ADVANCING EARLY CHILDHOOD DEVELOPMENT IN TAJIKISTAN LESSONS FROM A REVIEW OF INTERNATIONAL EXPERIENCE IN EARLY CHILDHOOD DEVELOPMENT PROGRAMMING Part 1: Key Features of Early Childhood Development Programs Naomi Rupasinghe, Lorna Benton, Kate Mandeville, and Mutriba Latypova April 2021 i ADVANCING EARLY CHILDHOOD DEVELOPMENT IN TAJIKISTAN LESSONS FROM A REVIEW OF INTERNATIONAL EXPERIENCE IN EARLY CHILDHOOD DEVELOPMENT PROGRAMMING Part 1: Key Features of Early Childhood Development Programs Naomi Rupasinghe, Lorna Benton, Kate Mandeville, and Mutriba Latypova April 2021 ii © 2021 The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved This work is a product of the staff of The World Bank. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. The World Bank encourages dissemination of its knowledge, therefore, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. All queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. ACKNOWLEDGMENTS This report was written by Naomi Rupasinghe, Lorna Benton, Kate Mandeville, and Mutriba Latypova. The team is grateful for the peer review provided by Adelle Pushparatnam and Sarah Coll-Black, as well as the guidance of Marcelo Bortman, Jakub Kakietek, Tania Dmytraczenko, Kathleen Krackenberger, Sara Giannozzi, Saodat Bazarova, Veronica Silva Villalobos, Gabriel Francis, Ian Forde and Angela Mazimba. The team would also like to thank Shahlo Norova and Parvina Mahmadziyoeva for their excellent administrative support and guidance. i CONTENTS CHAPTER 1: Why is advancing early childhood development (ECD) a priority for Tajikistan? 1 CHAPTER 2: What can be learnt from a review of international experience in ECD programs about the key features of an ECD program? 4 CHAPTER 3: How did we conduct this review? 9 CHAPTER 4: Considering age groups when designing an ECD program 12 CHAPTER 5: Considering target participants when designing an ECD program 14 CHAPTER 6: Considering settings when designing an ECD program 17 CHAPTER 7: Considering service providers when designing an ECD program 21 CHAPTER 8: Considering timing (program length and intervention frequency) when designing an ECD program 29 REFERENCES 33 ANNEX 1: Summary of studies 35 ii 1 CHAPTER 1: WHY IS ADVANCING EARLY CHILDHOOD DEVELOPMENT (ECD) A PRIORITY FOR TAJIKISTAN? KEY POINTS • A child born in Tajikistan today will be 53 percent as productive when she grows up as she could be if she enjoyed a complete cycle of education and access to high-quality health services. • Tajikistan has the highest birth rate, of 29 births per 1,000, in the Europe and Central Asia (ECA) region and children under 6 years old comprise 17 percent of the population. Early Childhood Development (ECD) investments are some of the most cost-effective interventions and given the demographic context, Tajikistan can benefit more than other countries by building its human capital and empowering its workforce of the future. Globally, investing in children to build human capital is one of the best investments a country can make to eliminate extreme poverty, boost shared prosperity, and build the workforce of tomorrow. Our experiences in early childhood have a profound and lasting impact on brain development, determining our ability to learn, be healthy, work effectively, and communicate with others. Globally, one third of children currently do not obtain adequate nutrition, stimulation, and protection to reach their developmental potential (Center on the Developing Child 2016), reflecting the need for increased investment. Tajikistan’s Human Capital Index (HCI) Score of 0.50 is lower than average for the region, reflecting that much more can be done to support children in achieving their potential. Tajikistan’s score in the 2020 HCI is significantly below the ECA average of 0.71, and regional comparators such as Armenia (0.58), the Kyrgyz Republic (0.60), and Kazakhstan (0.63). High levels of stunting, with almost 1 in 5 children in Tajikistan suffering from stunting, is one of the main drivers of the low HCI score. The role of education in driving human capital is a critical factor. While Tajikistan has made progress in recent decades, several key indicators (shown overleaf in Table 1) across health, nutrition, and education indicate the scope for future progress and point to unmet needs. The findings of a nationally representative survey of children from 0 to 6 years old conducted in 2019 using the Early Human Capability Index (eHCI)1 found that the overall child development outcomes score is low at 0.54 on a scale of 0 to 1; however, scores were particularly poor in some 1 The Early Human Capability Index (eHCI) is a holistic measure intended to capture early child development across diverse cultures and contexts. Further details can be found here: https://ehci.telethonkids.org.au/. The Human Capital Index (HCI) and quantifies the contribution of health and education to the productivity of the next generation of workers. Further details can be found here: https://www.worldbank.org/en/publication/ human-capital 2 of the more formal or academic domains of development like pre-literacy and pre-numeracy, likely in part due to the low preschool attendance rates. While child development scores improve as children get older, even children aged 6 years old still only scored 0.68 overall. TABLE 1: ECD outcomes in Tajikistan Maternal and child health outcomes Infant mortality rate (deaths per 1,000 live births) 30 Under-5 mortality rate (deaths per 1,000 live births) 34 Maternal mortality ratio (deaths per 100,000 live births) 33 Low birth weight (<2,500g) 6 Maternal and child health nutritional outcomes Underweight (% children 0–59 months*) 7.6 Stunting (% children 0–59 months) 17.5 Wasting (% children 0–59 months) 8.9 Overweight (% children 0–59 months) 5.6 Underweight/Thinness (BMI<18.5 kg/m2), % amongst WRA 7 Overweight (BMI 25.0–29.9 kg/m2), % amongst WRA 24.3 Obesity (BMI >30 kg/m2), % amongst WRA 13 Child development outcomes Physical 0.78 Communication 0.78 Cultural 0.72 Socio-emotional 0.68 Perseverance 0.43 Approaches 0.65 Numeracy 0.47 Reading 0.14 Writing 0.24 Overall 0.54 Child development score for males 0.53 Child development score for females 0.56 Child development score for age 2 0.39 Child development score for age 3 0.50 Child development score for age 4 0.57 Child development score for age 5 0.63 Child development score for age 6 0.68 Child development score in Dushanbe 0.56 Child development score in Sogd 0.60 Child development score in Khatlon 0.51 Child development score in Region of Republican Subordination 0.53 Child development score, in Gorno-Badakhshan Autonomous Province 0.54 Sources: • Statistical Agency under the President of the Republic of Tajikistan. 2018. Tajikistan Demographic and Health Survey 2017. https:// www.dhsprogram.com/pubs/pdf/FR341/FR341.pdf. • Child development outcomes are based on a World Bank commissioned study led by Sally Brinkman using the eHCI. Note: BMI = Body Mass Index; WRA = Women of Reproductive Age. 3 With one of the fastest growing populations and the highest birth rates in ECA, ECD investments are a development opportunity. Globally, Tajikistan is among the top 25 percent of the fastest growing populations. Children under 6 years of age comprise 17 percent of the population, and today 1 out of 3 Tajiks are under 15 years of age. At 29 births per 1,000 people in 2016, Tajikistan has the highest birth rate in the ECA region. Between 2015 and 2025, the number of children aged 0–9 is expected to rise by nearly 23 percent, from about 2.2 million to 2.7 million. This demographic context presents a unique opportunity for Tajikistan to invest in its youngest population as a means of changing its growth and development trajectory. Investing in ECD is also critical to achieving the Sustainable Development Goals (SDGs) and mitigating losses in gross domestic product (GDP), and is a first, indispensable step in building human capital. A rigorous review of scientific evidence, the Lancet series on ECD, has highlighted the role ECD investments play on the broader SDG agenda (L. M. Richter et al. 2017). Missing opportunities to support child development has been associated with a 26 percent average reduction in adult income, on an annual basis (L. Richter et al. 2019). L. Richter et. al (2019) note that, assuming 125 million children are born each year with a global average of poor infant growth, the estimated annual global income loss is US$17 billion (Fink et al. 2016; L. Richter et al. 2019). This stands to negatively impact economic growth. The impact of stunting alone is estimated to be 7 percent of GDP, and this only constitutes one area of development delay (Galasso and Wagstaff 2019). Given the high economic and social returns to quality ECD investments, interventions to improve child health, access and quality of education, and cognitive and psychosocial development are vital to reap the benefits of this demographic shift and sustain growth through a productive workforce that can respond to a changing global economy (World Bank 2019). 4 CHAPTER 2: WHAT CAN BE LEARNT FROM A REVIEW OF INTERNATIONAL EXPERIENCE IN ECD PROGRAMS ABOUT THE KEY FEATURES OF AN ECD PROGRAM? KEY TAKEAWAYS • ECD programs can include nutrition, social protection, health, education, and water and sanitation components but, overall, the most successful and effective programs seek to bring together many sectors and support children and families in accessing a range of services. • From our review of 14 programs across 10 countries we identified key differences in age, target participants, settings, frequency and length, and service providers. The purpose of this report is to share these findings so that policy makers are better positioned to make key choices and design effective ECD programs. This report is intended to support policy makers in understanding the key features of an ECD program based on a rapid review of international ECD programs. Given the range of sectors and systems that are relevant to ECD, policy makers face a key challenge in determining how to maximize the benefits of interventions. By drawing on a descriptive analysis of features of previous programs, this report seeks to support policy makers and ministries in understanding how to build the key components of a longer-term and sustainable approach to ECD programming. This report presents findings from a scoping review of 17 published studies reporting on 14 ECD programs across 10 countries; at the start of each section we have identified key takeaways for Tajikistan. Figure 1 overleaf provides an illustration of the global reach of programs reviewed in this report (the following chapter provides details on how these programs were selected) and Table 2 provides a list of the programs. This set of programs drew from 17 studies, including: 4 studies from Bangladesh, 3 studies from Colombia, 3 studies from Pakistan, 2 studies from Kenya, 1 study each from Mexico, Malawi, China and Australia and 1 covering both Brazil and Zimbabwe. The search was not limited by geography or region and identified studies from diverse geographical settings, including 9 rural studies, 1 urban study, and 7 studies from mixed settings. A feature of successful ECD programs is their ability to facilitate a multisectoral approach across the life course. Programs can include the sectors outlined in Figure 2: nutrition, social protection, health, education, and water and 5 FIGURE 1: Geographical coverage of 17 peer-reviewed studies identified in a systematic search for this review sanitation. Black et. al. (2017) have highlighted the importance of adopting a life course approach in improving the quality of ECD programs and in realizing the economic benefits of ECD investments. The life course approach is an effective way of understanding how to maximize a multisectoral approach. As programs often straddle multiple sectors and can be challenging to design, a life course approach can often be helpful as a guiding framework. Moreover, the lifecycle approach can help to get to the question of what works or what services are needed from the perspective of the beneficiary and not a particular ministry of agency, allowing a government to move beyond silos. 6 Table 2: Program titles and countries Program Countries WASH Benefits trial Bangladesh A 10-month parenting program (Save the Children) Bangladesh Alive and Thrive Program Bangladesh PLA cycle with women’s groups Bangladesh GDMP Colombia Familias en Acción Colombia Hogares Comunitarios de Bienestar Colombia Action Against Hunger - cash transfer programs Pakistan Lady Health Worker Program Pakistan Oportunidades (now Prospera) Mexico CCD Malawi IECD China Reach Up Brazil and Zimbabwe Right@Home Program Australia Note: CCD = Care for Child Development; GDMP = Growth and Development Monitoring Program; IECD = Integrated Early Childhood Development; PLA = Participatory Learning and Action; WASH = Water, Sanitation and Hygiene. FIGURE 2: A range of sectors can be involved in ECD programs Nutrition Social Health Education Water and Protection Sanitation In this context, the life course conceptual framework refers to the importance of sustained intervention at several life points. This includes preconception and pregnancy, parenting and caregiving, newborn, early childhood, mid- childhood and adolescence, and adulthood. It builds on the nurturing care framework, highlighting the cross-sectoral importance of health, nutrition, security, caregiving, and learning interventions, as well as the broader enabling environment and socioeconomic and cultural context. The life course approach highlights the interface between protective and risk factors that influence health and other socioeconomic outcomes over a lifetime. It takes into account the way in which family planning, preconception nutrition, and maternal health lay the foundation for healthy pregnancies. It can also be used to guide implementation by assessing a given context. For example, by identifying a sector that has built strong links with parents and families and focuses on defining a package of services and developing referrals and links to services from that sector. 7 Across the life course, the targeting that ECD programs focus on can be viewed in terms of service type and specific participants. In terms of services, one approach is to consider how three different types of services are provided for populations: basic, differentiated, and specialized services (The World Bank 2018). Successful programming entails moving toward effective coverage of all three service types. The definitions of these services are provided in Figure 3 and different approaches have been taken to the provision of these services. Ideally, however, a successful ECD approach entails being able to provide universal coverage of basic services. This tier of services considers the principle of equality, such that everyone should have access to the services that help lay a strong foundation. It also entails strong provision of the second two tiers of services. These draw more heavily on the idea of equity—that those with differential needs require distinctive services to be successful. Children and families from poorer or disadvantaged backgrounds may need extra help to access basic services. Finally, the successful provision of basic and differentiated services is likely to identify children whose development has been delayed or who have specialized needs. Here, the ethical consideration to provide for these children prompts the need for ensuring the provision of specialized services. FIGURE 3: Definitions of different types of services Specialized services, for those with specific developmental requirements e.g. disability or mental health needs Differentiated services, which provide support for families in recognition of their different needs, socio-economic status and challenges in accessing services Basic services that are available to all children. This might include immunizations, growth and monitoring check-ups, antenatal care for mothers and access to clean water. Source: World Bank 2018. Our review identified a range of design options for policy makers across five areas of program design. We observed a range in terms of the following five areas: age of children, targeting (by type of service and beneficiary), settings, timing (program length and intervention frequency), and service providers. Within each of these areas policy makers face decisions about how to build and develop programming that responds to a given context and the needs of children. A summary of key considerations is provided below in Figure 4. A detailed description of all the studies and programs is provided in Annex 1. 8 Ideally, the end-goal of an ECD program is to develop integrated systems that can monitor and support the development of a child from birth onto adulthood, ensuring that the child and their family can access the services they need to thrive. However, in resource-constrained settings, the path for achieving this vision can often necessitate identifying starting points for engagement and for building effective collaboration between siloed systems that may not be familiar with each other. This presents the policy maker with sequencing decision and trade-offs. While these were not the scope of the inquiry, where possible guidance on the sequencing and cross-sectoral interactions has been noted. FIGURE 4. A summary of the key design features of an ECD program Covering the ages of 0-3 is the backbone of an ECD program and provides important first step for policy makers in Tajikistan. While the age Age range targeted in the studies included in the review was 0-10, in resource constrained settings, a focus on 0-3 is a critical starting point Ensuring that all children have access a basic package of ECD services is an important starting point. Targeting focuses on how services are provided and to whom they are provided, and three tiers of services are helpful to consider. Basic services, which are needed for all children. Differentiated services, which help to target those with differentiated needs, and specialized services Target for those with more complex needs. The review identified that programs participants draw on a range of options, they can focus on mothers, father, communities, children of varying ages. In the case of low-income or resource constrained settings, targeting families, primary care providers and those that can support children aged 0-3 can be helpful starting point, although this must be balanced against context-specific needs. Many settings can be used to achieved ECD objectives. Fully assessing the range of services available can help to optimize the allocation of resources and build effective systems. For example, considering the role of schools, Settings hospitals, community settings, centers and households is important, before making decisions about where to focus efforts, as this can help to avoid overlooking cost-effective settings. Creating an army for ECD can be done by considering and engaging a range of providers. Across different contexts and settings, various Service stakeholders can be mobilized for the benefit of ECD provision. This includes providers highly specialized staff in higher resource settings and community members and groups in lower resource settings. Frequency and follow up make all the difference. Providing short but Timing regular interventions stands to be more successful in maximizing child development. 9 CHAPTER 3: HOW DID WE CONDUCT THIS REVIEW? KEY TAKEAWAYS • We used a two-stage scoping review to provide policy makers with a practical overview. This was intended to provide guidance on the key characteristics of an ECD program design and drew on an umbrella review and a systematic search. • The study did not seek to be exhaustive, opting to focus on practical insights. Therefore, there are important limitations, as systematic steps assessing risk of bias and providing critical appraisal were not conducted, leading to a variation in the quality of included studies. In the first stage grey literature publications were reviewed and the search criteria were tested. A range of ECD landmark series publications (World Health Organization 2015; Center on the Developing Child 2016; Britto et al. 2017; L. M. Richter et al. 2017; Daelmans et al. 2015; WaterAid, SHARE, and Healthy Start 2016; UNICEF 2017; Perkins et al. 2017; Banerjee et al. 2019; Boggs et al. 2019; Milner, Bhopal, et al. 2019; Milner, Bernal Salazar, et al. 2019; Arregoces et al. 2019) were identified. In addition, several systematic reviews focusing on different program participants and testing different ECD components also informed our approach (Baudry et al. 2017; Grantham- Mcgregor et al. 2014; Rayce et al. 2017; Smith et al. 2018). A broad search of the literature informed thematic areas of interest for this review and structured the inclusion and exclusion criteria for step two. Sources include the World Bank, Google Scholar, and the United Nations Children’s Fund (UNICEF) evaluation database. A second stage, which involved a systematic search of peer-reviewed literature, was conducted to scope the strength of existing evidence and to capture any major studies that were not included in the latest available systematic review. We identified publications that evaluated ECD interventions with both an education and a nutrition component (Box 1), published since February 1, 2013, in continuation of the work of Grantham-McGregor et al. Sources include Medline and ProQuest (Agriculture Science Database, Education Database, Engineering Database, Psychology Database, Public Health Database). Inclusion and exclusion criteria are presented in Box 2. 10 BOX 1: Search terms and inclusion Education - ECD OR parenting OR preschool OR preprimary OR early learning OR stimulation OR educational intervention AND evaluation OR assessment OR early childhood education OR early childhood care Nutrition - Nutrition OR breast feeding promotion OR responsive feeding OR micronutrient OR macronutrient OR nutritional supplementation Mental development - Child development OR cognition OR language OR behavior OR behavior OR physiology OR socio-emotional development OR motor development Health - Health OR nutrition OR micronutrient OR malaria OR integrated OR morbidity OR worms OR human immunodeficiency virus (HIV) Life course elements - Mental health OR depression OR sanitation OR hygiene OR poverty OR (cash transfer OR WASH adapted from Grantham-Mcgregor et al. (2014)) BOX 2: Inclusion and exclusion criteria for a systematic approach to reviewing ECD studies published since 2013 Inclusion Exclusion • English language only • Evaluating specific tools or individual • Published after February 1, 2013 components rather than a program itself • Contains both nutrition and education • Focus on a specific vitamin or supplement components AND/OR is one of our ECD intervention program types across the life course • Focus on obesity or physical activity • Evaluates the efficacy or the programs implementation of the program • Presents data on evaluation tools in an ECD setting • Assess factors affecting ECD implementation (that is, ECD quality) The approach had important limitations, as it did not seek to be exhaustive but intended to focus on practical insights. The requirements of a scoping/ umbrella review do not include systematic steps such as critical appraisal or risk of bias assessments that are usually found in a systematic review, leading to variation in the quality of included studies. Furthermore, it can be difficult to establish boundaries within a broad scope on a discipline as large as ECD and related fields such as Early Childhood Education and Care (ECEC). There can also be variation in the terminology and definitions used between study areas. Search terms may have excluded some topics, if they did not include both nutrition and education components (hence social protection or cash transfer studies may be underrepresented in this discussion). Finally, date and language limitations may have excluded studies published in a 11 non-English language or before 2013. To extend this review further, a more systematic review of empirical studies could be undertaken to evaluate the current evidence for integration and implementation of high-quality ECD ‘life course’ programs. A more systematic search may have identified empirical research that could hold further evidence on the efficacy of programmatic components and implementation lessons. 12 CHAPTER 4: CONSIDERING AGE GROUPS WHEN DESIGNING AN ECD PROGRAM KEY TAKEAWAYS FOR TAJIKISTAN • Ages 0–3 are the backbone of an ECD program as they cover the critical periods of cognitive, motor, socio-emotional, and linguistic development. An immediate focus on programming to support maternal and child health in these stages is priority for Tajikistan. Once these age groups are well covered, expanding to other groups would be an important next step. • Many programs were missing the opportunity to maximize their approach by adopting a life course approach. Taking this approach into consideration will help build a sustainable and step-by-step path to a comprehensive ECD offering, which focuses on adolescents and pregnant mothers, healthy babies, and thriving school children who are well equipped for the world of work to come. Covering the ages of 0–3 is the backbone of an ECD program. Previous reviews and this review, as shown in Table 3, have noted the scope for ECD interventions to target a range of age groups, across the life course, with a focus on pregnant mothers and children up to 8–10 years of age. However, early intervention has been identified as the optimal period, with the nurturing care framework targeting years 0–3 (World Health Organisation 2018). The first three years provide a critical window in which up to 80 percent of brain development occurs (Center on the Developing Child 2016). This suggests that countries without programming for ages 0–3 would benefit from focusing there; however, for those with well-developed programming for that age group, other age groups may be more relevant. If ages 0–3 are well-covered, expansion to other age groups may be the next priority. National ECD initiatives often make a common division around three years of age, depending on the primary school starting age in a country. This is because a distinction is made between pre-primary services and early childhood education. However, in Colombia home-based care is provided until 6 years of age and in the Russian Federation, Poland, and Estonia children remain in preschool until starting primary school at 7 years old. While the start of primary school can offer a natural upper threshold for an ECD program, segmenting interventions at the boundary between preschool and early childhood education can compromise the opportunity to build a more comprehensive and integrated approach. Well-recognized programs such as Colombia’s GDMP spans birth to age 10 and is anchored within the 13 health system. Similarly, in Denmark and Estonia the focus is from birth to primary school age. As a result, these models have led the way in maximizing opportunities to facilitate child development, given the recognition that the first 1,000 days are critical, and a life course approach is the foundation to more successful programming. The studies identified in our scoping review, as shown in Table 3, primarily focused on children aged 0 to 5, indicating that the current state of ECD programming often misses the opportunity to adopt a life course approach. Overall, programs in the scoping review reflect a continuum of ECD interventions from birth to 8 years of age. This includes evidence from the WASH Benefits trial series in Bangladesh and Kenya, Familias en Acción conditional cash transfer program (12–24 months), Action Against Hunger Cash transfer program (6 months to 4 years), Alive and Thrive (0–4 years). This finding is significant in two respects: firstly, that the continuum stretched beyond the dichotomy of preschool and school; secondly, that it demonstrates the lack of a life course approach—from adolescence and conception through to age 10. This reflects a significant missed opportunity for countries to build the foundation for greater human capital accumulation and healthier, happier children. It also makes programs such as Colombia's stand out, given the breadth of their coverage. TABLE 3: Summary of age ranges covered by programs Program/ Study Age range WASH Benefits trial Pregnancy–24 months A 10-month parenting program (Save the Children) 0–24 months Alive and Thrive Program 0–47.9 months Women were eligible to become members of the PLA cycle with women’s groups women’s groups if they were 15–49 years of age and resided in the intervention areas. GDMP 1–120 months Familias en Acción conditional cash transfer 12–24 months program Hogares Comunitarios de Bienestar 0–72 months Action Against Hunger - cash transfer program 6–48 months Lady Health Worker Program 0–24 months Oportunidades (now Prospera) 24–72 months CCD 0–24 months IECD in rural China 0–36 months Reach Up 6–48 months Right@Home Program Pregnancy–24 months 14 CHAPTER 5: CONSIDERING TARGET PARTICIPANTS WHEN DESIGNING AN ECD PROGRAM KEY TAKEAWAYS FOR TAJIKISTAN • Targeting can be considered in terms of the service type and the specific beneficiaries or participants. One approach is to consider how three different types of services are provided for populations: basic, differentiated, and specialized services. • Building a successful ECD offering within Tajikistan will entail considering how to move toward effective coverage of all three service types. • Child Growth and Developmental Monitoring Programs can build universal coverage so that health, education, and social protection systems have a foundation that is able to reach all children, and then guide referrals toward differentiated and specialized services. Mothers, pregnant women, children, households, families, fathers and communities were the target of programs reviewed in the study. A range of participants were the target of programing reviewed in the study, indicating the range of options available to governments and policy makers. For example, in Mexico’s Oportunidades Program vulnerable households were the focus of the interventions. It has an overall goal of strengthening universal coverage of basic services, but was seeking to focus on a marginalized community. One of the goals of the conditional cash transfer was to contribute to the improvement of the volume, quality and diversity of food consumed by vulnerable households. TABLE 4: Summary of populations targeted by programs Program/ Study Population targeted Households were eligible for participation if: there was a woman in her second or third trimester of pregnancy; the woman planned to reside in WASH Benefits trial the community for at least two years; the woman could speak the local language or English. For year two assessments, children with disability in hearing, seeing, motor function, or other physical disabilities were excluded. A 10-month parenting Target beneficiaries were mothers of children <2 years in rural Bangladesh program (Save the Children) Alive and Thrive Program Pregnant women and mothers of children aged 2 years. PLA cycle with women’s Women were eligible to become members of the women’s groups if they groups were 15–49 years of age and resided in the intervention. GDMP The GDMP aims for universal coverage among children, nationally. 15 TABLE 4: Summary of populations targeted by programs (cont.) Program/ Study Population targeted Familias en Acción Children aged 12–24 months and their primary caregivers from 96 conditional cash transfer municipalities in Colombia, located across 8 of its 32 departments (where program incidence of anemia is high) Hogares Comunitarios de Low-income children under the age of 6 Bienestar Action Against Hunger - Poor and very poor households, with one or more children aged 6 - 48 cash transfer program months Community-based home visits to mother–child dyads living in a rural district Lady Health Worker of Pakistan (infant aged up to 2.5 months), followed at ages 12, 24, and 48 Program months Oportunidades (now Vulnerable households with children in extreme poverty in Mexico Prospera) (geographically targeted and means tested) CCD Caregiver-children dyads, with children <2 years (30 rural and 30 urban) IECD in rural China Families with children <2 years Reach Up Mother-child dyads Pregnant women facing adversity were enrolled into the study and visited Right@Home Program for the first two years of a child’s life. The Colombian GDMP’s ‘universal approach’ had the greatest impact in the poorest regions, reflecting the importance of basic and differentiated service provision. The GDMP was created in 2000 to provide comprehensive early childhood care, including anthropometric assessment, evaluating motor skills, hearing and personal-social language, and providing education to parents about balanced nutrition, immunization, early stimulation, accident prevention, oral hygiene, and acute diarrhea (Carrillo et al. 2015). It focused on young children up to the age of 10 and their families. The GDMP outlines a schedule for health assessment, vaccination, and iron supplementation until 84 months. Carrillo et al. (2015) found a universal positive impact of vaccination and the prevention of acute malnutrition in less prosperous areas of the country. Further assessment indicates that providing education on infant-care practices has the greatest impact in the poorest regions, where rates of malnutrition are higher, and the level of maternal education is low. This demonstrates that seeking a universal approach or universal coverage of basic services, can require drawing upon differentiated approaches to provide additional information and support to better engage poor populations that might otherwise be left out. Cash transfers have been used to improve the provision of differentiated services because they can be designed to target the most economically vulnerable households. Arriagada et al. have identified cash transfers as an effective way of supporting ECD outcomes by better enabling households to meet nutritional needs. For example, Pakistan’s Action Against Hunger 16 cash transfer program (Fenn et al. 2017) or Colombia’s Familias en Acción (Attanasio et al. 2014), which targets the poorest 20 percent of the country, were able to help prevent stunting and wasting in children up to six months of age. A key finding from Fenn et. al’s study, which evaluated the effect of different monetary values, was that eh financial value made a difference. The larger the cash transfer, the greater the effect on wasting. Providing instruction on early childhood care also had significant impact in the lowest- income regions where maternal education and nutrition rates are poorer. Moreover, in both Bangladesh and Colombia focusing interventions on young children and providing instructions to parents were also seen to be effective in responding to differentiated needs. With regards to specialized services, areas such as language development have seen more success, unlike maternal mental health, where less is known about successful implementation. For children with special language or education needs, intervention packages have been successful in improving language outcomes. While these children will be included in universal approaches, children with disabilities are also excluded from some trial designs, particularly for water and sanitation programming. Nevertheless, CCD intervention packages, which specifically target children with disabilities, have successfully improved language outcomes. Maternal mental health has been difficult to effectively target in a multifaceted intervention and few identified studies discussed implementation lessons in this context (Singla, Kumbakumba, and Aboud 2015). Gladstone et al. (2018) did not find any significant impact on maternal mental health outcomes, which the authors attribute to the intervention components not being targeted enough. Integrated provision of ECD services in low-income settings remains an area that could benefit from further evidence and implementation science analysis so that systems are better positioned to track and strategically provide services to children in need. Once systems have developed the provision of each type of service, the critical next step is ensuring that each child can be tracked and that the required service provision and referrals can be made to support a child. For example, if a speech delay is detected in the health system, it is important that the appropriate referrals are made to speech, language, and education specialists. This entails not only the integration of services but the ability to identify and channel children around an integrated network. Examples of how this has been achieved in higher- income settings demonstrate the importance of building these systems, but much less is known about how to build toward this goal in low-income settings, where service delivery is often much more fragmented. 17 CHAPTER 6: CONSIDERING SETTINGS WHEN DESIGNING AN ECD PROGRAM KEY TAKEAWAYS FOR TAJIKISTAN • ECD models can draw on many different settings, including primary healthcare centers, community groups settings, schools, households, crèches and kindergartens, and community groups. • Demonstrations, provided during home visits and in community group settings, are an important avenue to provide children and caregivers with the opportunity to practice using new materials and actively absorbing important information on ECD practices. • In the interest of reducing disparities, education policies that are targeted at specific at-risk groups (especially the youngest learners) can prove useful in resource-constrained environments. • In low-income settings, the manner in which ECD interventions are sequenced can have a significant impact on efficacy and sustainability. • Aligning intervention scale-up to government services can be useful for reaching children/targets at geographic distance. Given the multi-dimensional nature of ECD investments it is possible to draw on a range of settings to deliver services and different settings are more appropriate for different services. Figure 5 documents a range of settings. These include community-based preschool or centers-based childcare (crèches, kindergartens, nursery schools, nursery or kindergarten classes in primary schools, kids clubs, preschools, day care centers, and integrated centers), home-based childcare, supplementary feeding program, home visiting, and health program. Settings, as shown in Figure 5, range from home based to more formalized (center and healthcare based) and vary in terms of opening hours, from long days to half days. An ECD program may also be delivered through several partnership channels, including government, community, private sector, and civil society. 18 FIGURE 5: Settings used in the provision of ECD services Creche / Primary care Community pre-primary Households (health centers) group settings school / kindergarten Community Schools group & households, combined Existing evidence on program impact notes the value of home visits and community groups as good opportunities for delivering a parenting program and supplementary nutrition (World Bank 2017) and can be used to support the provision of differentiated and specialized services. Home visits and community groups are particularly useful for parenting programs that utilize demos, problem solving, and for caregivers and for children to practice together with new materials. They also allow for an integrated approach, visiting parents of children at appropriate ages and stages up to 6–8 years old. In terms of types of service provision. Community groups can be helpful in the design of differentiated programs and they can be targeted to communities with specific socio-economist needs. For example, dietary diversity may be a challenge at the community level and a community group that provides education and information on low-cost, healthy and diverse meals can provide a population with differentiated needs with curated information, which builds on general guidance on the importance of nutrition. 19 TABLE 5: Summary of settings, by program Program/Study Settings WASH Benefits trial Household visits A 10-month parenting program (Save Community group meetings and home visits for mothers the Children) 1. IPC, via household visits. Alive and Thrive Program 2. MM 3. CM PLA cycle with women’s groups Community women’s groups (CM) GDMP Health Center Familias en Acción conditional cash Community/household visits transfer program Hogares Comunitarios de Bienestar Home-based childcare Action Against Hunger - cash transfer Cash transfer program Lady Health Worker Program Community-based home visits and parenting group meetings Oportunidades (now Prospera) Educational grants, primary healthcare, and cash transfer CCD Community group meetings and household sessions IECD in rural China Clinic, center, and outreach based Reach Up Home visiting Right@Home Program Nurse-led home visiting program Note: CM = Community Mobilization; IPC = Intensified Interpersonal Counseling; MM = Mass Media. BOX 3: Colombia’s home-based ECEC centers Idea in brief: Traditional community-based welfare homes are led by a communitarian mother (MC), who provides childcare for up to 15 children, aged between 6 months and 5 years, in the home setting. Services provided: Childcare, food supplements, monitoring of nutritional status, psychosocial development activities, health promotion and health prevention, parental stimulation guidance. Financial support: Training and loans were provided to MCs to upgrade facilities in their home. Impact: After at least 15 months of exposure to the program, children scored higher on tests of cognitive development and socio-emotional skills compared to similar children who entered the program later. Source: World Bank 2017. Kindergartens, crèches, and preschool services can be delivered in various formats and can receive mixed public or private funding, and anecdotal information provides some insight into the variation. As noted in Box 3, Colombia's home-based ECEC centers provide a range of services in home- based settings. Preschool education centers and full-day kindergartens in Russia are funded through state and private sources but are unique in their range of specialisms, including child development, the correction of abnormalities in children’s development (compensatory kindergartens), 20 sanitation, hygiene, and health improvement. Some Polish kindergartens have integrated units or specialist services for children with disabilities and are either publicly or privately funded. Chile’s ECEC kindergartens (Jardines Infantiles [JUNJI]) provide a good example of publicly funded services, free at the point of care, which are targeted at children from low-income families. JUNJI is the main institution for a range of tailored kindergartens approaches that include kindergartens via fund transfer, community kindergartens, and private kindergartens with delegated administration. They serve children from birth to 3 years of age and operate for 11 hours a day. 21 CHAPTER 7: CONSIDERING SERVICE PROVIDERS WHEN DESIGNING AN ECD PROGRAM KEY TAKEAWAYS FOR TAJIKISTAN • ECD models can draw on many different settings, including primary healthcare centers, community groups settings, schools, households, crèches and kindergartens. • A key aspect of developing integrated ECD programs is building an interdisciplinary team inclusive of parents, extended families, community members, and healthcare professionals that can provide services across different settings. Account for parental affordability in programs is also important. • Among low-income countries, careful consideration of resource requirements including cadre of worker, training, and supervision is critical for further scaling implementation. • While much of ECD evidence includes a parenting program component, it is difficult to measure the impact on child development despite improved parent-child relationships and maternal sensitivity. Frequency of follow-up and the length of parenting programs make a difference. • In resource-constrained settings, adaptations and enhancements to existing community-based parenting programs encourage ownership and contribution and improve affordability. A cadre of ECD frontline workers can provide interventions. This includes parents, extended families, community members, and healthcare professionals. Different cadres include lay community members, teachers, preschool teachers, and health professionals, as described in Figure 6. More recent evidence has focused on the education and support provided by mothers, fathers, and primary caregivers. Less attention has been given to interventions focused on teachers and the role of schools. 22 FIGURE 6: Cadres of ECD frontline workers Extended Community Healthcare Parents Families Providers Professionals Mothers and Grandparents, Schools (teachers, Nurses, doctors, Fathers aunts and uncles teaching assistants). healthcare assistants Community group structures. Overall, a range of service providers can be involved in ECD interventions. Table 6 provides a description of the providers involved in each of the reviewed studies and demonstrates the range. Across these studies, one takeaway for the development of integrated programming is to focus on building interdisciplinary teams who can provide services across diverse settings. This includes qualified health workers, teachers, parents, and community members, and indicates how a range of different roles can be mobilized for the benefit of ECD. Where this can be challenging in low-income or resource- constrained settings, focusing on the interaction between two sectors can be beneficial. For example, improving health services in school settings and strengthening the communication and interaction between health care workers and teachers in these settings can be a helpful starting point. TABLE 6: Service providers, by study Program/Study Service provider(s) Community health promoters were nominated by local mothers and WASH Benefits trial trained for between 3 and 7 days. Thereafter, group-specific refresher trainings were conducted every 6 months. A 10-month parenting program A program implementer delivered parenting messages (Save the Children) Three types of frontline workers: a salaried health worker (Shasthya Alive and Thrive Program Kormi) and a community health volunteer (Shasthya Shebika), plus a new type of frontline worker: Pushti Kormi. PLA cycle with women’s A local woman of reproductive age with secondary education is selected groups as a paid female facilitator GDMP Usual healthcare providers • Beneficiary families and local program officials elected female community leaders aka ‘mother leaders’. These mother leaders are Familias en Acción conditional influential and well connected in their communities. cash transfer program • Home visitors were selected from mother leaders based on availability and literacy levels. Hogares Comunitarios de Childcare homes are led by an MC, a home-based childcare provider in Bienestar the same community 23 TABLE 6: Service providers, by study (cont.) Program/Study Service provider(s) Action Against Hunger - cash N/A (cash transfer) transfer program Lady Health Workers (LHWs) are local paid community health workers, Lady Health Worker Program provided with additional training Oportunidades (now Prospera) Usual education and primary healthcare service providers HSAs are a centrally recruited government-paid cadre. HSAs were CCD provided with 12 extra weeks of training on CCD. Village doctors, volunteers, women cadres, and maternal and childcare IECD in rural China workers as well as mobile experts in child healthcare, early education, and social work Home visitors, comprising community health agents and development Reach Up agents in Brazil and ECD teaching assistants in Zimbabwe Right@Home Program Qualified nurse MOTHERS AND FATHERS: PARENTING PROGRAMS FOR BEHAVIOR CHANGE Previous reviews have highlighted how parent programs are used to promote nurturing care and note the importance of length of exposure and follow-up. Home visits are effective for prenatal education on infant development, health, nutrition, and parenting skills. Interventions are known to be most effective when they include two or more techniques to tackle knowledge, behavior, social input, and memory. Much of the ECD evidence includes a parenting program component and demonstrates that the length of exposure and follow-up matters. The two most widely established programs are CCD and the Reach Up and Learn program, as described in Box 4. BOX 4: CCD and Reach up and learn CCD is a World Health Organization (WHO)-UNICEF program to promote the psychosocial development of young children through home visits, community workers, and facility-based providers. CCD evolved from a module in the Integrated Management of Childhood Illness initiative and has been proven as a cost-effective intervention to improve health, growth, and development in multiple settings (Schäferhoff et al. 2015; Richter et al. 2017). Reach up and learn is a structured training course based on an evaluation of the Jamaica Home Visit (JHV) program. The JHV program is one of the original and most widely regarded sources of evidence, having been tested and replicated in countries such as Bangladesh, Uganda, and Colombia (Singla, Kumbakumba, and Aboud 2015; Attanasio et al. 2014; Hamadani et al. 2006). 24 Evidence suggests that parenting programs improve parent-child relationships and maternal sensitivity, but it is difficult to measure the impact on child development. A 2017 systematic review of parenting interventions (Rayce et al. 2017) identified 19 papers representing 16 trials that investigated the effects of parenting interventions delivered to at-risk parents of infants aged 0–12 months until June 2016. They found a small but significant positive effect on overall child behavior, but no significant effects on child cognitive behavior or the child behavior subscales internalizing or externalizing. A medium-sized effect was found on overall parent–child relationship and maternal sensitivity. As more evidence emerges from Nurturing Care, CCD and Reach Up and Learn programs, we will gain further understanding of the impact of this curricula. Evidence around parenting interventions is not complete; there are still questions on quality implementation, the length of exposure, and follow-up. A meta-analysis (Baudry et al. 2017) focused on adult mothers examined whether early intervention could help foster more positive cognitive development in the 0- to 4-year-old children of adolescent mothers. Twenty-two studies involving 29 different intervention strategies were reviewed. Interventions yielded greater effect sizes when they focused specifically on the quality of parent–child interaction or included parent–child interaction, compared to those that emphasized maternal support and education. This scoping review confirmed the reported challenges of evaluating how parenting programs translate into anthropometric and health impacts. A 10-month responsive feeding and stimulation parenting program in rural Bangladesh effectively promoted cognitive and language development indicators but not morbidity and length-for-age. Of particular note, this community-based approach found that group sessions significantly reduced depressive symptoms compared to predominantly home or clinic visits (Yousafzai and Aboud 2014). COMMUNITY MEMBERS Existing literature documents show how community members have been successfully drawn upon as frontline ECD community workers in various low-resource or rural contexts. Well-known examples include the Colombian Madre comunitaria, a ‘community mother’ of the Hogares Comunitarios de Bienestar Program, and the community mother leaders of the Familias en Acción, a conditional cash transfer program, also in Colombia. The potential of participatory approaches to encourage community ownership and beneficiary contributions has also been helpful; however, affordability needs to be carefully understood. In the Early Childhood Education and Development Project (Indonesia), communities chose from a list of service options, including playgroups, formal kindergartens, health posts, community centers with integrated ECE services, and outreach services such 25 as home visiting or mother-child meetings. Furthermore, communities could combine more than one type of service. There will usually be nominal costs incurred by parents for goods such as uniforms and stationery. Models vary in funding strategies, from publicly funded (school based) to cost recovery fees (community, center, and home based) models. The most appropriate model will depend on affordability for parents. For example, home-based ECEC Centers in Colombia successfully operated whilst charging monthly fees from parents. Conversely, removing the fees parents pay for kindergarten was the most direct way to remove the affordability barrier to early education and increase participation of very poor children in the Springboard for School Readiness Project in Bulgaria. In a review of published studies, we found that CM can be effectively mixed with other messaging channels to improve language skills, nutrition, hygiene, and care-seeking practices. Frongillo et al. (2017) report findings from a comparative analysis of two approaches in Bangladesh. First, a comprehensive CM intervention (see Box 5 for details). Second, nutrition counselling via household visits to mothers of children up to two years of age. The analysis showed improved language and gross motor development in children aged 6–48 months, partially attributed to improved minimal dietary diversity and through consumption of iron rich foods. Other studies illustrate the benefits of community group and mobilization approaches. The PLA, a women’s group that is community based, also demonstrates the efficacy of CM to improve feeding, hygiene, and care-seeking behaviors, resulting in improved morbidity outcomes (Younes et al. 2015). However, failure to see differences in dietary diversity could reflect a lack of targeted messaging on this topic. BOX 5: The Alive and Thrive Program • IPC, MM, and a CM intervention • The MM component included national broadcast of seven television spots that targeted mothers, family members, health workers, and local doctors with messages on various aspects of infant and young child feeding (IYCF). • The CM included sensitization of community leaders to IYCF and community theatre shows with a focus on IYCF. Studies from this review identified that training community members often lasted less than two weeks, much less than that delivered to health and education professionals. Specific training program lengths span four days (Aboud et al 2013), 40 hours (Hogares Comunitarios de Bienestar) 3–7 days (WASH Benefits trials), one week (PLA), 10 days (Reach Up), and two weeks (Familias en Acción). The Lady Health Worker training program (Pakistan) stands apart, providing LHWswith 15 months of training in the prevention and treatment of common childhood illnesses. 26 Training curricula also varied greatly with program focus. The WASH benefits trial demonstrates wide curricula for community health promoters that span behavior change activities, hardware use, provision of supplements, anthropological measurement, and referrals (Null et al. 2018; Stewart et al. 2018). Pakistan’s BRAC training demonstrates an active shift away from more didactic education and counselling to an evolved approach that encourages LHWs to listen, ask questions, and solve problems. Refreshers or follow-up training are also common for programs with a delayed start or after about six months of duration (for example, Familias en Acción, Reach Up, WASH benefits). Previous studies, and lessons from studies within this review, identified supportive supervision as a consistent element of community-based programs. Supervisors are usually qualified professionals; Familias en Acción mentors have an undergraduate degree in psychology or social work or have relevant experience and undergo six weeks of training. Supervision teams may include different skill sets, as exemplified by the BRAC program strategy for regular monitoring and supervision by BRAC staff, district managers, headquarters staff, and an independent team of monitors. Approaches to supportive supervision varied across studies. Direct and indirect observations were utilized by BRAC and the WASH Benefits trials. Specifically, supervision under WASH Benefits was provided through in-person, one-on-one spot- checks by supervisors, and via phone and text message support. TEACHERS AND SCHOOLS Teacher training has been combined with mass communication campaigns to target child health, nutrition, and education, but evidence of impact is limited. It is often challenging to evaluate programs delivered to a mass population, without a control group, and where the intervention is part of a wider program. However, media and communication are thought to effectively target families and young children, including those in rural areas, and can be effectively implemented through private partnership (Kataoka et al. 2011). Mass communication approaches were scaled in Zanzibar and Mexico, with benefits for targeting areas with poor access to ECD and education. The Radio Instruction to Strengthen Education (RISE) (Zanzibar) improved literacy, math, and life skills amongst participating preschool to grade 2 (age 8 years) children during program delivery between 2006 and 2011. The post-project report for the program stated that by project close, RISE had “provided ECD opportunities to over 35,000 Zanzibari children and trained over 809 formal teachers and non-formal facilitators in IAI pedagogy, classroom management and other key areas. The project also distributed over 28,000 teaching and learning materials developed by the team” (Kataoka et al. 2011). 27 Evidence from our review highlighted the potential of retraining teachers in preschool curricula. In Pakistan, local women in the community and lower primary school teachers were trained to serve as preschool teachers and caregivers as part of the Releasing Confidence and Creativity Program. Studies provide strong evidence for the need for training and supportive supervision, including skills training and on-the-job coaching with regular refreshers. Supportive supervision strategies include modelling, problem solving, peer-to-peer learning, supervisory checklists, and feedback. HEALTH AND TEACHING PROFESSIONALS Existing evidence from middle- and high-income countries highlights that qualifications are one of the strongest predictors of quality amongst education staff. Qualifications refer to the level of specialization and practical training received, types of professional development, and years of experience. Three different types of staff typically work in center-based settings in Europe: education staff, usually with qualifications to the tertiary level: “educational staff are usually qualified to tertiary level (Bachelor’s level), care staff with a minimum qualification at upper or postsecondary non-tertiary education; and auxiliary staff/assistants who are not qualified or have a minimum qualification at upper secondary level...” (Bertram et al. 2016). Coverage and cost can limit the feasibility of such specially trained cadres under large national programs or targeted rural programs. Effective training modules have been developed to build skills in facilitation and communication amongst community health workers or educators on IYCF and care topics. Where adherence is a prominent challenge amongst programs reviewed here, the cultural acceptability and local influence of selected service providers have also demonstrated to be important determinants of how well parents will engage with and access an ECD program. Often, multiple frontline workers will need to be trained in the components of an integrated ECD program. Teachers and healthcare professionals were trained to deliver a range of services under the Oportunidades cash transfer program, including prenatal education, healthcare, newborn screening, nutrition services and supplements, parent education, and child assessments in Mexico. Studies urge health providers to be treated as partners with health managers to encourage ownership of program delivery. Liberia’s Early Childhood Development Community Education and Awareness Program (ECDCEAP) has trained community health workers and midwives, mental health professionals, preschool teachers, lawmakers and other professionals, and home visitors with the aim to raise community awareness and develop the skills of service providers. Anecdotal evidence suggests success in influencing cultural norms and encouraging the inclusion of men in the program, although formal evaluation is needed. 28 Country experiences from this review highlight the variation in training approaches, ranging from models of up-skilling professionals in one sector to integrated, multi-professional training program. Experiences from Malawi and China highlight the variation in approaches to training. In Malawi, the CCD program trains centrally recruited government-paid HSAs. HSAs receive 12 weeks of training and cover topics of community health, family health, environmental health, prevention and control of communicable diseases, and community case management, including HIV, malaria, diarrheal diseases, pneumonia, and under nutrition (Gladstone et al. 2018). In contrast, in rural China, an IECD program in rural areas of China recruited mixed cadre teams that include doctors, volunteers, women cadres, and maternal and childcare workers as well as mobile experts in child healthcare, early education, and social work (Liu et al. 2018). Finally, program nurses in the right@home home-visiting model (Goldfeld et al. 2018) undertook 4 hours of online training, 12 hours of face-to-face training, and an additional 7-hour training in the right@home focus modules. Finally, where multiple cadres of service providers can be utilized in an ECD program, remuneration varies according to cadre and setting. Community-based frontline workers may be unpaid (Familias en Acción), paid (Lady Health Worker initiative, Pakistan and PLA cycle, Bangladesh, WASH Benefits trial), or mixed (Alive and Thrive, Bangladesh). Alive and Thrive involved three types of frontline workers: a salaried health worker, a community health volunteer, and a new type of frontline worker (Frongillo et al. 2017). 29 CHAPTER 8: CONSIDERING TIMING (PROGRAM LENGTH AND INTERVENTION FREQUENCY) WHEN DESIGNING AN ECD PROGRAM KEY TAKEAWAYS FOR TAJIKISTAN • Longer programs with frequent intervention points were the key to success, because they allow for wider coverage and more follow-up. As Tajikistan seeks to build toward a comprehensive ECD offering, it will be important to consider how to sustainably develop services so that they allow for frequent contact with children and families. • Encouraging adherence to a program of interventions is also a challenge and considering a family’s incentives to continue to participate in a program is also an important consideration for long-term success. The length and frequency of programming has been highlighted as an enabler of success; longer program with frequent touchpoints achieve better results. The review of the IECD, from China, demonstrated an impact in rural areas only after participants had been enrolled for more than a year (Liu et al. 2018). However, program effects may not be observable after one year. The WASH Benefits trial is an example where an effect can be observed in some development scores after one year but was lost after two years, a finding that may be attributed to low intervention adherence. Less frequent contact (that is, monthly) between health promoter and caregiver is suggested to have limited the potential for improvements in developmental outcomes. Evidence from the Lady Worker Program also demonstrates that improved developmental scores may be observed at one year and again not detected after two years. A trend likely due to lack of continued exposure. Hence there is an argument for a continuum of care to maintain positive intervention effect. In the home-based ECCE Centers (Colombia), benefits were only seen after enrollment for 15 months. The optimal structure of a program varied by delivery setting and there is limited evidence to show the efficacy of fortnightly home visits with focused 30- to 60-minute sessions. In general, smaller groups were found to be more effective and group sessions can last between one and two hours. They can be supported by booster sessions, particularly for short programs. Our review of the literature, summarized in Table 7, found support for greater frequency of visits, particularly for enabling adherence in water and sanitation programming. The WASH benefits trial in Kenya and Bangladesh 30 drew on community health providers for monthly household visits as part of a wider intervention package that combined water, sanitation, handwashing, and nutritional intervention components (but notably, no responsive caring component). Evidence suggests that combined interventions might have improved child motor development after one year but there were no differences sustained beyond two years old or between other groups or mental development outcomes in the Kenyan context. Authors indicate that less frequent contact (that is, monthly) between health promotor and caregiver could have limited the potential for improvements in developmental outcomes, but is more realistic for a government-funded program. Within this review, adherence was identified as a barrier to efficacy, indicating the importance of incentivizing adherence. We reviewed programs with a minimum duration of six months to a maximum duration of 14 years. Trials ran for a median duration of two years and one study provided a follow-up after another two years. Similar to other studies, an evaluation of Colombia’s GDMP highlighted adherence as a key limitation. In response, one option proposed by Carrillo et al. was that of making ‘program participation conditional on limiting the number of absences over a given period of time’ to encourage participant retention. National programs such as the Brazilian Bolsa Familia and Mexico’s Oportunidades successfully attach conditions to ensure program participants receive the intervention. However, this approach comes at the risk of excluding participants in need of services and, overall, the approach has not been fully evaluated to understand the implications. Further evidence is needed to understand how to regulate quality and improve the consistency of evaluation of implementation processes. In terms of regulation of quality, it remains a question as to how much regulation is required and in what format. In Organisation for Economic Co-operation and Development (OECD) countries, inspection was used more frequently than accreditation; however, in resource-constrained environments this may not be feasible. In terms of implementation, evaluation of processes has not been consistent, and tools are not always standardized for different contexts. 31 TABLE 7: Summary of program length and frequency Program/Study Overall program length Intervention frequency WASH Benefits trial Trial length of 2 years • Monthly home visits A 10-month parenting Program length of 10 • 14 sessions, fortnightly for 4 months and monthly program (Save the months for 6 months Children) • Monthly home visits Alive and Thrive Program length of 4 • 7 television spots Program years • Ongoing sensitization PLA cycle with Program length of 21 • Monthly women’s groups months • Multiple touchpoints a) Infants are monitored via 3 to 4 check-ups between 1 and 12 months of life. b) Iron supplementation is provided for 30 days Program length of 10 GDMP every 6 months starting from 6 months of age. years c) Routine monitoring of children between 2 and 7 years, at: 25–30 months, 31–36 months, 37–48 months, 49–60 months, 61–66 months, 67–72 months, 73–78 months, and 79–84 months. Familias en Acción Trial length of 18 months • Weekly home visits • Daily homecare Hogares Comunitarios Established in 1972 • Participating parents are required to pay a monthly de Bienestar fee not higher than 25% of daily minimum wage Action Against Program length of 6 • The cash or voucher amount was given every month Hunger - cash transfer months over six consecutive months program Lady Health Worker Program length of 3 • Monthly home visits Program years • Occasional group meetings • Paid every two months in cash at distribution Oportunidades points located in towns at the time of this study;. Established in 1997 (now Prospera) Electronic transfer and prepaid cards introduced in 2011) • Six HSAs were instructed to provide at least 2 Program length of 6 CCD groups a month (12 groups in 6 months) and 2 visits months a month for each participant (120 visits in 6 months) • Clinic-based check-ups, monitoring, and counseling were provided at 7 days, 28 days, 3 months, 6 months, 8 months, 12 months, 18 months, 24 Program length of 2 months, 30 months, and 36 months IECD in rural China years • Center-based group sessions were conducted 3–4 days a week. • Outreach visits were conducted every two months 3 years (Zimbabwe) and • Fortnightly home visits of duration 30–50 minutes Reach Up 1 year (Brazil) (Zimbabwe) or 20–50 minutes (Brazil) First 2 years of a child’s • The program schedule includes a minimum of 25 Right@Home program life home visits (approximately 60–90 minutes) REFERENCES Arregoces, Leonardo, Rob Hughes, Kate M. Milner, Victoria Ponce Hardy, Cally Tann, Arjun Upadhyay, and Joy E. Lawn. 2019. “Accountability for Funds for Nurturing Care: What Can We Measure?” Archives of Disease in Childhood 104 (April): S34–42. https://doi. org/10.1136/archdischild-2018-315429. Banerjee, Anshu, Pia Rebello Britto, Bernadette Daelmans, Esther Goh, and Stefan Peterson. 2019. “Reaching the Dream of Optimal Development for Every Child, Everywhere: What Do We Know about ‘How To’?” Archives of Disease in Childhood. BMJ Publishing Group. https://doi.org/10.1136/archdischild-2019-317087. Baudry, Claire, George M. Tarabulsy, Leslie Atkinson, Jessica Pearson, and Audrey St-Pierre. 2017. “Intervention with Adolescent Mother–Child Dyads and Cognitive Development in Early Childhood: A Meta-Analysis.” Prevention Science 18 (1): 116–30. https://doi. org/10.1007/s11121-016-0731-7. Black, Maureen M., Susan P. Walker, Lia C. .H Fernald, Christopher T. Andersen, Ann M. DiGirolamo, Chunling Lu, Dana C. McCoy, et al. 2017. “Early Childhood Development Coming of Age: Science through the Life Course.” The Lancet 389 (10064): 77–90. https://doi.org/10.1016/S0140-6736(16)31389-7. Boggs, Dorothy, Kate M. Milner, Jaya Chandna, Maureen Black, Vanessa Cavallera, Tarun Dua, Guenther Fink, et al. 2019. “Rating Early Child Development Outcome Measurement Tools for Routine Health Programme Use.” Archives of Disease in Childhood 104 (April): S13–21. https://doi.org/10.1136/archdischild-2018-315431. Britto, Pia R., Stephen J. Lye, Kerrie Proulx, Aisha K. Yousafzai, Stephen G. Matthews, Tyler Vaivada, Rafael Perez-Escamilla, et al. 2017. “Nurturing Care: Promoting Early Childhood Development.” The Lancet 389 (10064): 91–102. https://doi.org/10.1016/ S0140-6736(16)31390-3. Center on the Developing Child. 2016. “From Best Practices to Breakthrough Impacts.” Harvard, 1–52. papers3://publication/uuid/E7B72260-6FCB-4F0D-8F58-13986B7B04F0. Daelmans, Bernadette, Maureen M. Black, Joan Lombardi, Jane Lucas, Linda Richter, Karlee Silver, Pia Britto, et al. 2015. “Effective Interventions and Strategies for Improving Early Child Development.” BMJ (Clinical Research Ed.) 351: h4029. https://doi.org/10.1136/bmj.h4029. Fink, Günther, Evan Peet, Goodarz Danaei, Kathryn Andrews, Dana Charles McCoy, Christopher R. Sudfeld, Mary C. Smith Fawzi, Majid Ezzati, and Wafaie W. Fawzi. 2016. “Schooling and Wage Income Losses Due to Early-Childhood Growth Faltering in Developing Countries: National, Regional, and Global Estimates.” American Journal of Clinical Nutrition 104 (1): 104–12. https://doi.org/10.3945/ajcn.115.123968. Galasso, Emanuela, and Adam Wagstaff. 2019. “The Aggregate Income Losses from Childhood Stunting and the Returns to a Nutrition Intervention Aimed at Reducing Stunting.” Economics and Human Biology 34 (August): 225–38. https://doi.org/10.1016/j. ehb.2019.01.010. Grantham-Mcgregor, Sally M., Lia C.H. Fernald, Rose M.C. Kagawa, and Susan Walker. 2014. “Effects of Integrated Child Development and Nutrition Interventions on Child Development and Nutritional Status.” Annals of the New York Academy of Sciences 1308 (1): 11–32. https://doi.org/10.1111/nyas.12284. 32 Milner, Kate M., Raquel Bernal Salazar, Sunil Bhopal, Alexandra Brentani, Pia Rebello Britto, Tarun Dua, Melissa Gladstone, et al. 2019. “Contextual Design Choices and Partnerships for Scaling Early Child Development Programmes.” Archives of Disease in Childhood 104 (April): S22–33. https://doi.org/10.1136/archdischild-2018-315433. Milner, Kate M., Sunil Bhopal, Maureen Black, Tarun Dua, Melissa Gladstone, Jena Hamadani, Rob Hughes, et al. 2019. “Counting Outcomes, Coverage and Quality for Early Child Development Programmes.” Archives of Disease in Childhood 104 (April): S3–12. https:// doi.org/10.1136/archdischild-2018-315430. Perkins, Jessica M., Rockli Kim, Aditi Krishna, Mark McGovern, Victor M. Aguayo, and S. V. Subramanian. 2017. “Understanding the Association between Stunting and Child Development in Low- and Middle-Income Countries: Next Steps for Research and Intervention.” Social Science and Medicine. Elsevier Ltd. https://doi.org/10.1016/j. socscimed.2017.09.039. Rayce, Signe B., Ida S. Rasmussen, Sihu K. Klest, Joshua Patras, and Maiken Pontoppidan. 2017. “Effects of Parenting Interventions for At-Risk Parents with Infants: A Systematic Review and Meta-Analyses.” BMJ Open 7 (12): 1–20. https://doi.org/10.1136/ bmjopen-2016-015707. REFERENCES Richter, Linda, Maureen Black, Pia Britto, Bernadette Daelmans, Chris Desmond, Amanda Devercelli, Tarun Dua, et al. 2019. “Early Childhood Development: An Imperative for Action and Measurement at Scale.” BMJ Global Health 4 (Suppl 4): e001302. https://doi. org/10.1136/bmjgh-2018-001302. Richter, Linda M., Bernadette Daelmans, Joan Lombardi, Jody Heymann, Florencia Lopez Boo, Jere R. Behrman, Chunling Lu, et al. 2017. “Investing in the Foundation of Sustainable Development: Pathways to Scale up for Early Childhood Development.” The Lancet 389 (10064): 103–18. https://doi.org/10.1016/S0140-6736(16)31698-1. Smith, Joanne A., Helen Baker-Henningham, Alexandra Brentani, Rose Mugweni, and Susan P. Walker. 2018. “Implementation of Reach up Early Childhood Parenting Program: Acceptability, Appropriateness, and Feasibility in Brazil and Zimbabwe.” Annals of the New York Academy of Sciences. https://doi.org/10.1111/nyas.13678. Statistical Agency under the President of the Republic of Tajikistan, Ministry of Health and Social Protection of Population of the Republic of Tajikistan, and ICF. 2018. “Tajikistan 2017 Demographic and Health Survey.” The World Bank. 2018. 10 Years of Chile Grows With You (Chile Crece Contigo). Washington, DC: World Bank. UNICEF (United Nations Children's Fund). 2017. Early Moments Matter for Every Child. WaterAid, SHARE, and Healthy Start. 2016. The Missing Ingredients: Are Policy-Makers Doing Enough on Water, Sanitation and Hygiene to End Malnutrition? World Bank. 2019. World Development Report 2019: The Changing Nature of Work. Washington, DC: World Bank. WHO (World Health Organization). 2015. The Global Strategy for Women’s, Children’s and Adolescents’ Health (2016-2030): Survive, Thrive, Transform. 33 34 ANNEX 1: SUMMARY OF STUDIES TABLE 1: Summary of studies Program Country Study References Brief description WASH Benefits Bangladesh Effects of water Intervention description: WASH Benefits was a cluster-randomized trial in Bangladesh and Kenya trial quality, sanitation, designed to assess the independent and combined effects of water, sanitation, hand washing, and hand washing, and nutritional interventions on child growth, health, and development after two years of intervention. Six nutritional interventions arms were tested in this trial: chlorinated drinking water; improved sanitation; hand washing with soap; on diarrhea and combined water, sanitation, and hand washing; improved nutrition through counselling; and provision of child growth in rural lipid-based nutrient supplements. Three control arms included a double-sized active control (monthly Bangladesh: A cluster household visits to measure child mid-upper arm circumference) and a passive control (no visits). randomized controlled Findings: trial (Luby et al. 2018) Diarrhea: Significant effects on growth with improved nutrition, reductions in diarrhea with improvements Kenya • Effects of Water in water, sanitation, hand washing, or nutrition in Bangladesh (Luby et al. 2018). Unlike in Kenya, there was Quality, Sanitation, high adherence (70%) and a significant impact on reduction of diarrhea in this Bangladesh study. Small Hand Washing, improvements were reported in growth with improved nutrition; no intervention arms reduced diarrhea and Nutritional prevalence in Kenya (Null et al. 2018). The difference may be attributed to adherence; for sanitation, hand Interventions on washing, and nutrition, adherence was more than 70% at year 1, hand washing fell to less than 25% at year Child Development in 2, and for water was less than 45% at year 1 and less than 25% at year 2. Rural Kenya (WASH Development: Limited evidence of independent and combined water, sanitation, hand washing, and Benefits Kenya): A nutritional intervention effects were also found on child development outcomes in Kenya (Stewart et al. Cluster-Randomized 2018). There were no apparent differences between groups for the communication, gross motor, personal Controlled Trial social, or combined Extended Ages and Stages Questionnaire (EASQ) scores, but they found small (Stewart et al. 2018) increases in motor, language, and personal social development across most of the individual and combined intervention groups. (These results differ to findings by Tofail et al. (2018) who reported improvements • Effects of Water in indicators of child development across all intervention groups. There were improvements in motor, Quality, Sanitation, language, and personal social development in both of the nutrition groups that received lipid-based Hand Washing, nutrient supplements amongst pregnant women in Bangladesh.) and Nutritional Interventions on Nutrition: Significantly higher length-for-age and weight-for-age Z scores, and correspondingly lower Diarrhea and Child prevalence of stunting and underweight were reported in the in the combined water, sanitation, hand Growth in Rural washing, and nutrition group and the nutrition only group (Stewart et al. 2018). Kenya: A Cluster- Interpretation: A combined water, sanitation, and hand washing intervention provided little additive benefit Randomized over single interventions but this depends on the setting. WASH Benefits tested the idea that health or Controlled Trial (Null WASH interventions can influence developmental outcomes. They include a range of behavior change et al. 2018) techniques as advised by the literature and provide a stipend plus supportive supervision. They do not find strong evidence given that, as with other studies, the impact was not sustained by year 2 across settings. ANNEX 1: SUMMARY OF STUDIES Program Country Study References Brief description A 10-month Bangladesh Effectiveness of a Intervention description: Parenting program providing a combination of group meetings and home visits for parenting parenting program community groups of mothers. This program delivered responsive feeding and stimulation messaging with program (Save in Bangladesh to illustrative cards for hygiene, responsive feeding, play, communication, gentle discipline, and nutritious foods. the Children) address early childhood Findings: This program reported a positive effect on cognitive, receptive language, and expressive health, growth, and language indicators of development, as well as parenting practices related to stimulation and knowledge development. (Aboud of development milestones. Dietary diversity improved significantly in this program but younger children et al. 2013) demonstrated more improved dietary diversity than older children. All children became more stunted; however, twice as more children were stunted in the control group compared to those enrolled in the program. Intervention mothers obtained better knowledge scores and depressive symptoms were significantly reduced among women who met as a group twice monthly compared to those who had mostly home or clinic visits. Interpretation: This parenting program was effective in promoting cognitive and language development of children, but not morbidity and length-for-age. The high prevalence of stunting reflects a challenging rural setting with typically poor dietary diversity but change differences reflect a non-significant positive impact of the program. Alive and Bangladesh Large-Scale Behavior- Intervention description: An ‘intensive’ program included intensified interpersonal counseling (IPC), mass Thrive Change Initiative media (MM), and community mobilization (CM) components compared to standard nutrition counseling Program for IYCF, Advanced + less-intensive MM + CM (non-intensive). The Pushti Kormi (PK), together with regular community health Language, and Motor workers called Shasthya Sebika, conducted multiple age-targeted IYCF-focused counseling visits to Development in a households with pregnant women and mothers of children aged two years of age. Cluster-Randomized Findings: This study found significant improvements in language and gross motor development in children Program Evaluation in aged 6–48 months. Developmental advancement at 6–23.9 months was partially explained through Bangladesh (Frongillo improved minimum dietary diversity and the consumption of iron-rich food. et al. 2017) Interpretation: A diverse set of child outcomes, including functional outcomes such as child development, can be improved through integrated nutrition program. PLA cycle Bangladesh The Effect of Intervention description: Women were eligible to become members of the women’s groups if they with women’s Participatory were 15–49 years of age and resided in the intervention areas. The PLA cycle is designed as four phases: groups Women’s Groups identifying and prioritizing under-5 health problems (phase 1), help identify possible strategies (phase 2), on Infant Feeding and support the planning, implementation (phase 3), and monitoring of the strategies led by the women's and Child Health group members (phase 4). Knowledge, Behavior, Findings: Women reported a significant increase in exclusive breast-feeding for at least 6 months and and Outcomes in decrease in under-5 morbidities, compared with control areas. However, no differences were observed in Rural Bangladesh: A dietary diversity scores or immunization uptake. Controlled Before-and- Interpretation: Authors conclude that "Community mobilisation through Participatory women’s groups after Study (Younes et can be successfully adapted to address health knowledge and practice in relation to child’s health, al. 2015b) leading to improvements in a number of child health indicators and behaviors." This may be true for many behaviors but there is currently little evidence that complementary feeding (beyond exclusive breast feeding) is one of them. 35 36 Program Country Study References Brief description GDMP Colombia Attainments and Intervention description: The GDMP was created in 2000 to reduce rates of infant morbidity and mortality Limitations of an Early by providing comprehensive early childhood care. The GDMP aims for universal coverage among children Childhood Program in and provides education to parents about balanced nutrition, immunization, early stimulation, accident Colombia (Carrillo et al. prevention, oral hygiene and acute diarrhea. Infants between their first month and first year of life are 2015) monitored through three to four check-ups performed annually. Findings: This study reported an impact of the GDMP on vaccination only, as it increases the probability of being immunized. Carrillo et al. reported no observable impact on malnutrition. However, when the sample was divided between the more prosperous regions and the rest of the country, authors reported a more positive impact on acute malnutrition in the latter. Interpretation: The GDMP is one of many ECD programs in Colombia, a unique and specific context. In this study, descriptive statistics indicate that families that access such schemes in Colombia tend to be better educated, of higher socioeconomic status, and urban based but education for parents on early childhood care has greatest impact in the poorest regions, where the level of maternal education is low. Another lesson from this program is on the issue of adherence—only 13% of participating children made the number of visits required by the GDMP. The authors therefore recommend establishing strict compliance conditions similar to the Brazilian Bolsa Familia and Mexico’s Oportunidades. Familias en Colombia Using the Infrastructure Intervention description: Familias en Acción is a conditional cash transfer program/IECD intervention, Acción of a Conditional Cash evaluated by Attanasio et al. (2014b) as a cluster randomized controlled trial. Familias en Acción is Transfer Program to novel in its approach, linking an ECD program to an existing social welfare system, designed so that Deliver a Scalable IECD children naturally “graduate” into the existing community care services as part of a scalable intervention. Program in Colombia: Intervention components include psychosocial stimulation (weekly home visits with play demonstrations), Cluster Randomized micronutrient sprinkles given daily, and both combined. The psychosocial stimulation program was based Controlled Trial. on the Jamaican home visiting model and the micronutrient supplementation consisted of Sprinkles (Attanasio et al. 2014b) (Hexagon Nutrition, Mumbai, India), designed for anemia. Findings: No intervention affected height, weight, or hemoglobin levels. Stimulation significantly improved cognitive scores and receptive language. Micronutrient supplementation had no significant effect on any outcome and there was no interaction between the interventions. Interpretation: This study adds to the literature on another combined intervention that shows evidence of improved developmental indicators with no impact on linear growth. The importance of identifying the underlying cause of anemia is clear from this program. The underlying assumption of this intervention is that anemia is the result of micronutrient deficiency, rather than chronic infections and hemoglobinopathies. However, this study showed no effect of micronutrient supplementation on hemoglobin levels. It is not possible to fully elucidate the effective components from this study, which did not measure adherence (to explore the surprising finding that supplementation had no impact on hemoglobin levels) or parental behaviors (suggested to be significant mediating factors). Authors recommend measuring iron status as a critical step toward the inclusion of WASH as an intervention component. Program Country Study References Brief description Hogares Colombia Subsidized childcare Intervention description: Hogares Comunitarios de Bienestar is a subsidized home-based childcare Comunitarios and child development program, established in 1972. The program delivers home-based childcare, supplemental nutrition, and de Bienestar in Colombia: Effects of psychosocial stimulation to 783,399 low-income children under the age of 6 (32% coverage) throughout Hogares Comunitarios most of Colombia’s 1,100 municipalities. This study was designed to identify the effects of exposure to de Bienestar as a Hogares Comunitarios de Bienestar on children’s nutritional status and cognitive and socioemotional function of timing and development, comparing beneficiary children who had been in the program for a long time with length of exposure beneficiary children who had been in the program for a month or less, by age group. They used a short (Bernal and Fernández parent form of the Early Development Instrument (EDI). 2013) Findings: Cognitive development improved after at least 15 months of exposure for children between 3 and 6 years of age. Socio-emotional skills improved for children older than 3 after at least 15 months of program exposure. No significant gains were found for nutritional status. Interpretation: Duration matters: this combined intervention including nutritional supplements and psychosocial stimulation did lead to small improvements in cognitive development and socio-emotional skills after at least 15 months of exposure. The cadre and training of the childcare providers may have contributed to the positive effect in this study but was not tested. Action Against Pakistan Impact Evaluation of Intervention description: Three intervention arms were as follows: standard cash (SC), a cash transfer of Hunger - Different Cash-Based PKR 1,500 (approximately US$14; PKR 1 = US$0.009543); double cash (DC), a cash transfer of PKR 3,000; cash transfer Intervention Modalities or a fresh food voucher (FFV) of PKR 1,500. The cash or voucher amount was given every month over six program on Child and Maternal consecutive months. The control group (CG) received no specific cash-related interventions. Nutritional Status Findings: Primary outcomes of interest were prevalence of being wasted (weight-for-height Z score [WHZ] in Sindh Province, < −2) and mean WHZ at 6 months and at 1 year. Significant differences on the primary outcome were seen Pakistan, at 6 Mo only at 6 months. The larger cash transfer had the greatest effect on wasting, but only at 6 months. All three and at 1 y: A Cluster intervention groups showed similar significantly lower odds of being stunted. An unintended outcome was Randomized Controlled observed in the FFV arm: a negative intervention effect on mean hemoglobin status. Trial (Fenn et al. 2017) Interpretation: Cash transfers are an effective way to treat wasting and stunting up to 6 months but the value matters. Caution is needed when applying restrictions to food vouchers in order to secure a diverse food basket that provides adequate macro- and micronutrients. Lessons can be learned from the unintended consequences of food vouchers on hemoglobin status in this study. 37 38 Program Country Study References Brief description Lady Health Pakistan • Review of Intervention description: Monthly home visits and parenting groups for young children younger than Worker Implementation two years and their caregivers. The intervention included four arms: nutrition education and multiple Program Processes for micronutrient powders (enhanced nutrition, Sprinkles micronutrient supplementation); responsive Integrated Nutrition stimulation (a local adaptation of the CCD approach); combined responsive stimulation and enhanced and Psychosocial nutrition; and routine health and nutrition services. The control group received routine LHW services, Stimulation delivered in monthly home visits and occasional group meetings, which included health and hygiene Interventions. advice, IYCF recommendations (basic nutrition education), child growth monitoring, and immunizations. (Yousafzai and Findings: Children who received responsive stimulation had significantly higher development scores on the Aboud 2014) cognitive, language, and motor scales at 12 and 24 months of age (Yousafzai et al. 2014) and at follow-up at 4 years of age (Yousafzai et al. 2016). Those receiving responsive stimulation also had higher social– • Effects of responsive emotional scores at 12 months of age but no difference was reported at further follow-up. stimulation and nutrition Children exposed to enhanced nutrition had significantly better height-for-age Z scores at 6 months and interventions 18 months than did children not exposed to enhanced nutrition, but these differences were not observed on children's at age 4 years. Those receiving enhanced nutrition also had significantly higher development scores on development the cognitive, language, and social-emotional scales at 12 months of age. However, only language scores and growth at remained significantly higher by 24 months of age. age 4 years in a There we no additive benefits of combining responsive stimulation with nutrition interventions in this study. disadvantaged Interpretation: This study reflects an effective adapted version of the CCD program. The results indicate population in that an effect of early intervention can be lost without sustained intervention. Authors note that the Pakistan: A Jamaican cohort (on which CCD is based) received sustained intervention until adulthood, unlike children longitudinal follow-up in this study. Importantly, there was no evidence to support the addition of an enhanced nutrition arm to of a cluster- improve development scores in this setting. randomized factorial effectiveness trial (Yousafzai et al. 2016) Oportunidades Mexico Distributional effects Intervention description: A government-assisted social assistance program established in 1997 and (now of Oportunidades on designed to increase human capital through investments in education, health, and nutrition for children Prospera) early child development in extreme poverty. Four indicators of cognitive development and one indicator of non-cognitive (Figueroa 2014) development were measured in this program evaluation in 2003. Findings: Authors reported positive (and significant) evidence of the program protecting children from nutritional vulnerabilities. The effect was higher for those exposed between the ages of 2 and 5 years old and different program effects were also reported by gender and between indigenous and non-indigenous children. Interpretation: This study is unique because it looks at the effect of a cash/food-for-work program without any requirements that participants invest in child nutrition or other human capital, demonstrating a positive influence on non-cognitive abilities through a social assistance program. Program Country Study References Brief description CCD Malawi CCD in Rural Malawi: A Intervention description: A pilot of the CCD program, using adapted training materials for Malawi Model Feasibility and delivered by six HSAs in group and individual sessions. Pilot Study (Gladstone Findings: This study reported improved language and social indicators, improvements to some family et al. 2018 care indicators, and no changes to maternal stress over the 6-month period. This study reflects on different approaches to implementation and lessons learned to scaling-up, which is a useful resource for implementers. Interpretation: It is clear that the CCD model works best with supervision and support and Malawi was one of the first African countries to have ECD support through Community-Based Child Care Centers for 3–5 years old endorsed by the Ministry of Gender, Children, Disability and Social Welfare, indicating a potential for the continuum of care in Malawi. However, it can be challenging to appropriately address maternal stress amongst those living with common mental disorders such as depression. This study reported huge challenges to adherence and fidelity in a number of community visits. IECD China Effects of early Intervention description: IECD project in poor rural areas of China targeting children younger than 3 years comprehensive in Shanxi province (northern China). Clinic-based services include child health check-up, development interventions on child monitoring, feeding and nutrition counseling, and CCD counseling, as well as nutritional supplementation neurodevelopment to children aged 6–24 months. ECD centers provided group-based early stimulation activities for 3–4 days in poor rural areas of a week. A team including health and education experts and a social worker conducted outreach every two China: A moderated months. mediation analysis (Liu Findings: Intervention families showed significantly higher overall ‘home measurement of the environment’ et al. 2018) scores than control families and children performed significantly better than children in control families in overall development scores. Interpretation: This program adds to the body of evidence emphasizing the importance of the home environment. It demonstrates that duration of more than one year is an important mediator of such a multi-armed intervention. Reach Up Brazil and Implementation Intervention description: An early childhood parenting program, based on the Jamaican home visiting Zimbabwe of Reach Up early model and delivering fortnightly visits of duration 30–50 minutes (Zimbabwe) or 20–50 minutes (Brazil). childhood parenting Findings: The study explored themes of agency and common adaptations; perceptions of acceptability; program: Acceptability, appropriateness and feasibility amongst mothers, home visitors, and supervisors; and the quality of visits in appropriateness, and Zimbabwe. feasibility in Brazil and Interpretation: This program shares useful learnings on how to adapt evidence-based interventions whilst Zimbabwe (Smith et al. maintaining the core components important for efficacy of a program. 2018) 39 40 Program Country Study References Brief description Right@Home Australia Designing, testing, Intervention description: A nurse-led home visiting program. The program schedule includes a minimum Program and implementing of 25 home visits (approximately 60–90 minutes) with focus modules to promote parenting, child a sustainable nurse development, and learning. home visiting program: Findings: This study showed improved aspects of parent care, responsivity, and the home learning Right@ Home (Goldfeld environment compared to usual care. Factor analysis demonstrates that the intervention was delivered in et al. 2018) accordance to the expected content. Implementation feedback did not change the mean number of home visits provided. Clinicians reported a positive change in their experience of service delivery, and high rates of fidelity and retention suggest that replication at scale is possible. Interpretation: A theory-based home visiting intervention can improve HOME indicators when provided by trained professionals.