Children in poverty and the use of early development instrument mapping to improve their worlds

Abstract

Population data can be used to help physicians better understand child poverty in relation to families, geography and access to community resources. The early development instrument (EDI) is a population m easure of kindergarten children’s early cognitive, social, emotional, language and physical development. Researchers and communities in British Columbia have used EDI and socioeconomic data to examine early child development trends across neighbourhoods, school districts and provincial geographies. It highlights that while vulnerabilities at school entry are more prevalent in poorer communities, they are present in all communities. Mapping EDI data and other information help to identify communities that are more vulnerable and ones that seem to be resilient. Physicians and community partners can identify local needs and interventions that can support parents and communities in promoting healthy development of children before their entry to the first grade.

Most paediatricians and primary care physicians focus on individual children and families, and frame child poverty from that point of view. Child poverty is associated with increased risk for health problems during childhood and adult life (1). Family poverty turns into chronic and intergenerational deprivation through its impact on early child development (2,3). If early child development outcomes can be ‘uncoupled’ from family socioeconomic status (SES), there is a chance for intergenerational breaking of the cycle. The reverse is also true – addressing early developmental difficulties or vulnerabilities can prevent downward mobility. Addressing the health care problems of individual poor children, and attempting to ameliorate the impact of poverty, is an important element of primary health care. However, it is difficult to have much impact on a child’s health when his or her daily life takes place in social and physical environments that compromise his or her health. Also, medical care for one child at a time will not make a big impact on the total number of poor children, which is estimated to be 1.2 million children in Canada between zero and 18 years of age (4).

Physicians can use population data to better understand child poverty and its relation to families, geography and the availability of community resources. The understanding of the power of population level data, especially in early childhood, in which the impact of poverty is most pronounced, will encourage physician participation in community level solutions. New measurement techniques, such as the early development instrument (EDI) are making it possible to examine what makes children healthy and what the impact of poverty is on children living in different communities.

POPULATION LEVEL MEASUREMENTS

In the 1980s, the Ontario Child Health Study (OCHS) (5) was the first large-scale survey to examine the health of a representative sample of children in the province. The results indicated that more than one in five children had a mental health problem that could require clinical intervention and that children living in poor families were significantly more likely to have health difficulties.

In the 1990s, the Government of Canada launched the National Longitudinal Survey of Children and Youth (NLSCY). The survey is repeated every two years; tracks the original sample of children (over 20,000) into adulthood, and adds additional cohorts of young children each cycle.

The results of the NLSCY indicate that by the time children enter grade 1, significant preventable inequalities in development have emerged within Canadian society. The impact of family income and SES repeats the findings of the OCHS – income matters and poverty is related to increased health problems. The same finding is clear from population-based surveys in the United Kingdom (6,7).

The NLSCY data (8) show a gradient in the risk of receptive language delay, increasing gradually from the children in the highest income decile of Canadian families (5.2% delayed) to the poorest decile (approximately 26% delayed). This pattern of gradually rising vulnerability with declining SES is seen for most developmental outcomes, and frames the challenge for early child development in Canada and many other societies. The implication is that, although those at the bottom of the socioeconomic spectrum are ‘most at risk’, ‘most of the children at risk’ are spread more thinly across the more numerous middle class (Figure 1). Also noteworthy is that most children in the poorest income group do not have a language delay. Improving the state of early child development involves finding ways to create access to the conditions for healthy child development across the entire socioeconomic spectrum, including those children living in poverty.

Figure 1.

Early development instrument – percentage of vulnerable children in Canada (low early development instrument scores) by socioeconomic status (SES). Adapted from reference 26

Physician visits address the impact of poverty on an individual child mostly by patching up immediate problems. Addressing equitable development for all children means reframing the discussion of child poverty beyond minimal survival needs to equalize opportunities for children (9–11).

The findings of the NLSCY align with current child development studies and other population-based surveys. Influences on early development include family factors such as income, parenting style, family structure and dynamics; community factors such as availability of community resources, neighbourhood safety and social cohesion; and access to early childhood and family support programs (8,12).

By kindergarten, development is influenced by factors at three levels of society – the family; the neighbourhood or local community; and the broader social, economic and political environment. At the level of the family, the qualities of stimulation, support and nurturance in intimate circumstances contribute the most (10,13). These qualities, in turn, appear to be influenced by the resources that families have to devote to child-raising (significantly represented by income); by their style of parenting; and by their tendency to provide a rich and responsive environment for language which is often, but not always, associated with parental levels of formal education (14,15) and play (15).

At the level of the neighbourhood, children growing up in safe areas where the community is ‘cohesive’ in relation to children – in which it mobilizes resources formally (creates programs) and informally (treats its children like they belong there) – are less likely to be vulnerable in their development than children from similar family backgrounds living in unsafe and noncohesive neighbourhoods. Children who have stable neighbourhood environments during their early years tend to have better development outcomes than those children who are constantly changing their place of residence. Similarly, children from family backgrounds with multiple developmental risk factors will do better growing up in mixed socioeconomic neighbourhoods than in poor ghetto areas (16).

Finally, at the level of the society, access to ‘quality’ programs matter (17–19). This includes the full range of childcare, family support and family strengthening programs; public health programs including vision, hearing and dental care, and special programs for high-risk children; and broader social safety net functions such as parental leave and housing programs. Thus, the state of child development in any society is an ‘emergent property’ of a complex array of factors, many of them modifiable, at the intimate, civic and societal levels that influences each child in unique combinations (20).

The NLSCY data allow researchers to show associations between child outcomes and between all of these factors across Canada. However, the data do not allow an examination about development at the community level.

EDI

Canadian communities have embraced the population level use of the EDI to measure kindergarten readiness and, hence, early child development.

The EDI was developed by Drs Magdalena Janus and the late Dan Offord and was released in 2000 (21,22). It assesses community outcomes in child development with respect to health, learning and behaviour.

The EDI has the following characteristics:

• It is completed by kindergarten teachers based on several months of observation;

• While reliable at an individual level, it does not provide a clinical diagnosis of a child’s developmental problems;

• It provides a population level measure – results can be interpreted for groups of children;

• The results may be used to identify the weak and the strong sectors of a community; and

• The results can be used by communities to mobilize for improved child outcomes.

The EDI assesses five child development domains:

1. Physical health and well-being;

2. Social competence;

3. Emotional maturity;

4. Language and cognitive development; and

5. Communication skills and general knowledge.

Children are deemed vulnerable if they are in the bottom 10th percentile in at least one of the EDI five subscales (domains). This indicates that they have a developmental, social-emotional, cognitive or physical problem that is likely to interfere with their success in school.

The validity and reliability of the EDI continues to be monitored on an ongoing basis. Analyses demonstrate that its structure is robust and its validity meets acceptable psychometric standards (23). The EDI data are collected for individual children, and EDI scores correlate reliably with other similar measures of child development. They are also predictive of later outcomes including academic achievement in primary school. However, the EDI is not designed to be a tool to diagnose individual delays or developmental problems. The EDI’s strength is allowing the aggregation of individual data to the group or community level, which makes it possible to integrate child development outcomes with other sources of data about children, families and communities.

The EDI is used in British Columbia, Manitoba and Ontario with full provincial coverage of children in kindergarten and in many communities in other provinces. In total, EDI data are collected from over 50% of all kindergarten (five-year-old) Canadian children. Furthermore, the EDI has been adapted for use in seven countries outside of Canada (12,24) and projects are ongoing in many others.

The EDI data collected across Canada and elsewhere to date confirm a wide variation between children’s abilities and development before entry to grade 1. Data from EDI match that from the NLSCY, demonstrating that approximately one-quarter of all children are vulnerable when they enter grade 1.

The analysis of Canadian EDI data and additional data on measures of children’s socioeconomic circumstances repeat the gradient effect found in NLSCY data. As indicated in Figure 1, approximately 32% of kindergarten children in the poorest families are vulnerable (that is, scoring in the lowest 10th percentile in at least one of the EDI subscales), while approximately 14% of children in the most affluent families are vulnerable.

COMMUNITY EARLY CHILD DEVELOPMENT REPORTING

Researchers and communities are using EDI data to examine the geography of opportunity for young children by mapping EDI and SES data at provincial, regional and neighbourhood levels (12). The British Columbia Atlas of Child Development (25) is a robust example of analyzing, examining and mapping EDI and SES data from a number of points of view. Examples of specific uses of the British Columbia data show physicians the power and the importance for early child development measurements in the context of the socioeconomic climate in neighbourhoods. The British Columbia Atlas of Child Development presents a visual summary of early child development trends across neighbourhood, school districts and provincial geographies in British Columbia. Colour maps depict information about the many intersecting environments in which British Columbia families live and young children grow, including the socioeconomic, community and policy environments. The Early Child Development Mapping Project at the Human Early Learning Partnerships, which produced the Atlas, aims to understand development in British Columbia neighbourhoods. Academic, government and community partners cooperate to develop neighbourhood-based maps related to EDI data, socioeconomic factors, and community assets and resources.

Between 2000 and 2004, 59 geographical school districts in British Columbia completed the EDI, comprising 93% to 100% of kindergarten children who were five years of age. The Atlas illustrates a visual summary of the impact of community and neighbourhood characteristics on child development and readiness for school. Colour maps layer 2004 EDI results with 2001 Census data to describe the social and economic circumstances of school districts and communities. Figure 2 is a cartogram illustrating EDI results across British Columbia. Each school district retains its approximate geographical shape, but is sized according to the number of children who live there. The side bar listing of the percentage of vulnerable children indicates that vulnerability cuts across British Columbia’s 59 school districts, ranging from 14% in the West Vancouver school district to almost 44% in the Central Coast school district. More than one-half of the school districts have 25% of children or more with a low score on at least one of the developmental domains. The districts with the highest percentage of vulnerable schools tend to be either urban or rural and isolated communities with a high proportion of Aboriginal communities.

Figure 2.

The early development instrument (EDI) in British Columbia’s school districts showing the percentage of vulnerable five-year-old children varying between 14% and 43.7%. SES Socioeconomic status. Adapted from reference 25

Figure 2 shows the relationship between a social index measure and EDI results. In this case, a green background shows areas in which, overall, children did well on EDI assessments; yellow indicates moderate numbers of children with low EDI scores and red indicates large concentrations of vulnerable children. A social risk index is used as the SES measure. The SES measure for each district is represented in a circle that is an overlay on the EDI results. The green circles indicate high SES (and low social risk), yellow circles are moderate SES (and moderate social risk), and the brown circles are low SES (and high social risk).

Vernon and Central Okanagan school districts are circled in red in Figure 2 and enlarged in Figure 3. Both districts have green circles, indicating higher SES. Central Okanagan, while having a higher SES measure, is midrange on the EDI vulnerability scale. Vernon, as a school district, does well on both measures. It has a high SES and low EDI vulnerability. Central Okanagan and Vernon have relatively low social risk but Central Okanagan has a higher proportion of children with developmental difficulties before entry to grade 1.

Figure 3.

Mapping of early development instrument (EDI) data at the neighbourhood levels showing vulnerable children present in all neighbourhoods even when the school district (SD) level of vulnerable children is low. SES Socioeconomic status. Adapted from reference 25

Figure 3 breaks these Vernon and Central Okanagan school districts into neighbourhoods, allowing for a closer examination. Vulnerability is evident across neighbourhoods in Vernon and Central Okanagan, and the general pattern of higher social risk associated with higher rates of vulnerability is evident (eg, Central Kelowna). Vulnerability is more concentrated in higher risk areas but the map shows that vulnerable children are present in all neighbourhoods. The pattern is broken by a few neighbourhoods with low social risk but higher vulnerability (eg, Black Mountain and Peachland) and ones with high social risk but lower vulnerability (eg, Vernon Central). Proportionally, both school districts have an approximately even number of low and high SES neighbourhoods. Central Okanagan has more neighbourhoods with high EDI vulnerability.

Neighbourhood maps in British Columbia help to identify patterns and trends in children’s early development in their immediate local context. In British Columbia cross-sector coalitions use the EDI mapping results to initiate new programming, advocate for policy and funding changes, and monitor the impact of major economic downturns on young children before entry to grade 1.

CONCLUSION

The EDI taps into the key domains of development. It is able to give a summary measure of vulnerability for children in a particular neighbourhood or region, is sensitive to change in community socioeconomic context or developmental opportunities, and predicts school success. Mapping EDI data shows geography and gradients of early child development, highlighting that vulnerabilities, while likely more prevalent in poorer communities, are present across all communities. EDI data help identify vulnerable and resilient communities. It is possible to distinguish SES influences from other influences and to inform strategic planning.

The compelling visuals of EDI data and other information offered by maps can capture wide audiences at community, provincial and federal levels. Hopefully, even though the assault on child poverty and the SES problem may require longer-term solutions, much can be done at the community level to mitigate and strengthen communities.

At the end of the day, mapping will also allow physicians and community partners to identify local needs and to decide on interventions that will help parents and communities create healthy, nurturing environments for young children, so that by six years of age, children are physically, socially and emotionally ready to succeed in school.