Objective. To present an epidemiologic profile of children with special health care needs using a new definition of the population developed by the federal Maternal and Child Health Bureau.
Methods. We operationalized the new definition using the recently released 1994 National Health Interview Survey on Disability. Estimates are based on 30 032 completed interviews for children <18 years old. The overall response rate was 87%.
Results. Eighteen percent of US children <18 years old in 1994, or 12.6 million children nationally, had a chronic physical, developmental, behavioral, or emotional condition and required health and related services of a type or amount beyond that required by children generally. This estimate includes children with existing special health care needs but excludes the at-risk population. Prevalence was higher for older children, boys, African-Americans, and children from low-income and single-parent households. Children with existing special health care needs had three times as many bed days and school absence days as other children. An estimated 11% of children with existing special health care needs were uninsured, 6% were without a usual source of health care, 18% were reported as dissatisfied with one or more aspects of care received at their usual source of care, and 13% had one or more unmet health needs in the past year.
Conclusions. A substantial minority of US children were identified as having an existing special health care need using national survey data. Children with existing special health care needs are disproportionately poor and socially disadvantaged. Moreover, many of these children face significant barriers to health care.
- NHIS =
- National Health Interview Survey •
- NHIS-D =
- National Health Interview Survey on Disability
Anumber of conceptual approaches can be used to classify and characterize children with chronic conditions and ongoing health problems.1 In particular, three approaches have been commonly used. Condition lists have been used for many years to define populations of children with chronic illnesses.2-8Functional status assessments are used to identify children whose chronic conditions cause impairments in basic functions, such as hearing or seeing, or impairments in higher level functioning required to conduct activities of daily living, such as eating, bathing, and dressing.9-11 Limitation in socially defined roles, such as school or play, due to chronic conditions has been used for more than 40 years to identify children with disabilities.12-15
Historically, all three approaches—condition lists, functional impairments, and disability—have been used to identify target populations for public programs serving children with chronic conditions. However, as the mission of the federal and state Title V programs for children with special health care needs has expanded in recent years, the usefulness of these traditional approaches has diminished. The shift in focus from a historically narrow conception of the target population to a broader conception is visibly evidenced by the changing terminology used to describe children in the target population; over the past two decades, the term crippled children has been replaced by children with special health care needs. Yet, despite the emergence of this broader conception of the target population and the new terminology, no uniform definition of the population has been available.
In response to the need for a definition that could be used for planning and advocacy purposes, the federal Maternal and Child Health Bureau's Division of Services for Children With Special Health Care Needs developed the following definition:
Children with special health care needs are those who have or are at increased risk for a chronic physical, developmental, behavioral, or emotional condition and who also require health and related services of a type or amount beyond that required by children generally.
The derivation of the definition and a discussion of the terminology used is presented in the June 1998 issue ofPediatrics (see reference 23).
In this article we present an epidemiologic profile of children with existing special health care needs. We operationalize the new definition using a large nationally representative survey of children designed in part to identify children with special health care needs. The profile includes national estimates of the number and proportion of US children with special health care needs at the time the survey was conducted in 1994, as well as descriptive information on the demographic and socioeconomic characteristics of affected children and a description of their health status and service use patterns.
Our presentation is limited to profiling the population of children with existing special health care needs and does not include the component of the population that is at increased riskfor a special health care need. Presently, there is no accepted approach to identifying the at-risk segment of the population. Indeed, the development of acceptable methods for operationalizing the concept of risk and estimating the size of the at-risk populations remains an important research priority.
We used the National Health Interview Survey (NHIS) to operationalize the new definition.16 The NHIS is a continuing nationwide household survey that is conducted by the Bureau of the Census for the National Center for Health Statistics. Its purpose is to collect information on the health status and use of health services by the US civilian noninstitutionalized population. Each year the survey instrument consists of a core questionnaire on health status, health care use, and demographic characteristics of the population, as well as modules on topics of current interest. During 1994, two special supplemental questionnaires on disability, collectively referred to as the National Health Interview Survey on Disability (NHIS-D), were included as topical modules.
The NHIS-D was conducted in two phases. The Phase I disability questionnaire was administered at the same time as the NHIS core questionnaire and collected basic data on disability and special needs for all household members. It was also used as a screening device to determine eligibility for a more in-depth follow-back survey (Phase II) of those individuals identified with disabilities or special needs.
We used data collected in Phase I of the 1994 NHIS-D to produce the estimates of US children with existing special health care needs presented in this report. The estimates are based on 30 032 completed interviews for children <18 years old. The overall response rate for the 1994 Phase I survey was 87%. An adult knowledgeable about the health of the sample child, usually the mother, served as a respondent for this survey. Children >16 years of age were permitted to respond for themselves.
Phase I of the NHIS-D was originally designed as a comprehensive screening tool that could be used to identify a wide range of children and adults with chronic conditions, impairments, disabilities, and elevated service needs. Although the NHIS-D was not designed specifically for our purpose, the breadth of the Phase I screening instrument and the large nationally representative sample make it a useful tool for producing a epidemiologic profile of children with existing special health care needs.
The Phase I screening instrument contained hundreds of questions and probes relevant to ascertaining the functional status and service needs of children. As a first step toward producing the epidemiologic profile, a series of decisions had to be made concerning which questions and probes would be used in identifying children with special health care needs. To accomplish this goal, a work group consisting of federal and state directors of programs for children with special health care needs, parents, practitioners, and public health analysts (ie, the authors of this report) reviewed the screening questions for children. Specifically, work group members were asked to independently review NHIS-D questionnaire items for children and identify those questions and probes that should be used in operationalizing the new definition. Only questions and probes with an agreement rate higher than 75% among work group members were included for this purpose. A positive response to any of these items resulted in inclusion of the sample child in the population with existing special health care needs.
The work group identified two subgroups of children as part of this process. The first and larger group included children who were reported at the time of the interview to be using services at an elevated level due to a chronic physical, developmental, behavioral, or emotional condition. Included in this group were children who were reported as using the services of physicians, therapists, and other health providers/facilities on a regular basis for the treatment of a chronic health problem; children with special diets, requiring prescription medications on a long-term basis, or needing personal assistance or using special equipment aids, or personal assistance for mobility, hearing, vision or other needs; and children receiving special education or early intervention services. However, not all children with elevated service needs actually receive services because of access barriers, lack of knowledge on the part of parents, and other factors. Consequently, the work group identified a second and smaller group of children who had a presumed need for services resulting from the presence of functional limitations or disabilities. Children in this group experienced developmental delays, learning disabilities, vision and hearing impairments, and other types of functional limitations and disabilities but were not identified as using services at an elevated level at the time of the survey. Both groups were combined in the epidemiologic profile that follows.
All estimates presented in the text and tables have been statistically weighted to reflect national population totals. Because the NHIS is based on a stratified cluster sample design, standard errors used in computing test statistics were calculated using a Taylor series linearization variance estimation technique developed by the Research Triangle Institute.18 Unless otherwise indicated all differences discussed in the text were significant at the .05 level or higher.
In addition to bivariate comparisons, multivariate analyses were also conducted to identify independent demographic and socioeconomic predictors of special health care needs. Logistic regression analysis was used for this purpose. Only the summary results of the multivariate analyses are presented here; copies of the complete equations are available from the authors.
A list of the NHIS-D variables selected to serve as inclusion criteria for the new definition is presented in the Appendix . The NHIS-D data set is available from the National Center for Health Statistics on CD-ROM.17 The variable list in the Appndix can be used with the data set contained on the CD-ROM to replicate the epidemiologic estimates presented here. An algorithm for doing so is available from the authors.
Current Prevalence Estimates
Results from the 1994 survey indicate that 12% of US children <18 years old had a chronic physical, developmental, behavioral, or emotional condition and also used health or related services beyond those required by children generally. Our analysis indicates an additional 6% of children had a presumed need for health or related services beyond those required by children generally, even though they were not identified as using an elevated level of services at the time of the survey. The two groups—those with existing elevated service use and those with presumed elevated services needs—were combined in this profile of children with existing special health care needs. Thus, we estimate that 18% of US children <18 years old in 1994, or 12.6 million children nationally, had a chronic physical, developmental, behavioral, or emotional condition and required health and related services of a type or amount beyond that required by children generally.
Demographic and Socioeconomic Correlates
The prevalence of existing special health care needs varied by demographic and socioeconomic characteristics of the child population (Table 1). Prevalence increases in a stepwise fashion with age, such that school-aged children were twice as likely as toddlers to be classified as having an existing special need. Boys were about one third more likely than girls to have a special need. Differences are also apparent by race and ethnicity. Specifically, African-American children were most likely although Hispanic children and other minorities were least likely to be categorized as having an existing special need using the new definition.
Children from families with incomes at or below the federal poverty level ($12 320 for a family of three in 1994) were about one third more likely than children in families with incomes above poverty to have an existing special health care need. A gradient in prevalence was also found by educational attainment of the head of household, with children from less educated households exhibiting a higher likelihood of experiencing a special health care need at the time of the survey. Family structure was also related to prevalence; children in single-parent families were about 40% more likely to have existing special health care needs as children from two-parent households.
Differences in prevalence across age, gender, race, income, parental education, and family structure remained significant in a multivariate analysis (not shown), indicating each is independently associated with presence of an existing special health care need. Thus, children from disadvantaged families are disproportionately represented in the population of children with existing special health care needs. Additional analysis is needed to understand how different social and demographic factors contribute to the incidence and manifestation of special health care needs for children.
Health Characteristics of Children With Special Health Care Needs
As shown in Table 2, there are substantial differences in health status, as measured by the impact of illness, for children with and without existing special needs. Children with existing special health care needs had three times as many days spent ill in bed and three times as many school absence days as other children. These differences in rates translate to an additional 52 million days spent ill in bed and 58 million school absence days annually for children with existing special health care needs compared with other children.
Access to and Use of Services
Several dimensions of access to care are presented in Table 2 for children with and without existing special health care needs. Although most children with an existing special health care need had health insurance coverage in 1994, an estimated 11.2% were uninsured. The vast majority of children with existing special health care needs had a usual source of care in 1994, but 5.6% were without a usual place to go when the child was sick or when the family needed advice about the child's health. Nearly 1 in 5 respondents to the survey reported not being satisfied with at least one dimension of care provided to their child with an existing special health care need. Rates of dissatisfaction were higher for children with existing special needs than children without. More than 1 in 10 children with existing special health care needs was reported to have an unmet need for medical, dental, vision, prescription medications, or mental health care in the past year. Here again, children with existing special health care needs were at significant added risk of having unmet health needs compared with their counterparts without special needs.
The final set of data presented in Table 2 pertain to use of physician and hospital services by children with and without existing special health care needs. Substantial differentials in health care use were found for both physician and hospital services, with the greatest differences being apparent for hospital-based care. Children with existing special health care needs had more than twice as many physician contacts and five times as many hospital days as other children in 1994. Additional information on use of specialized services by children in this population will be available soon when data from the Phase II follow-up survey are released by the National Center for Health Statistics.
The New Definition in Context
In earlier studies using a variety of data sources, prevalence estimates for childhood chronic conditions have ranged from <5 to >30%.2-15 19 20 This degree of estimate variation has been explained by use of different approaches to defining chronic conditions, different severity thresholds, different ascertainment methods, and varying settings and purposes for conducting the studies.20-22 The higher estimates tend to include chronic conditions that have modest effects on children's activities or use of health and related services, although estimates at the lower end of the spectrum generally include only conditions causing substantial disability or need for personal assistance or special equipment in conducting the activities of daily living. Consequently, it should not be surprising that the prevalence estimate for the new definition of children with special health care needs is near the middle of this range.
Figure 1 illustrates the range of prevalence estimates derived for US children using national data sources. The left-most bar shows that 31% of children experience chronic conditions of a physical nature (no national estimates of the combined prevalence of physical and mental conditions are available).2 The next bar illustrates the prevalence estimate of children with existing special health care needs using the new definition. The subsequent bar shows that nearly 7% of children experience limitations in social role activities, such as school or play, because of chronic physical or mental conditions.16 A much smaller proportion of children, 0.2% nationwide, experience a need for assistance or special equipment in conducting the activities of daily living (eating, bathing, dressing, etc) because of a chronic physical or mental condition.9 Finally, the right-most bar shows that only about 1 in every 1000 children is institutionalized because of a chronic health problem.13
A variety of definitional approaches have been developed over the years to identify and classify children with chronic health problems. The definition presented here of children with existing special health care needs yields a prevalence rate that is higher than definitions based solely on functional limitations,9 but lower than definitions based simply on chronicity or duration of illness.2 However, the new definition yields a prevalence estimate that is similar to those independently developed by researchers at the Albert Einstein College of Medicine23and the National Association of Children's Hospitals and Related Institutions.24 Thus, there now appears to be some convergence by experts in the field concerning the number of children with significant ongoing health care needs. Specifically, all three definitions suggest that between 15% and 20% of US children have a significant ongoing health care need related to a chronic health condition. Additional children are at-risk of developing an ongoing health care need, but no estimates of their number are available at present.
Currently available definitions of the population of children with special health care needs are useful for epidemiologic study purposes. However, in present form they have limited practical usefulness for identifying and tracking children with special health care needs for quality assurance, risk adjustment, and other purposes. As more of these vulnerable children enter managed care arrangements, the need for identification approaches that can be applied in a simple and straightforward fashion has become more urgent. Efforts to develop tools with practical applications for managed care and other purposes should be accelerated. In addition, more research is needed to identify critical predictors of special health care needs to establish estimates of the size of the at-risk population.
This work was supported by the Maternal and Child Health Bureau, US Department of Health and Human Services (Cooperative Agreement MCU-06 MCP1).
- Received August 29, 1997.
- Accepted January 27, 1998.
- Address correspondence to: Paul W. Newacheck, DrPH, Institute for Health Policy Studies, 1388 Sutter St, Suite 1100, San Francisco, CA 94109.
- Perrin EC,
- Newacheck PW,
- Pless IB,
- et al.
- ↵Haggerty RJ. Epidemiology of childhood disease. In: Mechanic D, ed. Handbook of Health, Health Care and the Health Professions. New York, NY: Free Press; 1983
- ↵Walker DK, Gortmaker SL, Weitzman M. Chronic Illness and Psychosocial Problems Among Children in Gennessee County. Boston, MA: Community Child Health Studies (Harvard School of Public Health); 1981
- Cadman D,
- Bole M,
- Szatmari P,
- et al.
- Gortmaker SL,
- Walker DK,
- Weitzman M,
- Sobol AM
- ↵Collins JG. Prevalence of selected chronic conditions, United States, 1983–85. Vital and Health Statistics (Advance Data). Series No. 155. Hyattsville, MD: Public Health Service, National Center for Health Statistics; May 1988. US Department of Health and Human Services Publ. No. PHS 88–1250
- ↵Newacheck PW, Fox H, McManus M. Home care needs of chronically ill children. Caring. June 1988;VII(6):4–10
- ↵US Department of Health, Education and Welfare. Children and youth—selected health characteristics, United States—1958 and 1968. Vital and Health Statistics 10(62). Hyattsville, MD: National Center for Health Statistics, US Department of Health, Education, and Welfare; 1971
- ↵Aron LY, Loprest PJ, Steuerle CE. Serving Children With Disabilities—A Systematic Look at the Programs. Washington, DC: Urban Institute Press; 1996
- ↵McNeil JM. Americans with disabilities: 1991–1992. Current Population Reports. Washington, DC: US Bureau of the Census; 1993. Publ. No. 70–33
- ↵Adams PF, Marano MA. Current estimates from the National Health Interview Survey, 1994. Vital Health Stat. 10(193). 1995. Hyattsville, MD: National Center for Health Statistics. Department of Health and Human Services Publ. No. (PHS) 96–1521
- ↵National Center for Health Statistics, Department of Health and Human Services. 1994 National Health Interview Survey on Disability, Phase 1. CD-ROM Series 10–8. Hyattsville, MD: National Center for Health Statistics; 1996
- ↵Shah BV, Barnwell BG, Bieler GS. SUDAAN User's Manual, Release 7.0. Research Triangle Park, NC: Research Triangle Institute; 1996
- ↵Sulz HA, Schlesinger ER, Mosher WE, et al. Long-Term Childhood Illness. Pittsburgh, PA: University of Pittsburgh Press; 1972
- Weiland SK,
- Pless IB,
- Roghmann KJ
- Westbrook LE,
- Silver EJ,
- Stein REK
- Copyright © 1998 American Academy of Pediatrics