*Synthesis of Studies and Core Themes of This Paper*

Guided by a rights-based approach, this narrative review synthesizes selected studies and global initiatives promoting early child development. Selection for inclusion consisted of the most relevant initiatives to the areas listed in the introduction, decided by consensus, and published in English from 2001 to 2020. Specifically, we synthesized Initiatives addressing environmental factors influencing children's development, as well as prevalence and estimates of children at risk for developmental delay or loss of developmental potential in low-, middle- and high-income countries. Limitations of this approach are addressed in the paper.

Subsequently, a universal intervention framework of child-environment interactions is proposed for optimizing children's developmental functioning and progress drawing on the International Classification of Functioning, Disability and Health for Children and Youth (ICF-CY) [5,6]. Finally, we highlight the importance of adopting a global framework guiding assessments and evaluations of children's development across sectors. As such, the content of this study is organized in three themes as follows:


#### **2. A Global Initiative to Promote Early Child Development**

Although initiatives at country levels and programs at global levels by UNICEF, World Health Organization (WHO) and various non-governmental organizations (NGOs) over the last three decades have contributed to significant reductions in the scope and nature of childhood mortality and morbidity, challenges to the healthy development of young children remain pervasive, particularly in low- and middle-income countries, where the largest proportion of children in the world are found. In contrast to earlier populationbased studies focusing on specific child or maternal conditions, more recent research is increasingly recognizing the significant role of inadequate or deprived physical and social environments associated with poor developmental outcomes. Increased recognition of the significance of environmental factors is evident in the review by Black et al. [7] of

research publications since 2000 indicating that the number of publications on stimulation (*n* = 1121), micronutrients (*n* = 936) and nutrition-related issues (*n* = 508) were 2 to 8 times more frequent than studies of specific conditions such as malaria (*n* = 255), abuse and neglect (*n* = 298) or maternal depression (*n* = 139). This increased research focus on the physical and social environment is consistent with the broader agenda of the MDG and the SDGs. This perspective is in keeping with the challenge of documenting equitable early development as proposed by Barros [2], "entailing that every girl and boy should have the same opportunities to fully develop their potential, which is only achievable if they have good nutrition, good health, and a rich and stimulating home environment" (p. e873).

Indicators defining the child's loss of potential and limited opportunities have increasingly become representative of the focus of studies estimating risk for developmental delay, disability or disorders complementing other markers of developmental morbidity. Representative of this perspective is the study by Grantham-McGregor et al. [8] examining developmental potential of children in the context of the first and second MDGs, namely eradicating hunger and poverty and the completion of primary schooling by children globally, respectively. A comprehensive analysis was made of data on children under five in developing countries drawing on the indicators of stunting, available in 126 of these countries, and poverty, available in 88 of the countries. Of the 559 million children under five living in the developing countries, 22% were found to live in poverty, 28% were stunted and 39% were identified as stunted, living in poverty or both. This latter group served as the estimate of disadvantaged children defined in terms of risk for loss of developmental potential. The prevalence estimates for each of the indicators varied widely across regions of the world, with the percent of children living in poverty being lowest in Central and Eastern Europe at 4% and highest in Sub-Saharan Africa at 46%. The range of corresponding values for stunting were 14% for Latin America and the Caribbean and 39% for South Asia. The widest range was found for the combined indicators defining disadvantaged children, with 18% found in Central and Eastern Europe and 61% in Sub-Saharan Africa. The prevalence of lost developmental potential of disadvantaged children was supported by country-specific findings linking stunting and poverty to fewer years of education and limited learning per year of schooling with implications for deficits of income in adulthood.

The estimates of disadvantaged children based on 2004 data by Grantham-McGregor et al. [8] were updated and new estimates were generated for 2010 data in a study by Lu et al. in 2016 [9]. The study used essentially the same approach in the identification of children under five years of age manifesting stunting or exposed to poverty. Recalculating estimates of 2004 data for 141 of the developing countries with improved analytic methods indicated a higher prevalence (51%) of disadvantaged children than the 39% reported in 2007. However, a comparison of estimates based on analysis of data for the 141 countries at two time points, revealed a reduction in the prevalence of disadvantaged children from 51% in 2004 to 43% in 2010. The reduction in the prevalence of children at risk for poor development was attributed largely to reduction in stunting and poverty in South Asia including India and China. Similar to the findings of the Grantham-McGregor et al. [8] study, prevalence estimates varied widely across regions of the world, with the percent of children living in extreme poverty being lowest in Middle East and North Africa at 3% and highest in Sub-Saharan Africa at 54%. The range of corresponding values for stunting were 16% for Latin America and the Caribbean and 47% for South Asia. As in the 2007 study, the widest range was found for the combined indicators defining disadvantaged children, with 21% found in Latin America and the Caribbean and 70% in Sub-Saharan Africa. Lu et al. [9] conclude that findings reflect progress in global efforts to reduce the risk for poor development of young children, but the developmental potential of disadvantaged children continues to be significantly limited in developing countries, particularly in Sub-Saharan Africa. Challenges remain to promote children's development by reducing stunting and poverty through improved health and increased access to education, reinforcing the premise by Rippoin et al. that "increasing the educational level of their population will lead to better nourished populations, and the ability to improve gross domestic product (GDP)" [10].

The above studies have documented the scope of risk to developmental potential of young children using proxy variables of stunting and living in poverty. In more recent research, surveys such as the UNICEF Multiple Indicator Cluster Survey (MICS) encompass a range of variables more directly linked to developmental aspects of the young child in the Early Child Development Index (ECDI) of relevance in documenting progress to achieving the SDGs. Accessing data from a study of 35 countries, Manu et al. [11] conducted a secondary analysis of literacy-numeracy skills of 100,012 three to five-year-old children. A literacy-numeracy index was derived from the ECDI based on naming 10 letters of the alphabet, knowing four simple words and names and symbols for the numbers 1-10. The index was used to define an outcome variable, classifying children dichotomously as on-track or not on-track. Other variables of interest were availability of children's books in the home, urban or rural residence, and demographic variables of child age and gender, maternal education and a home wealth index. Analyses revealed that just over half (51.8%) of the children had one book in their home and less than a third (29.9%) met the criteria of being on track for literacy-numeracy. The facilitating role of the home environment was evident in the fact that having a book in the home almost doubled the likelihood of a child meeting the criteria for literacy-numeracy, after adjusting for maternal education, wealth index and demographic variables.

The broader aspects of development as assessed by the ECDI also served as the basis for a study by Gil et al. [12] to document the prevalence of children at developmental risk in low and middle-income countries. In this study, data were available from the administration of the MICS and Demographic and Health Surveys between 2010 to 2016 to families of 330,613 children, ages 3 to 5 years, in 63 low and middle-income countries. In addition to the EDCI, data was also obtained on contextual variables of rural/urban residence, maternal education, child gender and wealth inequality indicators. As in the previous survey studies of children in low and middle-income countries, large variability was found for prevalence of developmental risk between and within world regions as well as between countries. The prevalence of suspected developmental delay by region, based on the ECDI, ranged from a low of 10.1% for Europe and Central Asia and a high of 41.4% for West and Central Africa. Within the West and Central Africa region, the variability ranged from 24.9% for Ghana to 67.3% for Chad. The role of country income on prevalence of suspected developmental delay based on EDCI values yielded parallel findings with a prevalence of 41.2% for low-income countries and only 9.7% for high income countries. Prevalence estimates of suspected delay by assessed domains across all countries was lowest for physical development (3.5%), followed by learning (9.2%) and social-emotional development (24.0%).

Although the issue of promoting early child development as a global initiative is appropriately focused within the framework of inequalities faced by children in low and middle-income countries as illustrated above, initiatives of the SDGs are universal, applying to countries defined by higher incomes as well. Thus, the nature of developmental problems faced by young children, and how those problems are defined and surveyed in highly developed countries do differ, but the focus is the same, that of identifying children at a population level at risk for developmental delay or loss of developmental potential. Estimating the prevalence of children at developmental risk in higher income countries may involve several different surveys as is the case in the U.S. [13]. The National Survey of Children's Health (NSCH) generates prevalence estimates of 20 specified health conditions, as well as the status of overweight/obesity, risk for developmental delay and having a specified health care need. Based on data of a nationally representative sample of U.S. children 0–17 years of age (N = 91,642), Bethell et al. reported the prevalence of children with a chronic condition to be 43%, an estimate was increased to 54.1% with the inclusion of children with the status of overweight or obesity and risk for developmental delay [14]. Based on screener data to identify children with special health care needs such as needs for medication, additional health, mental health, or educational services, 19.2 % of the children were documented to have special health care needs. The National Health

Interview Survey (NHIS) is also administered to nationally representative samples of households with 3–17 year-old children to estimate the prevalence of one of ten diagnosed disabilities (attention deficit hyperactivity disorder [ADHD], autism spectrum disorder [ASD], blindness, cerebral palsy, hearing loss, learning disabilities, intellectual disabilities, seizures, stuttering or stammering and other developmental delay). In a comparative analysis of data from the 2009–2011 to the 2015–2017 administrations of the NHIS, Zablotsky et al. [15] reported an increased prevalence of any diagnosed disability from 16.2% to 17.8%. Among children 3–5 years of age, developmental problems were reflected in prevalence estimates of 10.55% for any disability, 2.73% for stuttering/stammering, 3.30% for learning disabilities, 2.13% for ADHD and 4.67% for other developmental delays.

Reporting on NHIS data for the period 2006 to 2010, Schieve et al. found the prevalence of diagnostic conditions in children 3–17 years of age to be, learning disabilities [LD] (7.8%), ADHD (7.9%), other developmental disabilities [ODD] (4.3%), ASD (0.09) and intellectual disability [ID] (0.07%) [16]. As some children were identified with more than one diagnosed condition, Schieve et al. (2012) derived prevalence on the basis of assignment of children to one of four mutually exclusive groups. This resulted in decreased prevalence estimates as follows, ASD only (0.9%), ID without ASD (0.5%), ADHD without ASD or ID (7.3%), and LD and ODD without ADHD, ASD or ID (5.0%).

As referenced above in the study by Bethell et al. [14], estimates of loss or delay of developmental potential has also been estimated on a functional basis of identifying special health care needs of children. Children with special health care needs were defined by McPherson et al. as children 0–17 years of age "who have or are at increased risk for a chronic physical, developmental, behavioral or emotional condition and who also require health and related services of a type or amount beyond the required of children generally" [17]. Prevalence estimates for children with special health care needs are derived from the inclusion of a screener in the administration of the National Survey of Children with Special Health Care Needs (NS-CSHCN), combined with the National Survey of Children's Health (NSCH) in 2016. The five screener items are "(1) need for or use of prescription medications, (2) above-routine use of medical, mental health, or educational services compared with other children, (3) daily activity limitations, (4) need or use of specialized therapies; and (5) need or use of treatment or counseling for emotional developmental or behavioral conditions" [18]. Use of the screener in the NS-CSHCN, the NSCH and the Medical Panels Survey have yielded prevalence estimates of children with special health care needs ranging from 12.8% to 19.3% [13]. A consistent finding of the prevalence estimates derived in these US studies is the role social determinants of poverty and limited parental education, a role shared with risks to development in low and middle-income countries.
