Analytical Studies: Methods and References
Identification of Children with Disabilities in the Survey on Early Learning and Child Care Arrangements: Children with Long-term Conditions and Disabilities

Skip to text

Text begins

Acknowledgments

The authors would like to acknowledge the valuable feedback received from experts in child disability and child care who provided input on the Survey on Early Learning and Child Care Arrangements – Children with Long-term Conditions and Disabilities and the analytical approach used in this study.

Abstract

Using data from the Survey on Early Learning and Child Care Arrangements – Children with Long-term Conditions and Disabilities, the present study examined and compared four measures that assessed child difficulties, long-term conditions, activity limitations or disabilities to shed light on the use of non-parental child care among young children with disabilities at a national level. Overall, the results demonstrated the usefulness of integrating the medical and social models by using multiple measures of disability when identifying child disability. While the four measures independently identified children who varied in their general health and need for extra support at their main child care arrangement, the use of child care was largely similar for all children. To identify a group of children who may be at greater risk of participation restrictions in child care, further analyses were conducted on children with activity limitations identified by the Global Activity Limitation Indicator or by at least two of the other three measures (n=1,189): the Disability Screening Questions filter questions, the presence of a long-term condition, or a parent-reported type of disability. While these children were less likely to have excellent or very good general health and more likely to need extra support at their main child care arrangement than the children who were excluded from the final identified sample, the use of child care was similar between both groups of children.

Executive summary

Background: Early learning and child care (ELCC) is particularly important for children with disabilities, as it provides an opportunity to be integrated into social settings while giving parents a range of opportunities and supports, such as participation in the labour force. Children with disabilities around the globe have fewer opportunities to participate in ELCC programs, compared with their peers without disabilities. Relatively little is known about ELCC for children with disabilities in Canada. The lack of data and research on the inclusion of children with disabilities in ELCC is largely because there is no established and consistently used survey measure of disability for children, particularly one that integrates the medical and social models of disability outlined in the International Classification of Functioning, Disability and Health framework.

Objectives and measures: The purpose of the current study was to improve the approaches used to measure child disability in future surveys and to inform inclusivity in ELCC. To this end, four survey measures that capture difficulties, long-term conditions, activity limitations or types of disabilities were examined (independently and concurrently) to explore which ones identified children who may be at greater risk for participation restrictions in child care. To examine child care participation, the measures were compared individually and in combination to determine whether a single item or measure would be sufficient for identifying disability, or whether a combination of measures would better do so. In the absence of an established measure of child disability, it was hypothesized that a single item or measure would not capture the medical and social models of disability and therefore multiple measures would be required. It was further hypothesized that mild long-term conditions, such as asthma or eczema, may be captured by a single item or measure but would not necessarily lead to an activity limitation or alter child care participation. Therefore, the combination of parent-reported disability measures (i.e., an affirmative response to two, three or all four measures) warranted exploration.

The four measures of child disability included

• the Disability Screening Questions filter questions (DSQ filters),
• the presence of a long-term condition
• the Global Activity Limitation Indicator (GALI)
• a parent-reported type of disability.

Data: Of 20,005 parents who were contacted for the Survey on Early Learning and Child Care Arrangements – Children with Long-term Conditions and Disabilities (SELCCA-CLCD), about 62% (n=12,411) responded to the survey. Of these parents, about 16% (n=2,016) provided an affirmative response to at least one of the four measures. Hence, the final sample included 2,016 children across the 10 provinces of Canada who were younger than 6 years as of June 30, 2023, and who may have one or more long-term conditions or disabilities.

Results: About 40% of children were identified as having a long-term condition or disability by a single measure (e.g., 569 children were identified solely by the DSQ filters). Compared with the children identified by a single measure, the children identified by multiple measures were less likely to have excellent or very good general health (72% versus 89%) and more likely to need extra support at their main child care arrangement (68% versus 33%). About 15% of children were identified by all four measures. The most frequently reported long-term conditions among children identified by all four measures were related to mental, behavioural or neurodevelopmental disorders and to developmental anomalies.

A majority (86%) of the sample was identified by the DSQ filters. As the DSQ filters aim to identify children who are likely to have a disability, further analysis exploring whether these children could be confidently identified as having one or more long-term conditions or disabilities was needed. Notably, because the DSQ filters were developed for an adult population aged 15 years and older and relatively little is known about their suitability for a child population, further analysis considering other measures was essential to improve the measurement of child disability in the SELCCA-CLCD.

Compared with children identified solely by the DSQ filters, children identified by the DSQ filters and other measures were less likely to have excellent or very good general health (71% versus 90%) and more likely to need extra support at their main child care arrangement (65% versus 32%). Among children identified solely by the DSQ filters, about 58% had learning, remembering or concentrating difficulties, which could be part of their normative growth and development. These results suggest that children identified solely by the DSQ filters may need to be excluded from further analyses to focus on those more likely to have participation restrictions in child care.

Similarly, children identified by the presence of a long-term condition and other measures were less likely to have excellent or very good general health (70%) and more likely to need extra support at their main child care arrangement (68%), compared with children identified solely by a long-term condition (90% and 37%, respectively). The three conditions most frequently reported among children identified solely by the presence of a long-term condition appeared to be somewhat less complex (less severe), including conditions related to diseases of the respiratory system, diseases of the immune system and developmental anomalies, which may not necessarily lead to an activity limitation or participation restriction. Based on these results, children identified solely by the presence of a long-term condition were excluded from the final identified sample of children with disabilities.

Children identified solely by the presence of a parent-reported type of disability had a similar pattern of results and were also excluded from the final identified sample of children with disabilities.

Just under one-quarter of the sample (23%) was identified by the GALI, a participation restriction measure, while a large majority (over 95%) was also identified by another measure. Children identified by the GALI appeared to be least likely to be in excellent or very good health (59%) and most likely to need extra support at their main child care arrangement (84%).

The final identified sample of children with disabilities (n=1,189) included children who were identified by the GALI or at least two of the other three measures, including the DSQ filters, the presence of a long-term condition and the presence of a parent-reported type of disability. In other words, children identified solely by a single measure (except for the GALI) were excluded from the final identified sample of children with disabilities.

In terms of child care participation, the use of child care did not differ among the groups of children identified by each of the four measures individually, between those identified by a single measure and those identified by multiple measures, or between the full sample and the final identified sample.

Conclusions: The DSQ filters identified a broad group of children with difficulties and long-term conditions that may not be disabilities. Given some unexpected findings, further conceptual and methodological work on the DSQ filters is needed to ensure their suitability for a child population. For example, parents who reported their child as having any other health problem or long-term condition on the sixth DSQ filter did not report the presence of a long-term condition.

In contrast, the GALI identified the lowest proportion of children, with the majority needing extra support at their main child care arrangement. However, the GALI is not a measure of disability per se but a global measure of participation restriction. Future child surveys may consider including the GALI to further assess its utility in identifying children with disabilities in comparison with other measures, such as the Child Functioning Module in the Canadian Health Survey on Children and Youth.

The parent-reported type of disability was a useful measure for identifying a considerable proportion of children with communication disabilities. This information could inform potential conceptual work on the DSQ filters. However, caution is needed to avoid falsely identifying children with difficulties that are part of their normative growth and development.

Finally, according to the medical model of disability, the presence of a long-term condition must be considered when identifying child disability. However, the presence of a long-term condition alone may not lead to activity limitations or participation restrictions.

In the absence of a national survey measure of disability for children, using multiple measures and integrating medical and social models may address the challenge of identifying children with disabilities to inform inclusivity in ELCC via national surveys such as the SELCCA-CLCD. Relying solely on the presence of a long-term condition (medical model) or limitations in daily activities (social model) could exclude children experiencing participation restrictions in child care. The present study provides preliminary support for the idea that integrating the medical and social models can be achieved by using multiple existing measures.

1 Introduction

The significance of early learning and child care (ELCC) in the promotion of children’s development and well-being is well established (United Nations Educational, Scientific and Cultural Organization, 2021), such that access to high-quality early childhood care and education for all children is a target monitored under Goal 4 of the United Nations’ 2030 Agenda for Sustainable Development (United Nations, n.d.a). ELCC is particularly crucial for children with disabilities, as it provides an opportunity for participation and integration into social settings while giving parents a range of opportunities and supports, such as participation in the labour force (Halfon & Friendly, 2013). The Convention on the Rights of Persons with Disabilities (CRPD) (United Nations, n.d.b) and the Convention on the Rights of the Child (United Nations Children’s Fund [UNICEF], n.d.)  provide guidance to address barriers and achieve inclusion in ELCC for children with disabilities. Yet, children with disabilities continue to have fewer opportunities to participate in ELCC programs, compared with their peers without disabilities (World Health Organization [WHO] & UNICEF, 2023). Globally, it is estimated that 7.5% of children younger than 5 years have developmental disabilities^Note  (WHO & UNICEF, 2023), and children with disabilities around the globe are 25% less likely to attend early childhood education than children without disabilities (UNICEF, 2021). The present study attempts to shed light on the use of non-parental child care among young children with disabilities in Canada, with a primary focus on the utility of four measures to identify child disability in the context of child care.

In Canada, the Multilateral Early Learning and Child Care Framework (Employment and Social Development Canada [ESDC], 2017) provides a long-term vision for governments to commit to high-quality ELCC that supports children’s development and lifelong well-being. This framework recognizes diversity among children in Canada, including children who experience barriers to child care participation, such as children from low-income families or children with varying abilities, and strives to increase access to ELCC for all children. Therefore, one of the areas of investment for governments is the provision of additional supports in ELCC settings to facilitate the participation of children with varying abilities (ESDC, 2017). For this reason, it is critical that researchers explore ways to identify children with disabilities, focusing particularly on those who may experience barriers to accessing child care or need extra support at their main child care arrangement.

Despite governments’ commitment to the Canada-wide ELCC initiative’s inclusion aspect (Childcare Resource and Research Unit, 2022), data and research challenges remain in reporting and monitoring progress. A data gap on children with disabilities at the national level (Arim et al., 2016) has led to a lack of data and research that could inform inclusion of children with disabilities in ELCC (Halfon & Friendly, 2013). The last national survey on children with disabilities, the Participation and Activity Limitation Survey (PALS), was conducted in 2006. The findings indicated that 29% of parents who had a child younger than 15 years with an activity limitation used some type of non-parental child care (Kowalchuk, 2008). In addition, parents of children with mild to moderate disabilities (25%) were less likely to use non-parental child care than parents of children with more severe disabilities (34%). In 2012, the survey was renamed the Canadian Survey on Disability and excluded children younger than 15 years. More recently, the 2019 Canadian Health Survey on Children and Youth (CHSCY) included information on child care use, along with child functional difficulties identified by the Washington Group / UNICEF Child Functioning Module (CFM) (UNICEF, 2017). Several data limitations were observed, including the inability to identify children younger than 2 years with functional difficulties (because the CFM is appropriate for children aged 2 years and older) and report results for children with functional difficulties who were regularly in child care disaggregated at the provincial and territorial level because of small sample size.^Note 

The lack of national data on the inclusion of children with disabilities in ELCC could also be partly explained by data gaps and differences in provincial and territorial jurisdictions, where the responsibility for ELCC data collection lies. A recent extensive review of the literature has identified differences in administrative data collection at the provincial and territorial level as contributing to data gaps, specifically that “the provinces and territories^Note  do not collect the same data at the same time or in the same way, if they collect data at all” (p. 6, Philpott et al., 2019).

The Survey on Early Learning and Child Care Arrangements – Children with Long-term Conditions and Disabilities (SELCCA-CLCD), sponsored by ESDC, was designed in consultation with external subject-matter experts to gather information from parents and guardians across the 10 provinces of Canada on ELCC arrangements for children younger than 6 years who may have one or more long-term conditions or disabilities. The objective of the SELCCA-CLCD is to provide baseline data to facilitate advances in research and informed policy development on inclusive ELCC for children with long-term conditions or disabilities. Given the lack of a national survey measure of disability for children, different measures of long-term conditions or disabilities that were previously used for child populations were included on the SELCCA-CLCD.

Despite data and measurement challenges in identifying children with disabilities, decades of research on universal child care (Friendly, 1994) or inclusive child care (Irwin et al., 2000) exist in Canada, and the main conclusion still holds. “There is a long way to go before children with disabilities have the same opportunities to attend quality child care as do other children, with accommodations and adaptations that meet their unique needs” (p. 91, Irwin & Lero, 2020). At the provincial level, Quebec is often distinguished from other Canadian provinces by its reduced-contribution child care program for children aged 0 to 5 years, which has been in place since 1997 through a network of subsidized child care services (Dionne et al., 2023). Yet, a recent qualitative study in Quebec showed that children with disabilities continue to face barriers to accessing early childhood education and care settings (Beaudoin et al., 2022). This finding is particularly important given that the results from the Québec Survey of Child Development in Kindergarten showed that the proportion of children in 5-year-old kindergarten who are considered vulnerable in at least one developmental domain^Note  has been increasing over the past 10 years, from 25.6% in 2012 to 27.7% in 2017 and then to 28.7% in 2022 (Institut de la statistique du Québec, 2023). However, a major limitation of this body of work is the lack of a consistent measure of child disability, limiting the comparability and use of the results to inform inclusive child care.

To date, most studies on child disability in the ELCC context have focused on a single condition, such as autism spectrum disorder (Maich et al., 2019), or a single type of disability, such as hearing impairment, including deafness and hard of hearing (Underwood & Snoddon, 2021), despite a shift from a categorical (condition-specific) approach to a non-categorical (inclusive) approach (Stein & Jessop, 1989) in policy-relevant research on childhood health problems or disabilities. This shift occurred because many of the consequences (e.g., participation restriction) of childhood health problems or disabilities are independent of the specific condition, as these children experience common challenges and barriers to participation in society (Stein et al., 1993). In line with this shift, the present study uses a non-categorical approach to identify children with disabilities by focusing on the consequences of diverse long-term conditions or disabilities and thus recognizing commonalities in needs. However, caution is needed to ensure that the measure does not falsely identify difficulties that are part of normative growth and development. Despite the improved utility of using a non-categorical approach, the methodological challenge of identifying children with disabilities remains.

Commonly used measures to identify disability in survey data vary in the degree to which they represent a medical or a social model of disability. While the medical model treats disability as a medical problem that requires person-level intervention (e.g., rehabilitation), the social model focuses on the person–environment interaction and recognizes disability as a social problem where environmental barriers need to be eliminated to ensure equal opportunities for social participation (see Arim et al., 2016). Survey measures therefore often range from checklists of chronic conditions to questions on activity limitations or functional difficulties. However, each of these measures has limitations (Arim et al., 2016). For example, checklists may not include some rare conditions and therefore exclude some children from the group of interest (a limitation from a medical model perspective). In addition, checklists are not designed to consider how children may interact with their environment (a limitation from a social model perspective). The challenge in identifying children with disabilities is a measurement issue related to the necessary but complicated integration of the medical and social models, as depicted in the International Classification of Functioning, Disability and Health (ICF) (WHO, 2002).

Disability is a complex and evolving concept, and its measurement has been purpose-specific. For example, disability measurement may differ when the goal is clinical practice oriented (e.g., who will receive a health care service) as opposed to policy oriented (e.g., who will receive a social program or benefit). Given this, the goal of disability measurement should always be established a priori. As outlined in the ICF framework (WHO, 2002), disability can be an impairment in bodily function or structure (e.g., problems with using the muscles), a limitation in activity (e.g., inability to see), or a restriction in participation (e.g., exclusion from participation in child care). Thus, neither the medical model nor the social model alone is adequate to represent disability. A more appropriate model of disability, the biopsychosocial model (WHO, 2002), integrates both views without rejecting the usefulness of each. This comprehensive concept of disability emphasizes individual functioning and factors that could improve it, suggesting a measurement that is consistent with a non-categorical approach. The present study uses the SELCCA-CLCD to demonstrate the usefulness of integrating medical and social models of disability through measures that vary in the degree to which they represent each model of disability.

Relatively little is known about ELCC for children with disabilities in Canada. Existing studies have several important limitations that offer future directions for addressing data and information gaps. Few studies provide information at the national level. Even at the provincial and territorial level, challenges remain regarding consistent data collection that yields information on inclusion in ELCC. Given the different measures and data sources (Russell et al., 2023), concepts and definitions of child disability vary (Arim et al., 2016).

The present study aims to address these limitations by using SELCCA-CLCD data to compare four measures that identify difficulties, long-term conditions, activity limitations or disabilities (hereinafter called disability) and shed light on their utility in identifying children with disabilities who may need extra support in ELCC contexts. The focus on the need for extra support is intentional, as children with disabilities who need extra support in their child care arrangement may be more likely to experience participation restrictions. Overall, filling information gaps on access to, use of and barriers to ELCC is critical to understanding the use of non-parental child care among children with disabilities at a national level, informing the accessibility pillar of the Multilateral Early Learning and Child Care Framework (ESDC, 2017) and providing useful insights for improving the measurement of child disability in future surveys.

The study answers the following research questions:

1. What proportions of children are identified as having a long-term condition or a disability by each of the four measures? Are there sociodemographic differences among the various groups of children? Are they associated with differences in their need for extra support at their main child care arrangement?
2. Do these four measures overlap (i.e., how many children are identified by all four measures)? Do they differ (i.e., how many children are identified only by a single measure)? Are any children systematically included in or excluded from any of the four measures (e.g., are children with a specific type of disability, such as a communication or other type of disability, mostly excluded from one of the measures used in the survey)?
3. What proportions of children identified by each measure are in child care? What are the associations between these measures and the use of child care after considering children’s individual and family characteristics?
4. Given the varying scope of the four measures, is there a subsample of children who may be at greater risk for participation restrictions in child care (e.g., more likely to have poorer general health and need extra support at their main child care arrangement)? Which measures best capture this subsample of children?

For methodological purposes and given that children in the study sample could have one or more long-term conditions or disabilities, it is important to examine the questions under (1) and (2), as they explore the utility of each measure in identifying children with disabilities. Without a single standardized measure of disability for children, the disability concept and how it is measured in the survey are important to define. The answers to the questions under (3) aim to clarify restrictions in child care participation. This is not only a primary focus of this study but also essential for conceptual and methodological reasons. Finally, there is a well-established literature on the social, economic and health characteristics of children with disabilities. For example, when compared with healthy children, children with disabilities are more likely to be older, have poorer general health, live in households with lower income or with parents who have a lower educational attainment, and encounter barriers that restrict participation (Arim et al., 2012; Irwin & Lero, 2020; Kohen et al., 2007). This knowledge helped address the questions under (4) to determine which measures best capture the subsample of children who may be at greater risk for participation restrictions in child care.

Section 2 discusses the data, key measures and methods. The results are presented in Section 3, and conclusions are provided in Section 4.

2 Data, measures and methods

Data

This study used data (N=2,016) from the 2023 SELCCA-CLCD, which was collected between April 20 and June 30, 2023. The SELCCA-CLCD is not a representative survey of children with disabilities.^Note  However, the survey collects information from parents and guardians (those knowledgeable about the child care arrangements, hereinafter called “parents”) of children aged 0 to 5 years who may have one or more long-term conditions or disabilities. The survey asks parents about their usual child care arrangement (child care use); the associated expenses, barriers or difficulties they may have encountered when looking for or accessing care; and their preferences for child care.

The sampling frame of the SELCCA-CLCD was based on two sources: the 2021 Census (to include children aged 3 to 5 years) and the Canada child benefit (CCB) (to include children aged 0 to 2 years who were not born at the time of the 2021 Census). In total, 20,005 parents of children younger than 6 years (as of June 30, 2023) across the 10 Canadian provinces were contacted for the survey. About one-third of the target population was derived from the 2021 Census (n=6,611) and the remainder (67%) was from the CCB (n=13,394). The response rate was 62%, with 12,411 parents completing the first portion of the survey. Of these, 10,395 (84%) were considered out of scope for the remainder of the survey because they did not provide an affirmative response to any of the questions identifying long-term conditions or disabilities.^Note  Thus, the final study sample included 2,016 respondents (16%) who were in scope. This percentage is comparable with recent statistics from the United States on children with functional difficulties and children with special health care needs. Specifically, in the 2022 National Survey of Children’s Health, 22% of children younger than 6 years were identified as having one or more functional difficulties and 11% were identified as having special health care needs (Child and Adolescent Health Measurement Initiative, n.d.). Figure 1 presents the sampling and target population.

Measures and methods

Child disability measures

Four measures of disabilities as reported by parents were included in the 2023 SELCCA-CLCD: (1) the Disability Screening Questions filter questions (DSQ filters), which are also asked on the Census of Population to identify people who are likely to have a disability; (2) the presence of a long-term condition that has lasted or is expected to last for six months or more; (3) the Global Activity Limitation Indicator (GALI), which assesses participation restriction; and (4) the presence of a type of disability, where different types of disabilities can be selected. These measures were agreed upon by a group of experts with knowledge and expertise in child disability and child care during the development of the survey content. Each measure, along with the identification criteria and the proportions of children identified in each response category of each measure, is presented in Appendix A.

The DSQ filters are derived from the DSQ, which are a disability survey measure based on a social model (Grondin, 2016). The DSQ filters (six items) were first introduced in the 2016 Census long-form questionnaire to identify people who reported having difficulties doing certain activities or long-term conditions lasting or expecting to last for six months or more. People with affirmative responses were considered to have a greater likelihood of having a disability. However, because the DSQ were developed for an adult population aged 15 years and older, relatively little is known about their suitability for a child population (Charters et al., 2022).

To date, two studies have examined the sociodemographic profile of children aged 0 to 14 years with an affirmative response to the DSQ filters using data from the 2016 (Charters et al., 2022) and 2021 (Charters et al., 2024) censuses. While both studies corroborate the suitability of the DSQ filters for identifying children who are likely to have a disability, some unexpected findings were noted. These warrant further research and analysis to assess the suitability of the DSQ filters for generating an appropriate sampling frame for prospective child disability surveys. The present study is the first to compare the DSQ filters with three other child disability measures using data from a single survey, allowing for direct comparison within the same group of respondents. A comparison of child disability indicators based on different models of disability is key to better understanding whether different indicators identify different groups of children, and whether and how these groups differ (see Kohen et al. [2007] for an example). This information has implications for considering and refining various disability concepts and definitions based on outcomes of interest.

The second measure, which is the presence of any long-term conditions that have lasted or are expected to last for six months or more (hereinafter called long-term conditions), represents a medical model of disability and has been commonly used in surveys, including child surveys (see the 2023 CHSCY for a recent example). In most surveys, the question is presented with a checklist (see the 1994/1995 National Longitudinal Survey of Children and Youth [NLSCY]) or a write-in option (see the 2006 PALS). In the SELCCA-CLCD, the single question was followed by a write-in option, enabling the coding of the conditions with the 11th version of the International Classification of Diseases (WHO, 2022).

The third measure, the GALI, which consists of two items, was developed as part of the instruments used to monitor mortality and the different facets of health in European surveys (Robine et al., 2003). It has also been used to support the European Union’s implementation of the CRPD (European Commission, 2017). A systematic review showed that, as a global measure on participation restriction, the GALI has good psychometric properties (Van Oyen et al., 2018) and can be used to evaluate general health status, disability,^Note  and related inequalities and health care needs at the population level (Eurostat, 2016). The GALI was adapted for children and included in the children’s health module of the 2017 European Union statistics on income and living conditions (Bogaert et al., 2018). The GALI measures participation restriction through a long-standing limitation (and its severity) in the child’s activities because of health problems (Eurostat, 2017). The GALI can be considered a measure that integrates the medical and social models of disability to some extent, given that it asks about the presence of a health problem, but follows up with a question on whether activities have been limited for at least six months. It also recommends that a general health question be included for children (see Appendix A). While this question is not considered part of the GALI, it can be used to determine the health status of children (see the section on children’s general health). To the authors’ knowledge, the SELCCA-CLCD is the first child survey in Canada to include the GALI, although similar questions have been previously asked in the NLSCY (e.g., activity limitations at home and school).

The fourth measure (hereinafter called type of disability) is a single question that asks parents to report the presence of a disability and provides a list of different types of disabilities, including examples of related conditions (e.g., autism for developmental disability). Recent Statistics Canada surveys, such as the Impacts of COVID-19 on Canadians – Parenting During the Pandemic, have included similar items (see Arim et al., 2020). Given the wording, this question may represent a medical model; however, it may also represent a social model as it is based on respondents’ perceptions and interpretations of having a disability. The present study clarified this ambiguity by examining the overlap between this measure (i.e., type of disability) and the other measures, specifically the DSQ filter questions, which represent a social model of disability.

An additional derived variable that determines whether a disability was identified based on a single measure or multiple measures was created. Children who met the identification criteria for two or more of the four disability measures (e.g., long-term condition and the GALI) were considered to have been identified by multiple measures, while children who met the identification criteria for only one of the measures were considered to have been identified by a single measure (e.g., a long-term condition but not the DSQ filters, the GALI or the type of disability).

Children’s child care participation, need for extra support and general health

The use of any non-parental child care arrangement in the past three months (yes or no) was the main variable of interest to examine the participation of children with disabilities in child care. Children’s need for extra support (yes or no) at their main child care arrangement (i.e., managing behaviour, learning, communicating, interacting socially, using motor skills, performing self-care activities, addressing nutritional or dietary concerns, or other), including support with medical needs, was considered in the analysis. Specifically, a derived variable (yes or no) was created based on the presence of any need for additional or medical support. This derived variable was important in relation to participation restriction, particularly if children identified by the four measures of child disability experienced differences in need for extra support at their main child care arrangement. Finally, a binary variable was created to classify children’s general health, with responses of “excellent” and “very good” distinguished from responses of “good,” “fair” and “poor.”

Other child, parent and family characteristics

Data on the child’s age group, the parent’s main activity (working at a paid job or business during the past three months or otherwise), the parent’s educational attainment, the parent’s marital status (based on the 2023 CCB file) and household income (above or below the low-income measure threshold based on tax-reported data from the 2021 T1 Family File) were available^Note  and included in the analysis. These sociodemographic characteristics are known to be associated with the presence of childhood health problems and disabilities (Arim et al., 2012) and may also be associated with child care use.

Analytical plan

First, descriptive statistics were produced (separately for the full sample and for the children identified by each of the four measures of child disability) on the use of child care arrangements; the child’s need for extra support at their main child care arrangement; the child’s general health; and other child, parent and family characteristics. These characteristics were also examined for those identified by a single measure and for those identified by multiple measures. The findings were interpreted to evaluate the utility of the four measures based on whether the observed proportions are meaningful and consistent with the established literature (e.g., each measure is expected to identify more older than younger children). To examine the differences between groups of children identified by each of the four measures, including those identified by single or multiple measures, 95% confidence levels and results from comparison tests (i.e., chi-square tests) were used. Given the number of comparison tests, a conservative p-value (p < 0.001) was used to determine statistical significance in observed differences. In all analyses, person and bootstrap weights were used to account for the complex survey design.

Next, a Venn diagram was used to examine the overlap among the four measures that identified children with disabilities. If the four measures assess distinct constructs and therefore identify distinct samples of children, the circles in the diagram should show minimal overlap. Conversely, if they measure the same construct with varying degrees of precision, the diagram would display overlapping circles. The most conservative measure (smallest in size) would be fully encompassed by the next most conservative measure, while the least conservative measure (largest in size) would include all three other measures. If the four measures do not differ in their identification of children with disabilities, they should be nested within each other. There should also be a pattern in the size of the circles, with the most conservative measure identifying the children with the poorest health (smallest group) and the least conservative measure identifying those with the best health (largest group), on average.

Finally, multiple logistic regression analyses with marginal effects were conducted to examine associations between the four measures and the use of non-parental child care before and after considering other child, parent and family characteristics. A separate multiple logistic regression analysis was performed using a single measure, compared with multiple measures. The analytical plan was extended from a preliminary version of the analysis to include an in-depth analysis on the four measures. This was necessary to provide useful information to researchers identifying children with disabilities using the measures from the SELCCA-CLCD data and to inform future data collection activities.

3 Results

Descriptive statistics

Individual measures of child disability

Table 1 presents the descriptive statistics of the full sample and the proportions of children identified by each of the four measures of child disability (i.e., the DSQ filters, a long-term condition, the GALI and the type of disability). Overall, about 70% of the full sample were children aged 3 to 5 years, with the remainder being younger than 3. Approximately 8 in 10 children had excellent or very good general health, as reported by parents. About 61% were in child care and, among those who were in child care, just over half (54%) needed extra support at their main child care arrangement. Regarding parent and family characteristics, about 61% of parents were working at a paid job or business, and more than three-quarters (79%) had at least a high school education. Finally, a majority (81%) of children were living in two-parent families, and about 22% were living in low-income families.

About 86% of children in the full sample were identified by the DSQ filters (n=1,711), which was the measure that identified the greatest number of children. Of these children, about 19% had sensory (seeing or hearing) difficulties; 16% had physical difficulties; 55% had learning, remembering or concentrating difficulties; 27% had emotional, psychological or mental health conditions; and 48% had other health problems or long-term conditions. Just under half (49.1%) of the children in the full sample were identified as having a long-term condition (n=1,017). The three most frequently reported long-term conditions were related to mental, behavioural or neurodevelopmental disorders (e.g., attention deficit hyperactivity disorder); diseases of the respiratory system (e.g., asthma); and developmental anomalies (e.g., structural developmental anomaly of the heart). Compared with children identified by the DSQ filters, children identified as having a long-term condition were less likely to have excellent or very good general health (72% versus 78%) and more likely to need extra support at their main child care arrangement (65% versus 57%). These findings are unsurprising given that the DSQ filters aim to identify a broader group of children with difficulties and long-term conditions.

The proportion of children in the full sample identified by the GALI (23%, n=419) was the smallest of all the measures. The children in this group were the least likely of all four groups to have excellent or very good general health (59%) and the most likely to need extra support at their main child care arrangement (84%).

Finally, about 47% of children in the full sample were identified by their parent as having at least one type of disability (n=835). Of these children, about 8% had a seeing disability; 8% had a hearing disability; 10% had a physical disability; 30% had a learning, behaviour or emotional disability; 29% had a developmental disability; 62% had a communication disability; and 9% had another type of disability. Compared with the children identified by the DSQ filters, these children were less likely to have excellent or very good general health (70% versus 78%) and more likely to require support at their main child care arrangement (70% versus 57%). Compared with the children identified as having a long-term condition, these children were more likely to need extra support (70% versus 65%).

Notably, the use of child care was largely similar among the groups of children identified by each of the four measures. Overall, about 6 in 10 children participated in child care regardless of the measure.

In terms of sociodemographic characteristics, the proportion of older children (aged 3 to 5 years) was slightly higher among those identified by the GALI or as having a type of disability, compared with children identified by the DSQ filters or as having a long-term condition. Few differences were noted in the parent and family characteristics. For example, a higher proportion (83%) of parents of children identified as having a long-term condition had at least a high school education.

Overlap between measures of child disability

Figure 2 depicts the number of children with disabilities identified by each of the four measures. The overlapping areas of the ovals indicate the number of children identified by more than one measure. The size of the oval is proportional to the number of children identified by the measure. The total area of all four circles represents the full sample of 2,016 children.

About 40% of children in the full sample were identified by only one of the four measures. More specifically, about 28% of children (n=569) were identified by the DSQ filters only, 5% (n=114) by a long-term condition only, few (suppressed) by the GALI only and 7% (n=144) by the type of disability only.

Among children identified solely by the DSQ filters, about 11% had an affirmative response to the filter question related to physical difficulties; 19% to that related to sensory (seeing and hearing) difficulties; 58% to that related to learning, remembering or concentrating difficulties; 21% to that related to emotional, psychological or mental health conditions; and 12% to that for other health problems or long-term conditions. Of note, those who had an affirmative response for any other health problems or long-term conditions (the sixth DSQ filter item) were not identified as having a long-term condition.

Among children identified solely by a long-term condition, the three most frequently reported conditions were related to diseases of the respiratory system (e.g., asthma), diseases of the immune system (e.g., food hypersensitivity) and developmental anomalies (e.g., structural developmental anomaly of the heart). Because conditions related to mental, behavioural or neurodevelopmental disorders (e.g., attention deficit hyperactivity disorder) were excluded, while conditions related to diseases of the immune system were included in the three most frequently reported conditions, children identified solely by a long-term condition may have less complex (less severe) conditions (i.e., conditions that do not result in activity limitations).

Among the children identified solely by the type of disability, about 76% had a communication disability.^Note  This finding regarding communication disabilities, which include being understood by and understanding others, warrants further research. While it highlights the importance of considering communication difficulties (e.g., speech delay) in young children, this question may falsely identify young children who are experiencing normative development.

Children identified by only one of the four measures were classified as identified by a single measure. The remaining children in the full sample, approximately 60%, were identified by two or more measures. These children were classified as identified by multiple measures. The fact that this group included more than half of the sample suggests considerable overlap between these measures.

About 15% of children (n=258) were identified by all four measures. Among these children, the three most frequently reported long-term conditions were related to mental, behavioural or neurodevelopmental disorders (e.g., attention deficit hyperactivity disorder); developmental anomalies (e.g., structural developmental anomaly of the heart); and symptoms, signs or clinical findings (e.g., visual difficulties). This finding suggests that, as expected, children identified by all four measures may have more complex (more severe) conditions.

Compared with children in the full sample, children identified by all four measures were more likely to be older (aged 3 to 5 years) (79%) and less likely to have excellent or very good general health (55%). About 60% of these children were in child care, a figure that was similar to that of the full sample (61%). However, among those who were in child care, 90% needed extra support at their main child care arrangement, a figure much higher than that observed in the full sample (54%). These findings suggest that children identified by all four measures may be at greater risk for participation restrictions in child care.

The DSQ filters had the highest overlap with a long-term condition (about 40% of children in the full sample), followed by the type of disability (about 30%) and the GALI (about 20%). By comparison, the GALI had a slightly lower overlap with a long-term condition and the type of disability, representing about 15% to 20% of children in the full sample. The overlap between the type of disability and a long-term condition also represented about 20% of children in the full sample. Lastly, the type of disability had a higher overlap with the DSQ filters than with a long-term condition, suggesting that parents’ responses to the type of disability are more aligned with their responses to the DSQ filters (social model) than with their responses to the presence of a long-term condition (medical model). However, caution is warranted as the differences in overlap may simply result from the size of the sample identified by each measure. For example, the greater the sample identified by the DSQ filters, the greater the likelihood of overlap with another measure.

An additional analysis focused on the overlap between each DSQ filter question and the corresponding type of disability (e.g., the DSQ filter for difficulty seeing and the type of seeing disability). While there was a high agreement for seeing (95%), hearing (92%) and physical (89%) types of disabilities, a lower agreement was observed for learning, behaviour or emotional disabilities (58%). As expected, the proportion of children identified by the DSQ filters but not by the type of disability (49%) was much higher than the proportion of children identified by the type of disability but not the DSQ filters (9%).

Table 2 presents the characteristics of children identified by multiple measures, compared with those of children identified by a single measure. Children identified by multiple measures were significantly less likely (72%) to have excellent or very good general health compared with children identified by a single measure (89%). Moreover, among children identified by multiple measures who were in child care (68%), the proportion who needed extra support at their main child care arrangement was more than double that of their counterparts identified by a single measure (33%). The SELCCA-CLCD did not identify the specific type of support needed, and the proportion of children in the general population needing extra support is unknown. There were no other differences in any child, parent or family characteristics between children identified by multiple measures and those identified by a single measure. Overall, these results highlight the utility of incorporating multiple measures to identify children with disabilities.

Multiple logistic regression analyses with marginal effects

The final set of analyses examined the associations between the child disability measures and the use of non-parental child care, before and after considering other factors known to be associated with child disability and child care participation. In Table 3, results from Model 1 show that there was no association between the use of non-parental child care and each of the four child disability measures. This suggests that child care use did not differ among children identified by the four measures. After considering the effects of other child, parent and family socioeconomic characteristics, Model 2 shows that children identified by the DSQ filters were 8 percentage points more likely to participate in non-parental child care than those identified by any of the other three measures, although this difference was not statistically significant. The effects of the socioeconomic characteristics were in the expected directions. For example, after controlling for other factors, children of parents who worked at a paid job or business were 29 percentage points more likely to be in non-parental child care than children without a parent working. These results suggest that the use of child care did not differ among children identified by the four measures, even after controlling for socioeconomic characteristics.^Note 

In Table 4, results from Model 1 also indicate no association between the use of non-parental child care and child disability measures, as classified by single versus multiple measures. This suggests that the use of child care did not differ between children identified by a single measure, compared with those identified by multiple measures. The results remained the same in Model 2, after controlling for other child, parent and family socioeconomic characteristics. The effects of socioeconomic characteristics were in the expected direction, the same as in the results in Table 3. Overall, there were no differences in non-parental child care use between children identified by multiple disability measures and those identified by a single measure.

In-depth analyses on child disability measures

Given that a majority (86%, n=1,711) of the full sample was identified by the DSQ filters, and that these questions aim to identify children with a greater likelihood of having a disability, additional analyses were conducted on the group of children identified by the DSQ filters. These in-depth analyses explored whether the characteristics of children identified solely by the DSQ filters (n=569) differed from those of children identified by the DSQ filters and at least one other measure (n=1,142), with a particular focus on general health, use of child care and need for extra support at the main child care arrangement. Children identified by the DSQ filters and no other measure were expected to be more likely to have excellent or very good general health and less likely to need extra support at their main child care arrangement. This additional set of analyses was repeated for each of the measures, except for the GALI, because of low sample size, to determine whether a similar or different pattern of results existed for the other measures (Table 5).

As expected, compared with children identified solely by the DSQ filters, children identified by the DSQ filters and at least one other measure were less likely to be in excellent or very good general health (71% versus 90%) and, among those who were in child care, were more likely to need extra support at their main child care arrangement (69% versus 32%). Proportions of children in child care were similar across the two groups (Table 5). These results suggest that excluding children identified solely by the DSQ filters could provide a more meaningful sample of children with disabilities who may be more likely to have participation restrictions in child care.

Similarly, about 9 in 10 children identified solely by a long-term condition had excellent or very good general health and, among those in child care, over one-third (37%) needed extra support at their main child care arrangement. By comparison, 70% of those identified by a long-term condition and at least one other measure had excellent or very good general health, while 68% of those in child care needed extra support. The proportion of children in child care was lower (55%, compared with 63%), though this difference was not statistically significant. Because the three most frequently reported long-term conditions were related to diseases of the respiratory system (e.g., asthma), diseases of the immune system (e.g., food hypersensitivity) and developmental anomalies (e.g., structural developmental anomaly of the heart), excluding these children from the sample of children with disabilities may be appropriate, as the presence of a long-term condition may not, in and of itself, lead to an activity limitation or participation restriction.

Results for children identified solely by a parent-reported type of disability) had a similar pattern as those for children identified solely by the DSQ filters or a long-term condition (see Table 5). Therefore, the children identified solely by a single measure, except those few children identified solely by the GALI (who were shown to be least likely to have excellent or very good general health and most likely to need extra support at their main child care arrangement) were excluded from a final set of analyses conducted to examine the characteristics of children with disabilities. This final group of children (hereinafter called the final identified sample) included children identified solely by the GALI or by at least two of the three other measures (i.e., the DSQ filters, a long-term condition or the type of disability) (n=1,189, 59% of the full sample).

Compared with children who were excluded from the final identified sample (i.e., those identified solely by the DSQ filters, a long-term condition or the type of disability; see the full sample in Table 5), children in the final identified sample were less likely to have excellent or very good general health (72% versus 90%) and, among those in child care, more likely to need extra support at their main child care arrangement (68% versus 32%).^Note  The use of child care was similar between the two groups (62% versus 59%). All other child, parent and family characteristics were largely similar between the two groups. One additional variable of interest related to child care participation was considered. Among the children whose parents had difficulty finding child care, those in the final identified sample (24%) were more likely to have a parent who reported difficulty finding a child care arrangement that met their child’s special needs, compared with children excluded from the final identified sample (2%).^Note  Overall, these in-depth analyses^Note  demonstrated that it was possible to further disaggregate children with disabilities to identify those who were less likely to have excellent or very good general health and more likely to need extra support at their main child care arrangement.

4 Conclusion

The SELCCA-CLCD provided an opportunity to examine and compare four measures for identifying long-term conditions or disabilities in children that could be associated with participation restrictions in non-parental child care. The four measures considered in the study, which vary in the degree to which they represent the medical and social models of disability, are the DSQ filters, the presence of a long-term condition, the GALI, and a parent-reported type of disability.

Most children in the full sample were identified by the DSQ filters. More than half of them were identified by the DSQ filter question related to learning, remembering or concentrating difficulties (55%), and about half were identified by the DSQ filter question related to other health problems or long-term conditions (48%). Among the subsample of children identified solely by the DSQ filters, about 6 in 10 were identified by the filter question related to learning, remembering or concentrating difficulties. The high proportion of children with such difficulties was in line with previous research showing these difficulties to be the most prevalent type in children younger than 15 years of age, according to Canadian census data (Charters et al., 2022; 2024). The fact that these children were not identified by any other measure, specifically by the type of learning disability, suggests that their learning, remembering or concentrating difficulties may be part of their normative growth and development, or that their parents do not consider these difficulties to be a disability. Therefore, these children may be considered false positives (incorrectly identified) in the sample of children with disabilities, even if this aligns with the DSQ filters’ intention of identifying children who are likely to have a disability.

Children identified by the DSQ filters were the least likely to need extra support at their main child care arrangement. Further conceptual and methodological work on the DSQ filters is needed to ensure their suitability for a child population. For example, parent-reported communication disabilities should be explored, as this type of disability may be particularly relevant for children. Furthermore, additional analysis of the response categories for the DSQ filter questions may shed light on whether more conservative inclusion criteria (i.e., affirmative responses to the “often” and “always” response categories) could yield a more appropriate sample of children who are likely to have a disability, similar to the identification of functional difficulties by the Washington Group / UNICEF CFM (UNICEF, 2017).

About half of the children in the full sample were identified by the long-term condition measure. This result shows the importance of including the presence of a long-term condition in identifying child disability. The three most reported conditions among this group of children, specifically those with a long-term condition only, appeared to be less complex (less severe) and were related to mental, behavioural or neurodevelopmental disorders (e.g., attention deficit hyperactivity disorder); diseases of the respiratory system (e.g., asthma); and developmental anomalies (e.g., structural developmental anomaly of the heart). These findings align with previous research on the most prevalent diagnostic codes among school-age children with neurodevelopmental disorders and disabilities (NDD/D) and show that, while mental disorders are more common among children with NDD/D, physical health problems such as respiratory diseases are more common among children without NDD/D (Arim et al., 2017). While the group of children identified solely by a long-term condition indicated that the presence of such a condition may not lead to activity limitations or participation restrictions, just over one-third of this group required extra support at their main child care arrangement. This finding suggests that caution is needed and further analysis is warranted.

Parent-reported type of disability was slightly less frequently reported, with about 4 in 10 children in the full sample identified by this measure. Of these, about 6 in 10 had a communication disability; this figure was even higher (76%) among children identified solely by their type of disability. These findings highlight the importance of considering communication difficulties, which include being understood by and understanding others (e.g., speech delay, difficulties with receptive or expressive language), in young children. Indeed, functional communication difficulties were among the most prevalent functional difficulties identified among children aged 2 to 4 years in the 2019 CHSCY (Statistics Canada, n.d.). However, caution is needed to ensure that this measure does not falsely identify children who are experiencing these difficulties as part of their normative growth and development. Overall, the type of disability measure was the only one that directly identified children with communication disabilities, as reported by their parents. Given the considerable proportion of children identified with communication disabilities in this study (29% in the full sample and 36% in the final identified sample; Appendix A), future measurement of child disability could consider communication disabilities.

The Global Activity Limitation Indicator (GALI), as a participation restriction measure, identified just over one-fifth of the full sample. The majority (over 95%) of these children were also identified by another measure. One reason the GALI identified the lowest proportion of children may result from the assessment of limitations in activities for at least six months in a separate follow-up question. Notably, children identified by the GALI had the lowest proportion reporting excellent or very good health and the highest proportion needing for extra support at their main child care arrangement.

Another way of examining the groups of children was to focus on the children identified by a single measure versus multiple measures of disability. About 40% of children in the full sample were identified by a single measure, and a majority of them (except for those identified by the GALI) were reported as not being limited at all by their parents. In contrast, children identified by multiple measures were less likely to have excellent or very good general health and more likely to need extra support at their main child care arrangement. The fact that 60% of children were identified by multiple measures indicated considerable overlap between the four measures. About 15% of children were identified by all four measures, and the most frequently reported long-term conditions among these children were related to mental, behavioural or neurodevelopmental disorders and to developmental anomalies. While not directly comparable, this overlap (15%) aligns with findings from a U.S. study, where about 9% of children aged 0 to 17 years who met any disability definition met all five definitions in National Survey of Children’s Health data (Hagerman & Houtrow, 2021). About 9 in 10 children identified by all four measures in the current study needed extra support at their main child care arrangement, and just over half (55%) had excellent or very good health.

When the four child disability measures were considered in a model (i.e., simultaneously with other characteristics), the associations between the four measures and the use of child care did not statistically differ before or after considering other child, parent and family characteristics. Interestingly, the use of non-parental child care did not differ between children identified by a single measure and those identified by multiple measures, before or after considering other child, parent and family characteristics. Overall, these results raise the question of whether the further selection of a subgroup of children with disabilities could be useful for research purposes in the context of participation in child care.

In summary, while the four measures identified children who varied in their general health and need for extra support at their main child care arrangement, the use of non-parental child care was largely similar across all measures. About 6 in 10 children with disabilities were in non-parental child care, regardless of the identification measure used. Moreover, other child, parent and family characteristics were largely similar and consistent with previous research on children with neurodevelopmental disorders (Arim et al., 2012).

Together, these results demonstrate the usefulness of considering multiple measures and integrating the medical and social models in identifying child disability via a non-categorical approach, as recommended by the International Classification of Functioning, Disability and Health (ICF) (World Health Organization [WHO], 2002). Relying solely on the presence of a long-term condition (medical model) or limitations in daily activities (social model) may exclude some children who have participation restrictions in child care. The present study provides preliminary evidence that integrating the medical and social models can be achieved by using multiple existing measures. Without a national survey measure of child disability, using multiple measures could address the specific challenge of identifying children with disabilities. This approach could inform inclusivity in early learning and child care (ELCC) through national surveys such as the Survey on Early Learning and Child Care Arrangements – Children with Long-term Conditions and Disabilities (SELCCA-CLCD).

Results of in-depth analyses indicated that using a single disability measure did not necessarily identify children with lower general health or an increased need for extra support. Furthermore, many of the children identified by a single measure were reported to have conditions that may not be associated with reduced access or ability to participate in child care (e.g., diseases of the respiratory system, such as asthma). Therefore, it was necessary to explore the overlap between the measures to identify an analytical group of children with disabilities who may experience difficulties in accessing or participating in child care—specifically, those who were less likely to have excellent or very good general health and more likely to need extra support at their main child care arrangement. The final identified research sample from this survey (59% of the full sample) included children identified by the GALI or at least two of the other three measures: the Disability Screening Questions filter questions, a long-term condition and type of disability. However, the use of child care, as well as other child, parent and family characteristics, remained largely similar between these two groups.

Although it is beyond the scope of this study to determine whether this method of identification is aligned with the disability definitions used across different jurisdictions, the GALI assesses participation restrictions in children and is useful for international comparisons. However, it is not a measure of disability per se but a global measure of participation restriction. Recent research (e.g., Balbo & Bolano, 2023) has used the GALI as a tool to assess disability while acknowledging the challenges in measuring child disability. The SELCCA-CLCD appears to be the first Canadian survey to include the GALI. Future child surveys may consider including the GALI to further assess its utility in identifying children with disabilities in comparison with other measures such as the CFM used in the Canadian Health Survey on Children and Youth.

Ideally, researchers should aim to develop and validate a new national measure of child disability based on the biopsychosocial model depicted in the ICF (WHO, 2002). This could be useful for measuring disability in general surveys of child functioning and well-being, as well as in surveys and research focusing on the social participation of children with disabilities, such as the SELCCA-CLCD. Although the results of this study suggest that the GALI may be one such measure to include in future surveys, the results also suggest that the GALI is the most conservative measure of disability (identifying the smallest proportion of children) in the survey, and that other measures are similarly useful in identifying children with disabilities in child care, specifically when they are combined. Without a national survey measure, future research can examine whether these different measures vary across other indicators of participation in child care, such as the number of specific areas in which children need extra support. In the continued absence of a single disability measure that integrates the medical and social models of disability, future surveys should include multiple measures, such as those included herein, to identify a broad range of children with disabilities. However, further specification is needed for children who are identified by multiple measures, especially for those whose social participation, including child care participation, may be restricted by their condition, impairment or disability. This approach would continue to target the subpopulation of children who are of particular interest for social policies, including those aimed at creating and supporting inclusive child care.

Several limitations of this study should be acknowledged. First, the sample size for certain groups of children restricted the presentation of the findings and further analyses. Second, the study did not distinguish between participation in different types of child care (see Kerr et al., 2024). Third, the lack of a comparison group (i.e., children without disabilities) limited the contextualization of the findings.

Despite these limitations, this study informs future data collection activities and makes an important contribution to the efforts to adopt a national survey measure of disability for children to monitor progress on inclusivity in ELCC. Furthermore, it provides useful information for researchers regarding the individual and overlapping measures from the SELCCA-CLCD data to identify children with disabilities and to examine topics related to ELCC. If a single measure of child disability that integrates the medical and social models of disability is not yet available for future cycles of the SELCCA-CLCD, the results of the current study show that a single measure of disability would limit future work. The final sample used for this study—that is, children identified by the parent-reported GALI or at least two of the other three measures—not only considered both models of disability, but also identified children who were least likely to be in excellent or very good health and most likely to need extra supports in child care.

The availability of robust data on child care participation among children with disabilities at the national level remains a challenge. This is partly because of the complexities in identifying disability during childhood, when difficulties may be partly related to dynamic periods of childhood development, and because there are varying definitions of child disability used to deliver inclusive ELCC programs and services. A national approach to defining and measuring child disability could be useful for producing disaggregated information on children with disabilities and identifying opportunities and barriers to their participation in ELCC.

References

Arim, R., Findlay, L., & Kohen, D. (2016). What Statistics Canada survey data sources are available to study neurodevelopmental conditions and disabilities in children and youth? The School of Public Policy Research Papers, 9(29). (accessed January 3, 2024).

Arim, R., Findlay, L., & Kohen, D. (2020). The impact of the COVID-19 pandemic on Canadian families of children with disabilities. (accessed May 9, 2024).

Arim, R. G., Garner, R. E., Brehaut, J. C., Lach, L. M., MacKenzie, M. J., Rosenbaum, P. L., & Kohen, D. E. (2012). Contextual influences of parenting behaviors for children with neurodevelopmental disorders: Results from a Canadian national survey. Disability and Rehabilitation, 34(26), 2222-2233.

Arim, R. G., Miller, A. R., Guèvremont, A., Lach, L. M., Brehaut, J. C., & Kohen, D. E. (2017). Children with neurodevelopmental disorders and disabilities: A population-based study of healthcare service utilization using administrative data. Developmental Medicine & Child Neurology, 59(12), 1284-1290.

Balbo, N., & Bolano, D. (2023). Child disability as a family issue: A study on mothers’ and fathers’ health in Italy. European Journal of Public Health.

Beaudoin, A. J., Pratte, G., Gervais, M. H., Essalik, S., Camden, C. (2022). Children with disabilities and access to recognized early childhood education and care in Québec. Employment and Social Development Canada. (accessed January 5, 2024).

Bogaert, P., Van Oyen, H., Beluche, I., Cambois, E., & Robine, J. M. (2018). The use of the global activity limitation indicator and healthy life years by member states and the European Commission. Archives of Public Health, 76(1), 1-7. (accessed January 4, 2024).

Charters, T., Arim, R., & Findlay, L. (2024). An update on the profile of children with affirmative responses to the 2021 Census questions on difficulties with activities of daily living. Economic and Social Reports. Catalogue no. 36-28-001. (accessed June 26, 2024).

Charters, T., Schimmele, C., & Arim, R. (2022). A profile of children with affirmative responses to the 2016 census questions on difficulties with activities of daily living. Economic and Social Reports. Catalogue no. 36-28-001. (accessed January 22, 2024).

Child and Adolescent Health Measurement Initiative (n.d.). (2022). National Survey of Children’s Health (NSCH) data query. Data Resource Center for Child and Adolescent Health supported by the U.S. Department of Health and Human Services, Health Resources and Services Administration (HRSA), Maternal and Child Health Bureau (MCHB). (accessed February 2, 2024).

Childcare Resource and Research Unit (2022). A summary of the Canada-Wide Early Learning and Child Care agreements and action plans. (accessed January 15, 2024).

Dionne, C., Paquet, A., Lemire, C., & Girard, S. (2023). Inclusion in Quebec childcare centers: Financial support, adaptation, and training. Frontiers in Education, 8.

Employment and Social Development Canada [ESDC]. (2017). Multilateral Early Learning and Child Care Framework. (accessed January 3, 2024).

European Commission (2017). Progress report on the implementation of the European disability strategy 2010-2020. (accessed January 4, 2024).

Eurostat (2016). Glossary: Activity limitation. (accessed January 4, 2024).

Eurostat (2017). 2017 Module on Health and Children’s Health (Guidelines and questionnaire). (accessed January 4, 2024).

Friendly, M. (1994). Child care policy in Canada: Putting the pieces together. Don Mills, Ontario: Addison-Wesley.

Grondin, C. (2016). A new survey measure of disability: The Disability Screening Questions (DSQ). Statistics Canada. (accessed January 3, 2024).

Hagerman, T. K., & Houtrow, A. J. (2021). Variability in prevalence estimates of disability among children in the National Survey of Children’s Health. JAMA Pediatrics, 175(3), 307-310.

Halfon, S., & Friendly, M. (2013). Inclusion of Young Children with Disabilities in Regulated Child Care in Canada: A Snapshot: Research, Policy and Practice. Toronto: Childcare Resource and Research Unit. (accessed January 3, 2024).

Institut de la statistique du Québec. (2023). Child development in kindergarten 2022. (accessed January 5, 2024).

Irwin, S. H., & Lero, D. S. (2020). Inclusion quality: Children with disabilities in early learning and child care in Canada. Wreck Cove, Nova Scotia: Breton Books. (accessed January 3, 2024).

Irwin, S. H., Lero, D. S., & Brophy, K. (2000). A matter of urgency: Including children with special needs in child care in Canada. Wreck Cove, Nova Scotia: Breton Books. (accessed January 3, 2024).

Kerr, S., Findlay, L., & Arim, R. (2024). Child care for young children with disabilities. Health Reports, 35(10), 29-40.

Kohen, D. E., Brehaut, J. C., Garner, R. E., Miller, A. R., Lach, L. M., Klassen, A. F., & Rosenbaum, P. L. (2007). Conceptualizing childhood health problems using survey data: A comparison of key indicators. BMC Pediatrics, 7(1), 1-13.

Kowalchuk, K. (2008). Participation and Activity Limitation Survey 2006: Families of children with disabilities in Canada. Catalogue no. 89-628-X20008009. Ottawa. (accessed January 3, 2024).

Maich, K., Davies, A. W., Penney, S. C., Butler, E., Young, G. D., & Philpott, D. (2019). Young children with autism spectrum disorder in early education and care: The earlier we begin together, the better. Exceptionality Education International, 29, 77-91. (accessed January 8, 2024).

Philpott, D. F., Young, G., Maich, K., Penney, S. C., & Butler, E. (2019). The preemptive nature of quality early child education on special educational needs in children. Research Report. Memorial University of Newfoundland, St. John’s, Newfoundland and Labrador. (accessed January 12, 2024).

Robine, J.-M., Jagger, C., & Euro-REVES Group. (2003). Creating a coherent set of indicators to monitor health across Europe. The Euro-REVES 2 project. European Journal of Public Health, 13(3), 6-14. (accessed January 4, 2024).

Russell, L. A., Tinker, S. C., Rice, C. E., Ryerson, A. B., & Gonzalez, M. G. (2023). Variation in identifying children and adolescents with disability and developmental disability in population-based public health surveys. Disability and Health Journal.

Statistics Canada (n.d.). Table 13-10-0764-01.  Functional difficulties for children aged 2 to 4 years, Canadian Health Survey on Children and Youth 2019.

Stein, R. E., & Jessop, D. J. (1989). What diagnosis does not tell: The case for a noncategorical approach to chronic illness in childhood. Social Science & Medicine, 29, 769-778.

Stein, R. E., Bauman, L. J., Westbrook, L. E., Coupey, S. M., & Ireys, H. T. (1993). Framework for identifying children who have chronic conditions: The case for a new definition. Journal of Pediatrics, 122, 342-347.

Underwood, K., & Snoddon, K. (2021). IECSS Policy Brief No. 10: Deaf childhoods and inclusive early childhood education and care. Inclusive Early Childhood Service System project. (accessed January 15, 2024).

United Nations (n.d.a). Goals 4 Ensure inclusive and equitable quality education and promote lifelong learning opportunities for all. (accessed October 8, 2024).

United Nations (n.d.b). Convention on the Rights of Persons with Disabilities (CRPD). (accessed October 8, 2024).

United Nations Children’s Fund [UNICEF] (n.d.). Convention on the Rights of the Child. (accessed October 8, 2024).

United Nations Children’s Fund [UNICEF] (2017). Module on Child Functioning: Concept note. (accessed June 3, 2024).

United Nations Children’s Fund [UNICEF] (2021). Seen, counted, included: Using data to shed light on the well-being of children with disabilities. New York. (accessed January 3, 2024).

United Nations Educational, Scientific and Cultural Organization (2021). Right from the start: Build inclusive societies through inclusive early childhood education. Global Education Monitoring Report – Policy Paper 46. Paris: Licence: CC-BY-SA 3.0 IGO. (accessed January 3, 2024).

Van Oyen, H., Bogaert, P., Yokota, R. T., & Berger, N. (2018). Measuring disability: A systematic review of the validity and reliability of the Global Activity Limitations Indicator (GALI). Archives of Public Health, 76(25), 1-11.

World Health Organization [WHO] (2002). Towards a common language for functioning, disability and health: ICF.

World Health Organization [WHO] (2022). International Classification of Diseases, Eleventh Revision (ICD-11). Geneva: License: CC BY-ND 3.0 IGO. (accessed June 3, 2024).

World Health Organization [WHO] and United Nations Children’s Fund [UNICEF]. (2023). Global report on children with developmental disabilities: From the margins to the mainstream. Geneva: Licence: CC BY-NC-SA 3.0 IGO. (accessed January 3, 2024).