Children and Youth With Special Health Care Needs: A Profile

The NSCH is funded and directed by the HRSA MCHB and conducted by the US Census Bureau. The NSCH was significantly redesigned for 2016, combining 2 previously separate quadrennial surveys, the National Survey of Children With Special Health Care Needs (NS-CSHCN) and the NSCH.¹¹  Since 2016, the survey has been conducted as an annual, crosssectional, address-based, self-administered survey using both web and paper-based methodologies. The survey employs a 2-phased design: (1) a screener to assess the presence, basic demographic characteristics, and SHCN status of children in the home; and (2) a detailed, age-specific questionnaire (0–5 years, 6–11 years, and 12–17 years) that collects information about the health, health care, and related family and community-level factors that can impact health for 1 randomly selected child. The 2-phased design is notable because it facilitates an 80% oversample of CYSHCN in households with both CYSHCN and non-CYSHCN present.¹² 

Data are collected for approximately 9 months of the year (June–January) and released the following October. The 2016 NSCH yielded a total of 50 212 questionnaires, followed by 21 599 in 2017, 30 530 in 2018, and 29 433 in 2019. The overall weighted response rate in each year was: 40.7%, 37.4%, 43.1%, and 42.4%, respectively, and the weighted interview completion rates (reflecting the probability that a household that started the survey completed it) were 69.7%, 70.9%, 78.0%, and 79.5%. Approximately 75% of all questionnaires are completed online with the remainder completed via paper and mail. Participation in the survey is voluntary, and data are collected in English or Spanish from a parent or caregiver knowledgeable about the child’s health and health care. The design and administration have been described previously.^12–15  NSCH data and questionnaires are publicly available.¹⁶ 

CYSHCN are identified using the CSHCN screener which assesses: (1) need and use of prescription medication; (2) above-routine use of medical, mental health, or educational services compared with other children of the same age; (3) functional limitations in day-to-day life; (4) need and use of specialized therapies; and (5) need and use of treatment or counseling for an emotional, behavioral, or developmental condition. Following each question, respondents confirm the presence of a related physical, mental, behavioral, or other health condition that has or is expected to last at least 12 months (Table 1).^8,17  Type of SHCN was derived from responses to the screener items and grouped according to healthcare and functional impacts: (1) prescription medication need and use only, (2) elevated service need and use only, (3) elevated service need and use and prescription medication need and use, or (4) functional limitations.¹⁸ 

Sociodemographic characteristics considered include the child’s sex, age (0–5, 6–11, or 12–17 years), race and ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, non-Hispanic American Indian or Alaska Native, non-Hispanic Asian, non-Hispanic Native Hawaiian or Other Pacific Islander, or non-Hispanic Multiple race), primary household language (English or non-English), parental nativity (parent(s) born in United States or parent(s) born outside of United States), family structure (2 parents, married; 2 parents, unmarried; single parent; or other), household income-to-poverty ratio (<100% FPL, 100% to 199% FPL, 200% to 399% FPL, or ≥400% FPL), highest education among primary caregivers in the household (less than or equal to high school or general educational development (GED), some college or associate's degree, or greater than or equal to a bachelor's degree), insurance status and type (public only, private only, combination of public and private, or uninsured), military status of primary caregivers in the household (current active duty, previous but not current active duty, active duty for training in the Reserves or National Guard, or no military service), state of residence, geographic region (Northeast, South, Midwest, or West), and urban-rural residence, defined by metropolitan statistical area (MSA) designation (MSA Principal City, MSA nonPrincipal City, or non-MSA).¹⁹  Health-related factors include: daily activity limitations (never affected; moderately affected - some of the time; or consistently affected - often or a great deal) and the presence of 1 or more physical, emotional or developmental diagnosed conditions. The NSCH assesses the presence of 27 conditions based on parent and caregiver responses to the following question: “Has a doctor or other health care provider ever told you that this child has…;” affirmative responses are followed by condition-relevant follow-up questions.

The receipt of comprehensive, high-quality services in a well-functioning system of care is assessed using 34 separate survey items which comprise 5 to 6 core outcomes, depending on the child’s age: (1) families are partners in decision-making; (2) medical home access; (3) early and continuous screening for special health care needs (operationalized through receipt of preventive health care); (4) continuous and adequate health insurance; (5) access to community-based services; and for children ages 12 to 17 years, (6) receipt of healthcare transition planning services. A full list of survey items used to assess the core outcomes is presented in Table 2.

Public Use Data files for 2016, 2017, 2018, and 2019 were appended to obtain the largest possible sample of CYSHCN for a final analytic sample of 131 774 children ages 0 to 17 years old, including 30 301 CYSHCN. Bivariate associations between the covariates and SHCN status as well as the 6 core outcomes were examined using χ2 tests (Tables 3 and 4). Multivariable logistic regression was used to identify factors independently associated with SHCN status and receipt of care in a well-functioning system (Table 5). State-level estimates were compared with the national estimate using t tests for overlapping groups (Fig 1 and Supplemental Table 6). Urban-rural estimates were calculated using the restricted access file and approved for release by the US Census Bureau (disclosure avoidance approval number CBDRB-FY21-POP001-0179).

Sample weights are developed by the US Census Bureau annually to account for differences in the probability of household selection and differential nonresponse. Missing data for child sex, race, and ethnicity were imputed using hotdeck methods (missing <1%). Two additional variables were multiply imputed using sequential regression methods: federal poverty ratio (missing: 18.6% in 2016, 16.0% in 2017, 15.3% in 2018, and 17.6% in 2019) and respondent’s educational attainment (missing: 3.0% in 2016, 2.0% in 2017, 2.0% in 2018, and 2.4% in 2019). The application of these weights to analyses allow for the production of estimates that are representative of all noninstitutionalized children ages 0 to 17 residing in housing units at both the national and state levels. All analyses accounted for the complex survey design using SAS-callable SUDAAN, version 11.0.1. (Research Triangle Institute).

Nearly 14 million, or 18.8% of US children, were reported to have a SHCN from 2016 to 2019. As illustrated in Table 3, the sociodemographic profile of the SHCN population differed from the population without such needs. For example, although the child population without SHCN was roughly evenly split by sex and age, CYSHCN were more likely to be male (58.0% versus 49.5%) and older than 5 (88.2% versus 64.5%). With respect to race and ethnicity, CYSHCN were more likely than non-CYSHCN to be non-Hispanic Black (17.0% versus 12.4%), but less likely to be Hispanic (21.7% versus 26.1%) or non-Hispanic Asian (2.6% versus 5.1%). CYSHCN were less likely to live in a household where English was not the primary language (8.1% compared with 15.9%) or both parents were born outside of the United States (19.4% compared with 29.9%). CYSHCN were more likely to live in poverty (23.9% versus 19.4%) or have any public health insurance coverage (46.7% versus 32.9%) and less likely to live in a household where at least 1 adult had attained a college degree (46.6% versus 50.0%) or had 2 married parents (56.1% versus 67.7%). Geographic variation was observed by both region and state, with a lower percentage of CYSHCN compared with non-CYSHCN residing in the West (20.6% versus 25.1%) and state-level CYSHCN prevalence estimates ranging from 13.6% in Hawaii to 24.0% in West Virginia (Fig 1). No difference was observed by urban-rural residence.

After adjustment, observed bivariate associations generally remained, with a few notable exceptions (Table 4). The association between race and ethnicity and poverty with SHCN status were both attenuated and no longer significant after adjustment, whereas the association between parental education and SHCN status reversed such that children of parents with less education were less, rather than more, likely to have a SHCN.

CYSHCN are a heterogeneous population and the nature and impact of their health status and needs can be characterized in multiple ways. For example, using the 5 screening criteria in the CSHCN screener, nearly two-thirds (63.1%) of CYSHCN qualified based on meeting 1 or 2 screening criteria, whereas only 7.5% were reported to meet 5 criteria. The most commonly reported type of SHCN was prescription medication use without the need for or use of additional services (29.3%) followed by need and use of prescription medication service utilization at an above-routine rate (26.8%) and functional limitations (25.9%), whereas 18.0% needed and used elevated medical, mental health, or education services without some other qualifying impact (Table 3; Fig 2).

The most common of 27 captured health conditions among CYSHCN was allergies (42.0%), followed by ADHD (33.8%), and behavior or conduct problems (27.4%). Eight of the top 10 health conditions among CYSHCN were developmental, mental, or behavioral conditions (Fig 3). Among those with a condition, over one-third (35.4%) were reported to have no activity limitations, whereas 45.7% were reported to have moderately affected daily activities, and nearly one-fifth (18.9%) were reported to have consistently affected daily activities (Fig 4).

Overall, only 14.9% of CYSHCN were reported to receive care in a well-organized system (Table 5). The proportion of CYSHCN who met each of the core outcomes ranged from 19.9% for healthcare transition planning to 87.7% for ease of access to community-based services. Less than half of CYSHCN (42.7%) received care in a Medical Home whereas nearly two-thirds (63.0%) were reported to have both continuous and adequate insurance coverage. Significant differences in the proportion of CYSHCN receiving care in a well-organized system were observed. Specifically, rates were higher among younger, non-Hispanic White children living in the United States, born in English-speaking households with greater educational attainment and income. However, after adjustment, race and ethnicity, primary household language, and educational attainment were not significantly associated with receiving services in a well-organized system, whereas child age and SHCN type had the largest rate ratios indicating an approximately 50% to 75% lower likelihood among older CYSHCN and those with qualifying needs beyond prescription medication (Supplemental Table 2).

CYSHCN remain a sizable and diverse population within the United States, representing nearly one-fifth of all children. Our results highlight important sociodemographic and health-related patterns among the CYSHCN population and their families and are consistent with previous research regarding both the prevalence of SHCN and disparities in access to a high-quality, well-organized system of care. Importantly, race and ethnicity and poverty were not associated with SHCN status after adjustment. However, the finding that first generation children and those in non-English speaking households are about a third less likely to have a SHCN after controlling for other sociodemographic characteristics suggests that there may be persistent access barriers to diagnosis and/or that the screener may be less sensitive for these children. Unfortunately, only 1 in 5 CYSHCN receive care in a well-functioning system with the most opportunity for improvement for the medical home and adolescent transition to adult health care, supporting the longstanding selection of these indicators as Title V performance measures. That receipt of well-functioning systems of care is even lower for those who need it most—those with more qualifying criteria, more complex needs, and those with more activity limitations—continues to be striking and an area in major need for improvement. The lower rates of well-functioning systems of care for first generation and lower income CYSHCN even after adjustment indicates additional inequities in access to quality health care that must be addressed.

Of note, these analyses provide the foundation for the critical work featured in this Special Issue that investigates the intersection of factors such as race, ethnicity, condition severity, complexity, and poverty. Taken together, these studies and the blueprint provide a set of priorities and actionable recommendations to support the health and wellbeing of CYSHCN and their families.

One such mechanism will continue to be the use of the NSCH as the primary national and state-level surveillance mechanism for this population. Although the annual administration has the capacity to produce both state- and national-level estimates, and breadth of content remain significant strengths, the NSCH is not without limitations. Specifically, work may be needed to assure that the CSHCN screener appropriately captures the full breadth of the SHCN population. Previous work has found discordance between the proportion of children for whom a diagnosed condition or functional difficulty was reported and the proportion identified as having a SHCN based on the same parent or caregiver’s responses to the CSHCN screener.¹⁷  The drivers of this discordance may be varied, ranging from the transient nature of pediatric conditions and/or health needs, inaccurate diagnosis and reporting, and/or differences in cultural norms and expectations related to health status and care seeking.¹⁷  Further work is needed to assess who is captured using the screener, who is missed, and why.

Second, since 1998, the HRSA MCHB definition of CYSHCN has retained a reference to children “at risk” for developing SHCN and yet the survey continues to lack a standardized measure for this population. Newacheck et al proposed a conceptual model for the development of SHCN based on the presence of factors in multiple domains, operating at the child, family, and community levels, which together, over time, create the conditions necessary for a SHCN.^6,20  The application of this model to the first iteration of the NSCH found a wide range of factors to be associated with a higher risk for SHCN, as well as protective factors, but did not provide a clear risk profile, in large part because of the crosssectional nature of the data and a paucity of indicators in some domains.²⁰  Using this broad lens to revisit the first limitation noted above, the observed incongruity between the proportion of children meeting 1 or more criteria of the CSHCN screener and the proportion of those with a diagnosed condition or noted limitation may reflect some portion of the at risk population – those with a condition that does not result in the impact(s) required by the screener. Further analyses are needed to test and refine a model toward the specification of policy- and programmatic-relevant approaches to defining risk that can ultimately inform research and prevention efforts.

Data from the NSCH are subject to at least 3 additional limitations. (1) Information is reported by parents and caregivers and thus, susceptible to response bias. Additionally, many of the component measures which comprise the core outcomes can be subjective in nature and parent and caregiver reports may be influenced by prior experiences and expectations. For example, a parent may not report unmet healthcare needs if they do not know of or expect their child to have access to particular services. Social desirability may also impact parental reports of health risk behaviors and, in some cases, diagnoses. (2) Because of changes in both sampling and mode of administration, comparison of estimates from the redesigned NSCH to those from previous iterations of either the NSCH or the NS-CSHCN should not be made, thus precluding longer-term trend analyses. 3) Finally, the sampling strategy employed does not permit the collection of self-reported data from older adolescents (≥18 years) and young adults. As such, the capacity to assess the impact of healthcare transition planning is limited.

Since the early 2000s, the NSCH has been a key source of information on the health, wellbeing, and needs of CYSHCN. The 2016 redesign ushered in important advances; however, opportunities exist to extend the utility of the survey. One recent evolution is the capacity for states to support oversamples that can be used to produce more robust analyses of rare outcomes or more granular geographic estimates. Additionally, HRSA MCHB is now planning to administer a longitudinal follow-up survey of respondents to previous NSCH surveys to assess the lasting effects of the COVID-19 pandemic on the health, wellbeing, and development of children in the United States. Although it is expected that the survey will yield important insights for all children, information on the experience of CYSHCN, particularly those who may have lost access to service supports and specialized care, will be critical to the development of future programs and policies. Through these advances and the release of high-quality, timely data from the annual NSCH, HRSA MCHB will continue to support critical research on the health status and needs of this vulnerable population to inform the development of effective programs to serve CYSCHN and their families.

NSCH

National Survey of Children’s Health

HRSA

Health Resources and Services Administration

MCHB

Maternal and Child Health Bureau

CYSHCN

Children and Youth with Special Health Care Needs