of harmful practices including child marriage and female genital mutilation. Education offers children a path to a promising future but about 264 million children and adolescents around the world fail to enter or complete school (UNESCO, 2017). These figures are alarming in 24 conflict-affected regions with approximately 27 million children out of school (UNICEF, Goal Area 2 Every child learns-Global annual report, 2018). They are thwarted by poverty, discrimination, armed conflict, emergencies and the effects of climate change. Often the family and environmental risk factors a child experiences are beyond their control. However, the effects of these factors can be moderated and this is where opportunities to promote children's well-being and positive mental health lie.

| Consequences of shortfalls in child well-being
Childhood deprivation cannot just take childhood from children, but also have long-run consequences. Child undernutrition is associated with shorter adult height, less schooling, reduced economic productivity and lower offspring birthweight for women (Victoria et al., 2008). Lack of education is a major factor in households remaining poor (Baulch, 2011).
Early marriage is bad for the health of the mother-with greater risk of dying during childbirth-and her offspring who are at greater risk of having low birth weight and of dying prematurely (Nour, 2006).

| Addressing shortfalls in child well-being
The provision of services on health, education and safety to all children in the world irrespective of cast, creed, colour and ethnicity is a fundamental right enshrined in the UN Convention on the Rights of the Child (UNICEF, Convention on the Rights of the Child, 1989 strategies. Though child well-being interventions have been underway for decades; evidence on the effectiveness of these interventions is often scattered, the value is possibly underestimated and inclusion in national strategies and programmes is rare. Failure to effectively implement evidence-informed interventions represents a key obstacle in the progress of child well-being in many LMICs towards achieving the SDG's. This is partly due to a weak evidence base that does not give policy makers and programme managers the information needed to make decisions. Both international and national organisations should work together to fill the gaps in evidence and to gain a better understanding of what works and what does not in child well-being.

| The intervention
The included interventions cover all main strategies to improve child well-being outcomes. The seven intervention categories are: Child protection regulation informed policymaking is seen to be of increasing importance, many agencies commission systematic reviews to inform policy, but due to lack of a central global repository, systematic reviews are often duplicated.
Furthermore, they sometimes lack in quality to clearly and correctly inform policy and the existing evidence base around child well-being has major gaps. Evidence maps are an approach to providing an overview of the available evidence, with various approaches adopted to evidence mapping by different agencies over the years (Saran & White, 2018).
Since this map has such a broad scope-all of child well-being-we label it as a mega-map and map only systematic reviews and EGMs. As a subsequent step, the two organisations are worked together on an EGM on ending violence against children in LMICs.

| Existing EGMs and/or relevant systematic reviews
Since this is a mega-map and has a very broad scope, it is the first map of its kind in this area. But there are related EGMs that are included in the mega-map.

| Defining EGMs
Campbell EGMs are based on a review of existing mapping standards (Saran & White, 2018) which drew in particular on the approach developed by 3ie (Snilstveit, Vojtkova, Bhavsar, & Gaarder, 2013). As defined in the Campbell Collaboration EGM Guidance, a mega-map is a map of evidence synthesis, that is, systematic reviews, and does not include primary studies (Campbell Collaboration, 2020).
The primary dimensions are the rows and columns of the map which are, respectively, intervention categories (and subcategories) and outcome categories (and subcategories). Secondary dimensions, such as World Bank region, systematic review quality and population target groups were included as filters.

| Types of evidence
The mega-map includes only systematic reviews and EGMs of effects of interventions. The key characteristics for a review to be included as a "systematic review" are as follows: 1. A clearly stated set of objectives with predefined eligibility criteria for studies.

A systematic search that attempts to identify all studies that
would meet the eligibility criteria.
4. An assessment of the validity of the findings of the included studies, for example, through an assessment of risk of bias.

5.
A systematic presentation, and synthesis, of the characteristics and findings of the included studies.
Relevant ongoing systematic reviews were also included but they were not assessed for quality. We did not include qualitative studies and the search was restricted to English language only.
We also included EGMs and it was defined as an approach for systematic presentation of all relevant evidence of a specified kind for a particular sector, subsector or geography. They typically contain systematic reviews and primary studies, but may include only one of these.
There are many approaches to evidence mapping (Saran & White, 2018 and we included any map on child well-being interventions.

| Type of population
The primary population of interest for this mega-map is children 18 years and under as per the definition by United Nation Convention (UNICEF, Convention on the Rights of the Child, 1989) and includes children from LMICs 1 ). Different child age ranges; infants (0-3 years), child (0-10 years), adolescent (10-18 years) were added as filters.
Population subgroups of interest include: children in conflict-affected regions, children with disabilities, children from underrepresented communities (low income and ethnicity) and malnourished children.

| Types of interventions/problem
The framework-matrix of interventions and outcomes was primarily based on UNICEF's Strategic Plan (UNICEF, 2018).
The included interventions cover all main strategies to improve child well-being outcomes. The seven main intervention categories are as mentioned below. The description for these categories is given in Annex 1.

| Types of outcome measures
The categories cover seven broad outcomes as mentioned below:

| Search methods and sources
The mega-map was developed in three stages: • Stage 1: included a search of relevant systematic reviews and EGMs from 3ie databases. This initially included 99 studies.
• Stage 2: this mapped the studies from a full search of databases and included 302 systematic reviews The search was as comprehensive as possible, using (but not limited to) relevant bibliographic databases and EGM databases, web-based search engines, websites of specialist organisations and bibliographies of relevant reviews. Additionally, reference lists of the included reviews were reviewed.

| Stakeholder engagement
The framework was developed through a consultative process. An advisory board was formed comprising of key experts in the area: In consultation with the expert group, intervention and outcome categories were agreed and used to set the scope of the mega-map.

| Scope
Developing the scope is one of the critical stages in an EGM. Specifically, we identified intervention and outcomes to map that formed the EGM framework.
Stages in framework development: 1. The initial framework-matrix of interventions and outcomes was developed based on UNICEF's strategic plan (UNICEF, 2018).
2. We then referred to strategic plan documents from other major funders and implementers on child well-being programmes including UNESCO, WHO and the World Bank to identify any additional key intervention and outcomes of relevance.
3. The proposed framework was revised based on feedback received by advisory board (mentioned above).
4. We then pilot-coded the framework using 30 key systematic reviews from the 3ie database and the framework revised further based on the exercise.

A mapping workshop was organised in London
where key stakeholders and experts in the area were invited to participate. They reviewed the categories in an interactive exercise to fit the identified papers into the suggested categories and the feedback was reviewed.
The final framework aimed to provide an overview of existing systematic reviews and EGMs for child well-being interventions such as: 1. Early childhood interventions that addressed the period from pregnancy, childbirth and children up to 3 years of age.
2. Health and nutrition interventions that addressed issues such as maternal health, timing and spacing of birth, childbirth, nutrition, prevention and treatment of childhood diseases.
3. Educational interventions that aimed to address improved learning and achievements and skills development of children from 4 to 18 years of age.
4. Social work and welfare interventions that aimed to protect the child from violence and other risk factors.
5. Social protection interventions that aimed to provide financial support to mother, children and families to access the basic amenities for survival and living.
6. Environmental WASH interventions to ensure that every child lives in a clean and safe environment.
7. Governance and advocacy intervention strategies to ensure sustainability.
6.1.1 | Conceptual framework Figure 1 presents a simplified theory of change which may underlie improvement in child well-being outcomes. A crucial step is successful advocacy and social mobilisation. We assume that advocacy is one of the keys to setting priorities and improving policies to leverage child well-being outcomes. Additionally, social mobilisation can help engage a broad range of actors so that they become involved in driving change, by raising awareness and by creating the demand for better policies, services and accountability. Both these components together help identify different target population and creating a demand and supply for better policies and services. Child well-being encompasses numerous domains such as early child development, health and nutrition, education, social work and welfare, social protection and WASH and the interventions targeting each of them are represented in the conceptual framework. Following adoption of these interventions, several outcomes are then expected to improve such as on health, education, social, economic and equity. Since UNICEF's work is structured around five overarching areas of well-being for every child who is grounded in the 2030 Agenda for Sustainable Development, improvement in any of these child well-being outcomes, will directly or indirectly effect fulfilment of UNICEF's goals.

| Description of intervention
The interventions included cover all main strategies to improve child well-being outcomes. Table 1 lists the intervention subcategories under each of these headings.

| Description of population/geographic location
In addition to interventions and outcomes, the following filters were coded in the systematic reviews and mega-map: (1) Population subgroups of interest include: children with disabilities, children in conflict-affected regions, children from underrepresented communities (low income, ethnicity, race) and malnourished children.
(2) Regions: East Asia and Pacific, Europe and Central Asia, Latin America and Caribbean, Middle East and North Africa, North America, South Asia, Sub-Saharan Africa and conflict affected regions.
(4) Country classification by Income level: low-income: <1,005; lower- The screening of studies in relation to inclusion/exclusion criteria was undertaken in two stages in EPPI-reviewer.
The screening was carried out based on predefined eligibility criteria. We did not exclude any study based on study quality or outcomes ( Ideally the agreement rate should be 80% between reviewers. However, for this map we had to work with 70% agreement rate for coding as even with repeated rounds of discussion and number of revisions in categories, the team was unable to reach 80% agreement, mainly due to overlap in health related intervention and outcome categories.

| Data extraction and management
We used a standardised data extraction form (presented in Annex 4) to extract relevant information from reviews and EGMs. For each included study, two coders independently extracted the following information: bibliometric information, study descriptions, population and target groups, intervention, outcomes, regions and funders. All the included systematic reviews were subject to critical appraisal and quality rated.

| Tools for assessing systematic review quality
We used a standardised checklist, AMSTAR-2 (Shea, et al., 2017) to assess our confidence in the findings of each systematic review. The confidence ratings do not appraise the studies included in a review, but rather the methodology and reporting of the review. AMSTAR-2 is a 16 item checklist and covers: Items 2, 4, 7, 9, 11, 13 and 15 are termed "critical." Study quality is rated high if there is no more than one noncritical weakness and medium if there is one or more noncritical weakness but no critical weakness.
Studies with one or more critical weaknesses are rated low quality. As a result, a total of 356 records were included in the final map, of which 54 were identified from grey literature and reference checking. Of these, 333 3 are systematic reviews and 23 are EGMs. All of the studies identified through the search and screening process T A B L E 2 Eligibility criteria for systematic reviews and EGMs

Include Exclude
Literature type Systematic reviews and EGMs on interventions on child well-being.
From: -Reviews on clinical drug trials -Literature reviews -Rapid reviews -Prevalence reviews -Qualitative reviews -Association reviews -Commentary or editorials Published journal article Grey literature: -Working paper -Report -Thesis papers Population -Studies on children 18 years and below.
-Studies with multiple populations were included if they included children from age category as above.
-Systematic reviews were included if the "inclusion criteria" in the review was to include the studies conducted in LMICs.
-Studies specifically on populations above 18 years -Systematic reviews for which studies from only highincome countries were eligible for inclusion Interventions -Interventions to improve well-being of children: health, education, social protection, social work and welfare, Governance  Health and nutrition was the most reviewed sector, followed by early childhood development. This is possibly because 3ie started commissioning EGMs from 2010 onwards and many organisations like IRC adopted the 3ie methodology after 2010. Education was the most reviewed sector followed by health and nutrition and social protection.

| Distribution of studies across Intervention and outcome categories
The health quadrant of the EGM which maps the studies of the effects of health interventions on health outcomes-is the most heavily populated section of the map (Tables 3 and 4).  (132). Limited evidence was identified on learning and development (28), risk factor reduction (220 and economic impact (18) of these interventions. Only one systematic review was identified with an equity focus. About 30% of the reviews were identified to be of low quality (Figure 6).

EGMs
There were seven EGMs identified covering early childhood education and parenting, four on maternal education and parenting, three on early childhood nutritional interventions and one on early childhood health interventions.
Healthy development (eight) was the most commonly studies outcome, five on safety and five assessed equity as an outcome measure. Four EGMs assessed impact on health and three on learning and development and risk factor reduction each.

| Distribution of studies across health and nutrition
Systematic reviews The outcomes covered by systematic reviews mainly fall in two domains, health impact (188) and healthy development (162). About 26 studies were identified assessing the economic impact of these interventions. Only nine studies were identified assessing equity outcomes.
Over 100 studies relate to the main outcomes of mortality, morbidity, disability and childhood illness and nutrition. There were fewer studies relating to risk factors notably on childhood injuries, alcohol and substance abuse, handwashing, clean environment, diet and physical activity.
In terms of quality, 26% of child health reviews are classified as low-quality (Figure 7).

Evidence and gap maps
Thirteen EGMs were identified in the area of health and nutrition, with the majority covering community health interventions including CHWs (7), mental health programmes (5), antenatal care and postnatal care by TBA/SBA (4), mass media campaigns on health education (4). Healthy development (10) is the most common outcome covered, followed by health (7) and economic impacts (6) and five (5) assessing equity.

Systematic reviews
The most studied interventions in the education domain were school based-health interventions (38) In terms of quality, 22% of educational reviews were identified to be of low quality (Figure 8).

Evidence and gap maps
School based interventions (nine) and decentralization and local community participation (nine) and teacher training (seven) were the most covered interventions by EGM. Alternative schooling, school sanitation and WASH and teacher incentives and pedagogical approaches were covered by six EGMs each. No EGMs were identified assessing equity component.

Systematic reviews
This is a scarcely populated domain in the map. The largest concentration of reviews in this area were on gender-based violence interventions (21), also termed as violence against children and women, followed closely by interventions on child abuse (13) and child protection services (13). There was a lack of evidence on child trafficking interventions (one) and substance abuse prevention (two). The outcomes covered by systematic review were concentrated in three outcome domains-safety (30), health (13) and healthy development (15). Only three studies measured the economic impact of these interventions. On safety outcomes, violence against children (35) was the most studied outcome.
There was limited evidence on child rights (eight) and FGM prevalence (four). Insufficient evidence was found on child labor and child marriage. There considerable lack of reviews in this intervention areas with explicit focus on equity (one).
Twenty-two percent of reviews were identified to be of low quality (Figure 9).

Evidence and gap maps
Gender-based violence interventions (six) were the most covered by EGMs, followed by interventions on child abuse (two), child trafficking interventions (two) and child protection services (two). Safety (five) is the most studied outcome in EGMs as well, followed by health (four) and healthy development (five) and equity (three).

Systematic reviews
Social assistance intervention (44) is the most concentrated intervention area, followed by social insurance schemes (13). There was lack of evidence on labour market insurance (one).
Health (31) and healthy development (34) were the most studied outcomes, followed by learning and development (15) and safety (14). Economic impact was assessed in nine reviews and no reviews were identified on equity component.
24% of the reviews were identified to be of low quality in this area ( Figure 10).

Evidence and gap maps
Social assistance (11) was the most covered intervention area followed by social insurance schemes (four) and labour market insurance (four).
Healthy development (10), learning and development (eight) and safety (eight) and equity (six) were the most covered outcomes.

Systematic reviews
On the interventions, 24 reviews were identified on improved sanitation and water and 21 on hygiene education. There was limited F I G U R E 9 Number of studies for social work and welfare by study theme and quality F I G U R E 1 0 Number of studies for social protection by study theme and quality evidence on prevention of environmental tobacco smoke (seven), safe places to play (three), prevention of outdoor and indoor air pollution (one) and traffic calming (one).
On the outcomes, health impacts (33) and risk factor reduction (21) was the most commonly studies outcomes, followed by health development (11). Limited evidence was found assessing impact on equity (six), learning and development (two), safety (one), and economic impact (three) and only one review assessing equity.
Nine reviews were assessed to be of low quality and 13 medium of quality ( Figure 11).

Evidence and gap maps
Four EGM were identified each on improved sanitation and water and hygiene education and three on safe places to play.
On the outcomes, health (four) and healthy development (four) was the most commonly studied outcomes, followed by risk factor reduction (three), safety (three), equity (three) and economic impact (two).

Systematic reviews
Governance was the least studied area of the maps with only six reviews on legislations on child rights (six) and four on child protection legislations (four). Safety (seven) was the most studies outcome, followed by health development (five) and impact on health (four). No studies were identified on equity or economic impact.

Evidence and gap maps
We identified only two EGMs on child protection legislations. These EGMs assessed impact on safety (2), healthy development (2), and one each on health, risk factor reduction and economic impact. One EGM assessed equity outcomes ( Figure 12). Figure 13 shows areas in which there is ample evidence for UNICEF strategic goals related to child well-being. As clearly stated above, Goal 1 (Every child survives and thrives) and Goal 2 (Every child learns) are heavily populated. Goal 3 (every child is protected from violence and exploitation) and Goal 4 (every child lives in a clean and safe environment) were less populated and there is scant evidence in Goal area 5 (every child has an equitable chance in life).

| Equity
There are thirteen systematic reviews in the mega-map with an explicit equity focus. That is, reviews which either address equity or focus on a disadvantaged group, such as people with disabilities.
These are concentrated in health and nutrition (six), and education intervention (five) areas. There is just one study with respect to early child development, and none at all for social protection or rights and governance. The 13 reviews identified as addressing equity report a broad range of outcomes. Most of these outcomes relate to health and education. However, there are also a number related to the UNICEF goal that every child is protected from violence and exploitation, with child abuse and neglect being reported in three of the 13 studies.
Approximately half (six out of 13) of the studies are rated as high quality.

| Population subgroups
There is a fair amount of evidence assessing the effectiveness of interventions for all age-groups, though it is highly concentrated for children in the young child (249). There was limited evidence focusing F I G U R E 1 1 Number of studies for environmental health including WASH by study theme and quality specifically on vulnerable populations such as children with disabilities (12), children in conflict-affected regions (11) and malnourished children (8). There were limited systematic reviews focusing on gender and equity issues ( Figure 14).

| Region
Systematic reviews were found to be almost equally concentrated in all regions with sub-Saharan Africa (292) and South Asia (268) having the highest number of reviews. These were closely followed by Latin

America and Caribbean (245), East Asia and Pacific (243) and Middle
East and North Africa (234) (Figure 15).   Figure 18 provides a breakdown of funders of the included EGMs.

| Funders for EGMs
The United Kingdom's DFID is the most cited funder (eight) followed by USAID (four). The most cited producer is the 3ie. 9.2 | Areas of major gaps in the evidence

| Overall gaps
Some priority areas and gaps in evidence are listed in (Figure 19).
There priorities emerge from both the gaps in the map, and stakeholder consultations on the map findings.
-There is a clear gap in reviews which are explicitly focused either on equity or on programmes for disadvantaged groups and those who are discriminated against such as children with disabilities, children belonging to ethnic minorities and low income groups.
-Very few reviews were identified that conducted cost-analysis.
-The evidence synthesis which is available is mostly in the areas of health and education. More is needed in those areas. But more still is needed in other areas where there are limited reviews such as governance. LMICs (Saran, White, & Kuper, 2020) showed primary studies to be concentrated in very few countries, and few studies addressed the social aspects of disability necessary for inclusive development.

| Gaps by each sector
A gap was also identified on management of severe acute malnutrition, which is a huge issue in many LMICs. Since undertaking this map, the Bill and Melinda Gates Foundation has commissioned ten Campbell systematic reviews in this area.
There is less information on behavioural outcomes-that is parent- There were no studies identified highlighting the impact of birth registration on child safety and well-being.

Social protection
Social protection is well covered because of the amount of reviews on conditional cash transfers, but there is a lack of evidence on social insurance and labour market insurance. Micro insurance programmes offer a financially sustainable model of social protection so more understanding of their effectiveness is important to inform policy.

Risk factors and waSH
Child injuries are an important area for which there are few reviews.
There are also no reviews of safe places to play and traffic calming despite the fact that over 90% of traffic fatalities are in developing countries. There is a new ongoing map on road safety, from which preliminary findings show over 90% of evidence comes from North America. So primary studies of traffic safety measures in LMICs are an important gap.
There are no or few studies to help identify effective strategies to reduce the exposure of children to indoor and outdoor air pollution, to reduce the risk of accidents including road-related accidents, or to reduce risk of exposure to toxic materials.

| Limitations of the EGM
The search strategy was systematic but owing to the large scope of the map, there are limitations. Some studies may have been missed, although several steps were taken to reduce this risk. For example, the search was conducted across a wide range of academic databases and the search was validated by an expert; reference and citation checking was undertaken and key donor websites were searched to F I G U R E 1 9 Priority areas and gaps in evidence SARAN ET AL.
| 23 of 44 identify grey literature. However, hand-searching of literature was not undertaken.
Although we used a consultative process that included two roundtable events to develop the framework, experts did not fully agree on all the categories. For example, there were comments that child protection services and interventions for child abuse should be a single category. Similarly, child right interventions were not fully agreed on. We worked based on the definitions we had, which could have led to judgement bias for some studies.

| Implications for research
The mega-map demonstrates major gaps in our understanding, suggesting many practices in the promotion of child well-being are largely unsupported by evidence.
The mega-map is the first step in identifying priority areas for rigorous systematic reviews and EGMs on child well-being interventions. Based on the findings, a systematic research prioritisation should now be undertaken. As an immediate development from the mega-map, UNICEF-Innocenti and Campbell Collaboration are now working on an EGM on ending violence against children in low and middle income countries (Pundir et al., 2019). A similar collaboration is now needed in all the areas of weak evidence synthesis to help improve and advance research.
➢ There were gaps in reviews with explicit focus on equity however, there may be review of reviews that contain relevant evidence, but mapping that requires diving into the review contents in more detail. A next step could be to produce an equity-focused version of the mega-map which allows identification of which reviews present evidence disaggregated for priority groups.
➢ It would be useful for global and country partners to work together to achieve consensus on priority areas for evidence synthesis and to adopt a coordinated approach to undertaking and updating maps and reviews across priority areas.
➢ Efforts are also needed to identify optimal study design, key interventions and outcome areas so that a universal evidence base can be built and used.
➢ There were a few dated systematic reviews that were identified and these need updating. Examples include in the areas of FGM, early marriage and child labour.
➢ There is a good amount of high-quality systematic reviews in health and education and this implies that there is a substantial evidence base for governments and international agencies to draw upon for programme and policy design. The map is a starting point in guiding people to the evidence: further analysis and knowledge brokering is required for the evidence to be usable.

SOURCES OF SUPPORT
The mega-map is funded by UNICEF's Office of Research-Innocenti.
Interventions to reduce prejudice and enhance inclusion and respect for ethnic differences in early childhood: A systematic review. Annual Review of Psychology, 32 (4) (2014). Interventions to prevent or reduce violence against women and girls: A systematic review of reviews. The World Bank. Retrieved from http://documents.worldbank.org/curated/en/700731468149970518/ Interventions-to-prevent-or-reduce-violence-against-women-and-girls-asystematic-review-of-reviews Arnold, B., & Colford, Jm Jr. (2007). Treating water with chlorine at pointof-use to improve water quality and reduce child diarrhoea in developing countries: A systematic review and meta-analysis. Alternative schooling/nonformal education Nonformal education refers to education that occurs outside the formal school system. Nonformal education is often used interchangeably with terms such as community education, adult education, lifelong education and second-chance education. It refers to a wide range of educational initiatives in the community, ranging from home-based learning to government schemes and community initiatives. It includes accredited courses run by well-established institutions as well as locally based operations with little funding School sanitation and WASH Interventions to ensure child friendly water supply, toilet and hand washing facilities in the schools and promote behavioural change by hygiene education

Teacher Incentives
Seek to improve the working conditions in schools so that teachers are motivated to come to work and improve their performance Remedial education Also known as developmental education, basic skills education, compensatory education, preparatory education, and academic upgrading, is assigned to assist students in order to achieve expected competencies in core academic skills such as literacy and numeracy Pedagogical approach Interactions between teachers, students, and the learning environment and the learning tasks. This broad term includes how teachers and students relate together as well as the instructional approaches implemented in the classroom. Types include teacher-centred, learner-centred and learning centred pedagogy

Social work and welfare
Development and provision of public or private social services to promote social justice amongst individuals and groups of individuals. While the term social welfare refers more generally to the well-being of groups and individuals as well as the system of social service delivery, the term social work refers more specifically to the professional practice of delivering these social services Child protection services Child Protective Services (CPS) is a branch of a state's social services department that is responsible for the assessment, investigation and intervention regarding cases of child abuse and neglect, including sexual abuse. CPS typically takes cases where a child has been abused or is believed to be at risk of abuse by someone who has caregiving responsibilities for that child Intervention for child abuse Includes clinical treatment of physical and psychological injuries, family counselling, self-help services, the provision of goods and services such as homemaker or respite care, legal action against the perpetrator, and removal of the child or the offender from the home Child-trafficking prevention/interventions Supporting training of professionals working with children including social workers, health workers, police and border officials to effectively deal with trafficking. Providing children with social services, health care, psychosocial support, and reintegration with family and community Substance abuse prevention Programmes and strategies for substance abuse preventions among children and adolescents

Social protection
Is defined as the set of policies and programmes designed to reduce poverty and vulnerability by promoting efficient labour markets, diminishing people's exposure to risks, and enhancing their capacity to protect themselves against hazards and interruption/loss of income Social insurance schemes Microfinance, employability training, vocational training and savings programmes that aim to affect child well-being outcomes Labour market interventions Interventions covering work-related injuries of employees, broadly affecting household income generation and child health Social assistance interventions Unconditional or conditional cash transfer programmes that aim to affect child well-being outcomes

Environmental health including WASH
Improved sanitation and water Refers to access to adequate and equitable sanitation and hygiene for all and end open defecation, paying special attention to the needs of women and girls and those in vulnerable situations Hygiene education Setting up a comprehensive educational programme to ensure that the community: is aware of the importance of water quality and its relation to health, and of the need for safe water supplies; accepts the importance of surveillance and the need for a community response; understands and is prepared to play its role in the surveillance process; has the necessary skills to perform that role.

Prevention of outdoor and indoor air pollution
There are myriad forms and sources of indoor pollution, including combustion (tobacco smoke, stoves, fireplaces, and wood stoves), household products, construction materials, biologic agents (e.g., microbes, pets), off-gassing from water, and soil gas. In particular, soil gas is the origin of most indoor radon Prevention of environmental tobacco smoke Includes intervention such as to reduce exposure to ETS (smoking bans and restrictions), interventions to reduce tobacco-use initiation (increasing the unit price for tobacco products and multicomponent mass media campaigns), and interventions to increase cessation (increasing the unit price for tobacco products; multicomponent mass media campaigns; provider reminder systems; a combined provider reminder plus provider education with or without patient education program; multicomponent interventions including telephone support for persons who want to stop using tobacco; and reducing patient out-of-pocket costs for effective cessation therapies) SARAN ET AL.

| 41 of 44
Prevention of exposure to toxins such as lead, mercury and pesticides Includes all the strategies to protect children from existing toxins, such as lead, mercury, and tobacco such as the US EPA and FDA need more authority and resources to regulate and reduce emissions and exposures Safe places to play Creating safe places for children like playgrounds, where they are protected from violence, road accidents and other injuries Traffic calming Includes strategies as speed reducing initiatives near urban residential areas include speed cushions, humped pelican crossings, raised junctions, narrowing the road(s), gateways at the entrances to the area, build-outs to protect on-street parking spaces, and mini-roundabouts

Governance
Governance can be seen as the exercise of economic, political and administrative authority to manage a country's affairs at all levels. It comprises the mechanisms, processes and institutions, through which citizens and groups articulate their interests, exercise their legal rights, meet their obligations and mediate their differences Child rights It provides a set of principles and standards covering Children's entitlements to such essentials as education, healthcare and the right to be heard, as well as protection from abuses such as unjust treatment and exploitation It places an obligation on states to ensure that all children within their jurisdiction (including noncitizens such as refugees) enjoy these rights Legislative reforms Develop and adopt national legislation to implement child protection laws Child protection regulation Measures that protect access to resources, promote employment, and support the childcare role ANNEX 2: SEARCH STRATEGY

Search string/Key words (OVID platform)
Developing Country Free Text -(developing OR less-developed OR less* developed OR "under developed" OR underdeveloped OR under-developed OR middle-income OR "middle income" OR "low income" OR low-income OR underserved OR "under served" OR deprived or poor*) adj3 (countr* OR nation OR population OR world OR state OR economy OR economies).mp -("third world" OR L&MIC OR L&MIC OR LAMIC OR LDC OR LIC OR lami countr* OR transitional countr*).mp -(Africa OR "Sub-Saharan Africa" OR "North Africa" OR "West Africa" OR "East Africa" OR Algeria OR Angola OR Benin OR Botswana OR Burkina Faso OR Burundi OR Cameroon OR "Cape Verde" OR "Central African Republic" OR Chad OR "Democratic Republic of the Congo" OR "Republic of the Congo" OR Congo OR "Cote d'Ivoire" OR "Ivory Coast" OR Djibouti OR Egypt OR "Equatorial Guinea" OR Eritrea OR Ethiopia OR Gabon OR Gambia OR Ghana OR Guinea OR Guinea-Bissau OR Kenya OR Lesotho OR Liberia OR Libya OR Madagascar OR Malawi OR Mali OR Mauritania OR Morocco OR Mozambique OR Namibia OR Niger OR Nigeria OR Rwanda OR "Sao Tome" OR Principe OR Senegal OR "Sierra Leone" OR Somalia OR Somaliland OR "South Africa" OR "South Sudan" OR Sudan OR Swaziland OR Tanzania OR Togo OR Tunisia OR Uganda OR Zambia OR Zimbabwe).mp. -("South America" OR "Latin America" OR "Central America" OR Mexico OR Argentina OR Bolivia OR Brazil OR Chile OR Colombia OR Ecuador OR Guyana OR Paraguay OR Peru OR Suriname OR Uruguay OR Venezuela OR Belize OR "Costa Rica" OR "El Salvador" OR Guatemala OR Honduras OR Nicaragua OR Panama).mp -("Middle East" OR "South-East Asia" OR "Indian Ocean Island*" OR "South Asia" OR "Central Asia" OR Caucasus OR Afghanistan OR meta analysis/ -((systematic* adj2 review*) or "meta-analy*" or "meta analy*") Evidence map key words -Evidence maps evidence adj3 map -Evidence mapping -Evaluation map -Evaluation mapping -Systematic map -Systematic mapping -Descriptive mapping -Scientific uncertainties Child well-being key words -[(child abuse* OR child well-being OR child protect* service* OR foster care OR child* adj3 abus* OR schoolchild*, adolescen*, teen*)] Target age group key words -(young child* OR children OR pre-schooler* OR pre-schoolers*, kindergarten* OR kindergartener*, early child OR childhood, and early year OR years) -("adolescen*" OR juvenile OR minors OR youth OR "young adult" OR "young women" OR "young men" OR girl* OR boy* OR (school adj6 student*) OR teen* OR schoolgirl* OR schoolboy*)