Using Existing Platforms to Integrate and Coordinate Investments for Children: Summary of a Joint Workshop by the National Academies of Sciences, Engineering, and Medicine; Centre for Health Education and Health Promotion; and Wu Yee Sun College of the Chinese University of Hong Kong (2015)

2

Coordinated and Integrated Approaches to Investing in Young Children

In the first panel of the workshop, two speakers provided broad overviews of the issues involved in coordinating and integrating early childhood development policies and programs. While coordination and integration can confer advantages to children and their caregivers, they also can pose risks to the delivery of needed services. These potential advantages and disadvantages vary from one place to another, but their existence points to broader principles that extend across locales.

SCRUTINIZING ARGUMENTS FOR COORDINATION AND INTEGRATION

The consensus among early childhood development experts is that children develop holistically, observed Jan van Ravens, a senior policy maker and consultant affiliated with the Child Study Center at Yale University. This requires services or interventions from multiple disciplines located in multiple sectors. Typically these services or interventions are administered by multiple ministries of government, which raises the following question: What are the implications of the consensus regarding holistic childhood development for the organization, management, and governance of early childhood development programs and policies?

One argument is that synergy exists among such services and interventions. This argument hypothesizes, for example, that if nutrition services are integrated into a preschool program, a child will learn more and benefit more from that program. Evidence for such a conclusion typically

emerges from study designs that consider four groups of children: children who have access only to a nutrition program, children who have access only to a preschool program, a control group of children who do not have access to either intervention, and children who receive the integrated program. If the last group has the best outcomes, this is seen as evidence in favor of integrated programs.

However, this argument is not logically sound, observed van Ravens, unless a fifth group of children is considered: children with access to a preschool program from one provider and nutrition services from another. Children in such “combined” programs have access to multiple but not integrated services and interventions. “There is no evidence, to the best of my knowledge, that children in combined programs fare less well than children in integrated programs, because they are never compared in a research setting,” said van Ravens. Researchers have found synergy in integrated programs because that is where they have looked for it, not necessarily because it does not exist in combined programs. “It is a classic example of research bias.”

Combined programs are very common. Many children receive early childhood education, medical care, nutrition services, social support, and other services from multiple providers that are not integrated or necessarily even coordinated. Admittedly, a wide range of services is found most often in richer countries, said van Ravens. “But by the year 2030, or the end of the Sustainable Development Goals, I think there is a very good chance that many children, even in developing countries, will have access [to] combined programs if we work on that.”

Another argument for integration could be referred to as the attraction argument. It posits that the integration of two interventions in one program will attract families and children more than will single interventions. For example, conditional cash transfers or the provision of school meals could attract families and children to medical, parental education, or other programs, van Ravens observed.

However, the attraction argument often fails in practice, he continued. For example, a preexisting program may reach most or all of a target population, but when it is integrated with another program it reaches only part of the population. For example, a nutrition program, a parental education program, and a preschool program may each reach separate populations, but the populations may not overlap extensively, so an integrated program ends up serving fewer people. It also may be difficult or costly to integrate existing programs. And human resources can be a serious bottleneck in integrated programs. A relatively untrained person may be able to deliver some services but would be stymied in trying to deliver integrated services, and this problem is often worst in places with the greatest needs.

THE COSTS OF INTEGRATION

The number of countries with integrated early childhood development policies has grown rapidly in recent years, but results on the ground have often failed to materialize, van Ravens said (van Ravens, 2015). One reason is that the development and implementation of policies often suffer from delays and drawbacks even in single settings, and this problem is exacerbated in multisectoral settings. For example, one group may be opposed to some of the provisions of an integrated policy, which delays the entire policy. In addition, each ministry tends to have its own mechanisms for policy development, and these mechanisms can differ among agencies.

Another problem is that multisectoral policy development often does not fit into regular planning and budget cycles, van Ravens noted. As a result, it tends to take place in a political vacuum, which can frustrate endorsement and implementation. A multisectoral policy also can conflict with the policies of the line agencies, such as the education or health ministries, he added.

Finally, quantitative and financial analyses usually come after the finalization of the integrated policy, which can lead to funding gaps. To avoid such gaps, the costs and available funding must be considered from the start of policy development, van Ravens said.

The line ministries are often seen as the stakeholders who resist integration. But these ministries often have large budgets and a need to keep tight control over their operations. If things go wrong with the distribution of resources, the agency will be held responsible. If the streams of money are intertwined at various levels of the system, the tools for controlling those resources may be insufficient.

The ministries also need to maintain internal consistency, observed van Ravens. For example, health systems need to be able to refer patients from primary to secondary or tertiary health care, and school systems need to transition students from one level to the next. If these processes become subject to an external policy, the ministries may not be able to fulfill their responsibilities.

Finally, investing in children is not the only “horizontal constituency” that claims territory and budgets from the line ministries; others include poverty reduction, urban development, environmental protection, youth criminality, and older adults. Few policy issues exist entirely within single sectors today. Line ministries would disintegrate if they ceded power to all these other constituencies, van Ravens said. Instead, early childhood development needs to learn from other horizontal constituencies about the best ways of getting things done, he observed.

REALIZING THE BENEFITS OF COORDINATION

Van Ravens concluded with five recommendations that reflect the realities of policies and programs.

The first recommendation he made is to universalize essential child services. As the coverage of these services increases, they will increasingly overlap, and more and more children will have access to more and more services, which will increase synergy. However, the services should remain in the sectors, van Ravens said, to ensure progress. What matters is that the services come together in the child.

His second recommendation was to tie services together at the local level but only as needed. Sensible people, operating within a culture of cooperation, do not need a lot of formal coordination mechanisms, he said. Whenever spontaneous coordination appears to be insufficient, more formal coordination could be provided through local coordination bodies, sensitization, training, and perhaps an early childhood development coordinator. Traditional structures and local leadership should be used as much as possible, he added.

His third recommendation was to treat home-visiting programs that lie within the health sector as an exception. Early stimulation and psychosocial development need to be integrated in such programs, he said, but these programs should remain in the health sector.

His fourth recommendation was that services be “linkable” at the local level. Therefore, services designed at the national level must be flexible enough for local adaptation and for linking them with other services as needed, he said. Vertical coordination can ensure that central authorities design policies and programs in this manner. This will often mean funding on the basis of results (for example, the numbers of children reached), setting some general quality standards, and ensuring room to maneuver at the local level.

His fifth recommendation was that coordination at the central level should be “light.” Two or three interministerial meetings could be held to decide which child services are essential (this will not vary greatly between countries) and to agree on a roadmap for their universalization. Targets can be used to monitor progress through existing sectoral systems of monitoring and evaluation, while existing sectoral councils can also be used for stakeholder consultation.

COORDINATION AND INTEGRATION AT THE NATIONAL LEVEL

At the national level, early childhood development policies, service programs, strategic plans, and laws are usually multisectoral, observed Emily Vargas-Barón, director of the RISE (Reconstruction and International Security through Education) Institute. However, very few of these

early childhood development programs are integrated at the national level. Rather, services are sometimes integrated around children at the local level to ensure holistic child development.

To ensure effective early childhood development policy implementation, multisectoral coordination structures are usually developed, with the participation of governmental agencies at all levels, nongovernmental organizations, and other civil society and private-sector organizations. Coordination also occurs within sectors, which are often multidisciplinary in structure. Examples of multisectoral programs include parent education and support programs, early childhood intervention services, health and nutrition, and social protection. Even preschool programs tend to be multisectoral, Vargas-Barón observed, because a preschool program may draw on health services, nutrition services, social protection services, and so on, whereas establishing these services within an education system can be very expensive. “The better approach,” she said, “is to have a good collaboration between the ministry of education and the health and nutrition areas of a health ministry.”

The ministries usually involved in early childhood development policy planning include education, health, nutrition, sanitation, protection, justice, planning, finance, gender, rural development, and interior, although these ministries may combine or overlap in different ways according to the institutional cultures of countries. Sectoral ministries usually maintain the managerial and fiscal control of their programs, although they may coordinate their services, sometimes closely, with other ministries. For example, they often have formal or informal inter-agency agreements; they may conduct mutual referrals and create a common database and tracking system using existing databases to the extent possible; and they may conduct joint planning, monitoring, and evaluation, though this is usually the weakest area of program implementation in most countries.

Few intersectoral ministries are found, however. The few that exist are mainly health and protection ministries, such as those in Georgia and Lesotho. Very few health and education ministries exist. One exists today in Bhutan, and Brazil created a health and education ministry in 1943, but it lasted for less than 4 years before splitting. “They really are very separate sectors,” said Vargas-Barón.

In general, “integrated” child development ministries that bring together multiple sectors have been few and short lived and have tended to lack power. For example, Venezuela established a Ministerio de la Inteligencia in 1979 that lasted for only a few years, Sri Lanka established a Ministry of Child Development and Women’s Affairs, and Tanzania established a Ministry of Community Development, Gender, and Children. However, integrated ministries tend not to have served as good

platforms for early child development policies, said Vargas-Barón. “What some people have dreamed of and continue to dream of—an early child development ministry in each country—is just not going to happen, and when people have attempted to create such a ministry, it has tended to fail.”

The lead early childhood development ministries tend to be education, health, or protection. To ensure good intersectoral policy coordination and implementation, the lead ministry needs a multisectoral body plus a coordinating unit, said Vargas-Barón. This multisectoral body can be at the high level of ministers or deputy ministers, it can be a technical body, or both may be created. “It is essential to create some form of multisectoral body for collaboration and coordination with regular meetings.” Similarly, the coordinating unit can consist of a separate institute, as in Colombia, an office in the presidency or prime minister’s office, or be part of a particular ministry, as in Ghana.

Early childhood development policies need to specify the structure, roles, and responsibilities of entities for policy implementation and coordination at all levels, according to Vargas-Barón. However, decentralization issues abound. National roles and responsibilities often are not retained as funds are decentralized; funds for children’s programs may be lost if investment targets for early childhood are not specified for local governments; and the lack of normative regulations and of systems to ensure quality, equity, and accountability usually established at national levels can undermine children’s programs.

Vargas-Barón presented findings from a study on the evolution of early childhood development policies since 2000 that included five country case studies regarding short-term results from policy planning processes (Vargas-Barón, 2015a). She noted that participatory and comprehensive planning approaches, when used, have been successful in achieving many expected and unexpected results to expand and improve services for young children and their families.

COORDINATION AND INTEGRATION AT THE REGIONAL AND LOCAL LEVELS

At regional and local levels, the situation is somewhat different. At those levels, early childhood development structures, agencies, services, and activities may be sectoral, multisectoral, or integrated. At the local level, services tend to occur in a continuum from integrated to multisectoral to sectoral (see Figure 2-1). The points on this continuum vary, especially in terms of administration, fiscal support, supervision, monitoring and evaluation, and referral procedures.

Local entities and activities tend to be developed before regional

FIGURE 2-1 Early childhood development services at the local level occur along a continuum from integrated to sectoral.
NOTE: M&E = monitoring and evaluation.
SOURCE: Vargas-Barón, 2015b.

structures and entities are established, Vargas-Barón noted. By anticipating the development of regional offices for multisectoral coordination, collaboration, supervision, and support, nations can avoid having later problems at the regional level.

In Latin America and elsewhere, a number of countries have been able to take integrated and multisectoral programs that have been evaluated to be successful and scale them up to the national level. One study showed that, to be replicated on a larger scale, integrated programs require national legal recognition, financial support, and assistance with planning, coordination, supervision, and monitoring and evaluation (Vargas-Barón, 2009). Examples include the Colombian Institute for Family Welfare, the Integrated Early Childhood Development centers in Bosnia (see Table 2-1), and early childhood intervention systems in Australia, China, Europe, New Zealand, and the United States.

Formal agreements for multisectoral early childhood development activities help to ensure that roles and responsibilities are clearly defined and energetically pursued at each level. In contrast, informal agreements tend to fall apart as leadership changes and interpersonal work relations evolve over time. Job descriptions with multisectoral coordination roles, rewards, and other incentives for personnel who engage in these activities can be very helpful. The institutional culture can promote and reward multisectoral coordination and partnerships with civil society and private-sector entities, Vargas-Barón said.

TABLE 2-1 Early Childhood Development (ECD) Centers in Bosnia Extend from Preconception Services to Preschool and Family Support Services

SOURCE: Vargas-Barón, 2015b.

An example of multisectoral early childhood development services is the Convergence Zones that have been set up in Cameroon. This system offers separate but fully coordinated services for education, nutrition, health care, sanitation, and child protection. “If a family walks in to one of these service providers, they are automatically referred to all the others,” said Vargas-Barón. “Community members conducted an analysis of their own needs and service structures and identified what other services they wanted to develop. . . . This system, although it was multisectoral and not integrated, functioned as well as an integrated system in many respects.” However, she noted, having multiple sectors also means paying for multiple administrative structures, and this may not be cost-effective over time.

THE STATUS OF NATIONAL POLICIES AND PROGRAMS

Vargas-Barón briefly described a study that was released shortly after the workshop on the main drivers and challenges facing nations and regional governments as they develop early childhood development policy and instruments (Vargas-Barón, 2015a). The study describes the evolution and status of policy planning for early childhood development from before the year 2000 to 2014 and includes five country case studies on initial policy impacts.

Before 1999, only five countries had adopted early childhood development policies, with Colombia being a leader, beginning in 1968. By 2014, a total of 68 countries had done so, with another 10 countries reliably reported to have developed policies and 23 more reliably reported to be in progress. As a result, more than 100 countries will soon have early childhood development policies. The policy documents have “different quality and content, but the policies do exist.” Southern Asia and Southeast Asia have the highest percentage of countries with early childhood development policies, while Western Europe and North America have the lowest (see Table 2-2). The latter countries have tended to adopt a sectoral approach, including early childhood within education, health, nutrition, and protection policy instruments. This sectoral approach has its strengths, according to Vargas-Barón, “but there are always going to be problems in the so-called safety net. There will be some gap areas . . . where inadequate attention is given to young children and their needs.”

Among the challenges to policy planning for early childhood development, the study lists the following:

• Authoritarian regimes
• Lack of political will
• Rapid governmental leadership turnover

• Poor decentralization systems
• Extreme sectorality
• Inadequate attention to one or more of the key elements for developing comprehensive, scalable, and sustainable early childhood development systems
• Poor policy implementation structures and processes

In the five case study countries—Bosnia, Brazil, Colombia, Myanmar, and Rwanda—policy planning for early childhood development has led to a greater focus on improving and expanding services. “In these countries, they are already achieving more than they would have, most likely, without a policy.” The case studies indicate that, where highly participatory early childhood development policy planning methods were used, implementation often begins before policies are adopted, greater national ownership is achieved, and policies are being successfully implemented. Furthermore, if a strategic plan or an action plan is developed in addition to an early childhood development policy, the policy is more likely to be well implemented. Relevant laws and bylaws also can reinforce key policy dimensions. In addition, appropriate organizational structures at all levels are required to ensure good multisectoral coordination and implementation, Vargas-Barón said. Indicators, targets, cost studies, and budget projections and simulations also are essential, she added.

TABLE 2-2 Adoption of Early Childhood Development (ECD) Policy Instruments by Region

SOURCE: Vargas-Barón, 2015a.

Participatory processes in planning policies can make the difference between success and failure. These processes need to involve people from all levels of society, said Vargas-Barón, including ethnic and linguistic minorities and people living in urban and rural areas. “Tons of policies [are] gathering dust in ministries all over the world. These sectoral policies went nowhere because people sat in an office and wrote them. Participatory approaches . . . are ones where you involve all of the critical governmental, civil society, and private sectors that need to contribute to policy planning.”

Vargas-Barón drew attention to several issues involving equity that arose in the case studies. When and where high-risk, impoverished families were officially targeted, they were served more fully. It may be concluded that ethnic and linguistic minority groups must be explicitly identified as requiring special attention or they tend to be forgotten, she said. Mother tongues will be included in early childhood development services if the country officially and unequivocally states they will be used. If children with developmental delays, disabilities, malnutrition, and chronic illnesses are prioritized, it is more likely they will receive screenings, early childhood intervention services, and inclusive preschool education. Children living in remote rural areas, violence zones, and areas prone to natural disasters also need to be explicitly targeted if they are to be reached, Vargas-Barón said.

FUTURE INITIATIVES

Vargas-Barón concluded with several suggestions for future initiatives. One is to expand research on policy planning and implementation for early childhood development. In addition, more well-trained and experienced early childhood development policy advisors and policy analysts are urgently needed, she said, along with at least one graduate university program for training early childhood development policy planners and analysts from all world regions.

Vargas-Barón also called attention to the importance of nongovernmental organizations, which often become the purveyors of pre- and in-service monitoring, quality assurance, supervision, training, standards setting, and planning. “Nongovernmental organizations can help ensure that quality is adequate to enable effective service growth and improvement.”

Issues of national investment in early childhood development need to be emphasized more pointedly, Vargas-Barón said. “There has been too much reliance on international flows of funds to support early childhood development work. The percentages of line ministries’ budgets for early childhood development programs need to be greatly expanded. Funda-

mentally, it is up to country leaders to have the political will to ensure their children develop well.”

Finally, Vargas-Barón emphasized that national leaders must perceive that improved child development and well-coordinated early childhood development services are truly in the national interest.

SUMMARY OF THE PRESENTATIONS

Panel moderator Hiro Yoshikawa briefly summarized the main points of the presentations on “Coordinated and Integrated Approaches to Investing in Young Children” by van Ravens and Vargas-Barón. He pointed to six concepts that cut across both presentations:

1. The issue of population level assessment and screening such as universal assessment of children at birth with periodical needs assessment
2. The idea of task-shifting—that a workforce can have duties that cut across sectors
3. The idea of interdisciplinary teams or communication and training across workforces
4. Topics that are represented in specific curricula in early childhood development, such as single programs that can encompass expertise and topics in particular areas
5. The need for better coordination when financing from different sectors
6. The ways in which local governance can include budgets that cover multiple sectors of services

THE ROLE OF PARENTS

In response to a question about the role of parents in promoting the integration of services, Vargas-Barón observed that parents must always be included in participatory policy planning processes. In addition, the parental role in early childhood intervention services or the transition from preschool to primary school is extremely important. “Empowerment of the parent is at the core of what we are trying to do to help them improve their lives and develop their children well.”

Van Ravens agreed, adding that parents have a double role. They influence their own children’s lives and have an impact on the local community—for example, by building support for preschool, health programs, and so on. “So many community-based early childhood development centers would not be able to function without the direct support of parents. I’m talking about very simple things, like building the structure, or

building a playground, or cleaning.” Also, integration is unavoidable with programs involving parents, whether home visitation or group sessions.

The involvement of parents points to a bottleneck in workforce issues, van Ravens observed. For example, parent programs are most effective with leaders from local communities, but the levels of education and training in some communities may be low, making it difficult to find enough leaders.

COORDINATION AND INTEGRATION FROM A BROADER PERSPECTIVE

The presentations of van Ravens and Vargas-Barón led to a more general discussion of coordination and integration that threaded through the entire workshop. Kofi Marfo, founding director of the Institute of Human Development at Aga Khan University, pointed out that coordination and integration are not either/or issues, because they occur on a continuum and can exist simultaneously. Similarly, replacing coordinated structure with integrated ones will not necessarily yield improvements. If existing structures are torn down in favor of integration, the services provided by those structures can be lost.

Vargas-Barón agreed in the need for flexibility and local adaptations, adding “The capacity of cultures to organize themselves must be recognized as the starting point.” Leadership and decision making are provided by the people of those cultures. “The tendency to want to bring evidence-based approaches and ‘cookie cutter’ approaches from one culture and impose them on others—which has been done by a lot of people—is really wrong.” For example, new capacity is emerging in parts of the world, such as Africa and Asia, and this capacity can be a strength for each region and also a benefit to the rest of the world as other regions learn from what is going on in those regions.

As Yoshikawa pointed out, the challenge of establishing a continuum of care across health, education, social protection, and child protection is such a large task that continual change and flexibility are essential. But once a comprehensive set of standards and a framework are in place, implementation can proceed bit by bit in different areas.

Pia Britto, senior advisor to the Early Childhood Development Unit of UNICEF, pointed to two forces that are mobilizing investment in young children by government: one is a political process, and the other is technical. Sometimes one leads the other or the other way around. In the past, the field of early childhood development has largely focused on the technical process and has not sufficiently addressed the political process. “Both those levels of decision making are required to increase investments and bring the type of capital we need for sustainability,” Britto

said. “Part of our thinking has to be what is it that we are discussing here that makes it compelling for the political-level decision makers to look at this as something they can gravitate to.” In some countries, such as Chile or Colombia, political leaders have been champions of the field, but elsewhere the argument has not been well enough articulated. “We have not taken our technical knowledge that we are sharing here to that level of political prioritization. As a forum, we have to bring that now into our conversation.”

Sara Watson, national director of ReadyNation, a business membership organization that advocates for early childhood, observed that in the United States, business leaders have been a major political factor to win investments and, in particular, to push for the multisectoral approach. In the states of California and Massachusetts, for example, the state early childhood advisory body is chaired by a business person, not by someone from the early childhood world. Because business leaders do not come from a particular early childhood sector, they can push for all different sectors. “That has been very effective in the United States.”

Howard Sobel, a representative of the Western Pacific regional office of the World Health Organization (WHO), asked for examples of success with concrete outcomes in low- and low-middle-income countries in intervention efforts for early childhood development. As a brief response, Vargas-Barón mentioned Brazil’s success in bringing the private sector into leadership roles for effective implementation of integrated early childhood development interventions.

DATA AND POLITICS IN POLICIES

Discussion of integration and coordination also led to a more extended discussion of the roles that data and politics play in establishing and implementing policies and programs. Constanza Alarcón, the Colombian National Coordinator of the Intersectoral Committee for Early Childhood, noted that the data on which policies should be based can sometimes be “quite distant from day-to-day affairs.” Policy makers are faced by economic pressures, political pressures, historical factors, and financial issues, all of which can make it difficult to generate and use data to render decisions.

To ensure that decisions are based on evidence, policy makers need to be better prepared and have greater capacity to do their jobs, she said. Governments and teams change, and new teams tend to change existing approaches. “If we do not affect those who make decisions on a day-today basis that affect the lives of women and children, pregnant women, and so on, we cannot create a process of transformation or impact on the public policy-making process.”

Vargas-Barón advocated setting up a multisectoral coordination unit, such as the one in Colombia, that is subject to the political realities of a country and is receiving constant feedback on what is happening at all levels—local, regional, and national. The challenge is to work in the “wisest manner possible with the greatest input.”

Van Ravens observed that providing access to data can nevertheless be of great assistance in influencing policy. “In so many countries, it is a big challenge simply to find out how many children are being vaccinated, how many children are birth registered, how many teachers there are, how many nurses are there.” Sometimes data are manipulated because of perverse incentives—for example, to report more children in preschool to acquire additional funding—but in many countries great progress is being made. Peru, for example, has a child-by-child system of monitoring that is integrated across sectors. Similarly, data on returns on investments in children that are specific to a program and a country can help convince policy makers to support those investments.

Zulfiqar Bhutta, Robert Harding Inaugural Chair in Global Child Health at the Hospital for Sick Children in Toronto and founding director of the Centre of Excellence in Women and Child Health at Aga Khan University, pointed out that one of the biggest gains in the past decade has been the increase in quantitative data around health, and particularly around survival rates. Similarly, quantitative data on the impact of integrated, partially integrated, or nonintegrated programs could help support the momentum for change.

Along those lines, van Ravens noted that another way to gauge the effect of integration would be to compare integrated and nonintegrated programs to determine if integrated systems provide greater benefits than combined programs.

Regarding the political factors in policy development and implementation, he said that early childhood development experts need to understand the political landscape and how it works so they are able to make the right moves. “We need more Machiavelli in our work.”

This page intentionally left blank.