Trends in Disability in Early Life
Ruth E. K. Stein*
This paper provides an overview of the trends in disability in early life and has five main sections. The first section identifies key demographics that may be important to the Institute of Medicine (IOM) Committee on Disability in America: A New Look. The second section reviews traditional measures of childhood disability in use over the past several decades and the trends that they show. The third section highlights newer approaches to the assessment of disability among children and youth and what is known about trends determined by the use of these newer techniques. The fourth provides some examples of changes in disability patterns, and the last section provides conclusions with some implications for the committee.
Children younger than 18 years of age constitute 25 percent of the population in the United States. Over the past half-century there has been an increase in the number of children in the population, from 44 million in 1950 to 73 million in 2003, and this number is expected to rise to 80 million by 2020.1 There has also been a dramatic change in the composi-
tion of the child population. While white non-Hispanic children represented 75 percent of the nation’s children in 1980, by the year 2020 they are expected to be a little over 50 percent of the child population.2 In contrast, the proportion of minority children is increasing and the Hispanic child population is growing at the fastest rate and will exceed one in five children within the next few years.3
These data are important for a number of reasons, especially because of the strong association between membership in a minority group and poverty. Poverty is a major correlate of poor child health and has been shown to have important long-term health consequences, such that morbidity and mortality are strongly associated with income. The proportion of children in poverty has remained relatively stable at about 16 percent, but most recently, in 2003 (the last year for which statistics are available) that proportion was 18 percent.4 Moreover, 29 percent of children live below 150 percent of the federal poverty level (FPL) and 39 percent live below 200 percent of the FPL.4
Although the absolute numbers and the recent suggestion that the percentage may be increasing are of concern, the more important point is that children are the one age group in U.S. society whose financial status has not improved over time.5 In fact, in the 1960s, before the institution of Medicare, elderly people were considered the most financially disadvantaged age group. However, over the ensuing years the rate of poverty among the elderly population has fallen by almost two-thirds, but the proportion of children who are poor today is about the same as it was in the 1960s. At that time, children were less likely to be poor than the elderly, but today, the rate of poverty among children exceeds that among elderly people, so that about two-thirds more children than elderly people are poor.5 Among minority children in 2003, 29 to 34 percent of black and Hispanic children were poor, whereas 9 percent of white non-Hispanic children were poor.5
A recent paper also suggests a growing intergenerational inequity in public spending.6 Between 1965 and 2000, per capita public spending grew more rapidly for elders than for children, so children today are getting a smaller share of the pie.
Children in poverty are much more likely to be rated as having poorer health than other children.7 Among the children in families with incomes below the FPL, 71 percent of the children are rated to be in excellent or very good health, whereas 89 percent of the children in families whose incomes are above 200 percent of the FPL are rated to be in excellent or very good health.4
Although the rates of disability among the young are considerably lower than those among people in older age groups, disability is neverthe-
less of great importance in the child population. Among the reasons for this importance is that those with disabling health conditions have long-term survival, with the overwhelming majority now surviving well into adulthood. Thus, disabilities in children result in extremely large cumulative costs to society, their family units, and the individual members of their families. Moreover, the health of children as a group is critical to society because of the key role that children play as perhaps the single most precious of society’s natural resources. Undoubtedly, therefore, the health of children is integrally linked to the health of the nation’s and society’s future.
In addition, as was recently highlighted in a report of the Board of Children, Youth, and Families, Children’s Health, the Nation’s Wealth: Assessing and Improving Child Health,8 a child’s health has been demonstrated to have effects that may reach far into adulthood. It is becoming increasingly evident that many, if not most, of the most important causes of adult disability have their origins during childhood, even though they may not be causing any obvious health problem or disabilities during childhood. This is a key point, and its implications will be discussed below.
The predominant notion of disability in the United States is derived from two central concepts based on the effects of injury and illness for adults: (1) the effects of having a condition on one’s ability to work to support oneself and (2) self-care as reflected in independence in activities of daily living (ADLs). Other concepts of disability, such as the International Classification of Functioning, are relatively new.9 This creates a problem when early life and disability are being discussed, because neither the concept of work nor the concept of independence in ADLs is an adequate way of defining disability among children. Independence in ADLs is not the norm, especially in younger children, who are fully dependent, and only a few of the very most severely impaired older children actually qualify as being disabled by virtue of limitations of ADLs. In 1987 the Office of Technology Assessment of the U.S. Congress attempted to estimate the number of children with dependence on technology for bodily function and ADLs. Its estimates varied up to 40-fold, depending on the level of technology dependence and the criteria used.10
Moreover, the functioning of children is always a moving target, as children mature at different rates, live in different cultures with different expectations of independence and self-sufficiency, and grow up in environments that vary markedly in the demands that they place on the performance of activities by children. All these factors make it hard to conclude whether a child’s function is or is not within the normal range by the use of a short set of questions or some other relatively efficient modality.11
These difficulties with the assessment of functioning in children leave only traditional measures of assessment that are built on the notion of work. For children, substitutes for work have customarily been activity
limitations in schoolwork for those over age 5 years and play for those under age 5 years. I would submit that there are very few even very impaired children who cannot play. So this leaves a rather significant paradox, in which very few children under age 5 appear to be disabled, even though many of these children are considered disabled as they age without experiencing changes in their health status or level of impairment. This suggests that current techniques for the identification of the level of disability in children are not optimal. Another reason for the rise in the rate of the disability with age is that additional children become disabled later as a result of either disease progression or the onset of new conditions or injuries.
Nevertheless, questions about functional limitations affecting participation in school and play have been used and are very similar to the types of questions used for adults that Jay Bhattacharya and colleagues reviewed in Appendix F. Such questions have been used rather frequently in national surveys over time. As mentioned in other appendixes in this volume, inconsistencies in the wording of the questions for children and adults have occurred across data sets and over time, but questionnaires have been a major means of tracking disabilities in children. Wording changes have been tracked in a number of studies12,13 and are not reviewed here, but it should be noted that they may contribute to some of the changes in trends over time. It should also be emphasized, however, that the trends have continued even during periods in which the wording and administration of data collection have been stable. In a 1998 paper, Newacheck and Halfon14 showed that in the 1960s the rate of functional or activity limitations in children was a little below 2 percent; in the early 1980s this rate went up to about 3.5 percent. These rates continued to increase to the present rate of about 7 percent.13
Another study that examined the differences between white and black non-Hispanic children showed differences in the raw percentages of functional or activity limitations in children. It appeared that higher proportions of black children and youth than white children and youth are affected, but poverty accounted for all of these patterns. When the data were controlled for poverty, the differences completely disappeared.13 The bottom line is that over a 40-year period, the proportion of children reported to have major limitations in their activities related to play and school has gone from less than 2 percent to close to 7 percent. Within the population, all studies show higher rates of major activity limitations among older children than younger children and higher rates among males than females. At present, children over the age of 5 years are consistently reported to have rates of major activity limitations over 8 percent.4
The overwhelming majority of limitations in major activities is related to school participation and the need for special education. The higher level of special education for males is responsible for most of the difference by
gender. Relatively few children are reported to be limited in major activities in any other way.4 Additionally, broader definitions such as the one adopted by Child Trends Data Bank report that as many as 18 percent of 5- to 17-year-olds have at least one limitation of activity.15
Another way of looking at the traditional measures of disability is examination of the rate of Supplemental Security Income (SSI) enrollment among children. These data provide only a rough notion of the rates of disability among children because they are limited to poorer children, a consequence of the program’s income restrictions that make children in middle-class homes ineligible for this benefit. In addition, the eligibility criteria for SSI require the child to have a significant impairment or a fatal condition. Nevertheless, according to those statistics, a rather significant rise in the number of children reported to have disabilities could be seen in the 1990s. In the early 1990s about 290,000 children16 were receiving SSI, but by the end of the decade that number rose to 960,000.17 However, that increase is due in large part to two concurrent changes: the U.S. Supreme Court decision in Sullivan v. Zebley, which broadened the criteria for eligibility,18 and a revision of the medical standards for assessing health mental impairment, which increased the number of qualifying conditions from 4 to 11.19
Although the methods described above are the traditional ways in which disability has been measured in children and youth, there are real problems with these work-related measures. First, there is a lack of a baseline against which functioning in children can be measured. Children are involved with the acquisition of function; except in instances of later onset of illness or injury, it is not possible to assess what the child would have been able to do if he or she did not have the condition that caused the disability. That is, children are involved in habilitation, not re-habilitation. To complicate matters more, wide variations in cultural norms and expectations exist across society, and these influence how children are assessed. Wide variations in what is accepted as normal development also exist.11,20 The standard that children must meet before they are considered to have a disability is actually pretty low (and one that most healthy children pretty easily exceed). In real instances children and youth are not considered disabled because they have previously functioned above normal, and when their functioning is impaired to the level that meets the minimum baseline expectations, they continue to be classified as being free of a disability.
A recent example is an adolescent who had won an athletic scholarship to a university and who had been recruited for a professional baseball team. He then developed a condition that impaired his arm so that he could not play baseball. This adolescent is going to lose both his educational opportu-
nity and a very major activity in his life, but he would probably not be counted as disabled by the standards that are now used.
However, the real problem with the use of school or play as the work analog for children and youth is that the real work of childhood is maturation and development of the child’s potential—the acquisition of new capacities and skills—not going to school and playing. This notion of child health was recently endorsed by the Board of Children, Youth, and Families (Institute of Medicine and National Research Council) report on the health of children. It defines children’s health as “the extent to which individual children or groups of children are able or enabled to develop and realize their potential; satisfy their needs; and develop the capacities that will allow them to interact successfully with their biological, physical and social environments”8 (page 33). In the assessment of disabilities in children, individuals who are unable to develop and realize their potential, satisfy their needs, or interact with their environments—rather than those who cannot play or participate in school—should be counted. That is quite different from the current standard.
An important consideration in the search for alternative ways of measuring disabilities in children is the measurement of chronic conditions that lead to disability over time in childhood as well as those that may portend disabilities in later life. This is critical, because it is known that many chronic conditions lead to disability, especially if they are inadequately treated. From a prevention focus, the time to identify and catch chronic conditions and minimize the chance that they will produce a long-term disability is early in the child’s life. This is important to avoid the extra burdens of disability over a lifetime.
In comparison to disabilities in adults, which tend to result from a cluster of relatively frequent conditions that account for a large portion of the burden of disability, the range of chronic conditions that produce disabilities among children is extremely diverse; and a far larger proportion of the burden of disabling conditions in childhood results from very rare conditions. The percentage of children with chronic conditions has been estimated by the use of a variety of tools and a large number of national surveys over the years and has varied from a low in the single digits to a high in the low 30s.21,22 In the Child Health Supplement of the 1988 National Health Interview Survey, conducted by the National Center for Health Statistics, the estimate was 31 percent.23 It should be noted that these percentages are based on counts of conditions, not of children.
A series of concerns surrounded these estimates, especially because a child, once he or she is labeled as having a condition, was counted forever after. Additionally, it seemed important to distinguish those who carried a label from those who were actually experiencing a consequence of having a condition. Furthermore, the condition lists used to identify children in such
surveys were long and cumbersome and inevitably were unable to list all conditions, so that those with serious and disabling but rarer conditions were likely being missed. These factors, combined with the growing evidence that raising a child with a wide variety of health conditions posed similar challenges and required similar assistance from service systems, led to the endorsement of what has been called the “noncategorical” or “generic” approach to the identification of children with chronic conditions. It is based on continuing concern that the large number of individual conditions cannot be inventoried, that there have been negative effects of providing services and advocacy for each condition separately, and that there are inherent inequities in doing so.24,25,26,27
In 1993, my colleagues and I at the Albert Einstein College of Medicine proposed that children with chronic conditions could be identified by the consequences of their conditions.27 Three major types of consequences were identified: whether the conditions imposed functional limitations compared to the functioning of their age-mates, whether they produced dependence on compensatory mechanisms or assistance, or whether they required more than the usual level of services. Others have since published similar definitions28 and endorsed similar concepts,29 and a variety of noncategorical or generic approaches have been used. One advantage of this approach is that it counts children, not conditions, and it has been estimated that between one-third and one-half of children have more than one condition.
A number of tools have been developed to operationalize these concepts, and three of these employ survey techniques: the Questionnaire for Identifying Children with Chronic Conditions (QuICCC), which has 39 items30; the Questionnaire for Identifying Children with Chronic Conditions—Revised (QuICCC–R), which has 16 items31; and the Children with Special Health Care Needs (CSHCN) screener, which is the shortest (and which some have called the “quickest”), with 5 items.32 Only the first of these instruments was specifically designed for epidemiological purposes. However, the brevity of the CSHCN screener, which was originally designed for the purposes of quality assessment in health insurance plans, and its ability to identify so many children with disabilities quickly has made it an appealing instrument for many other purposes, including estimation of the prevalence of disabilities within populations. It has now been incorporated into many large-scale surveys on a national basis, such as the National Survey of Children with Special Health Care Needs (2000, 2005), the National Survey of Child Health, the National Health Interview Survey, and the Medical Expenditure Panel Survey.
The most comprehensive of these instruments, QuICCC, was almost entirely incorporated into the National Health Interview Survey on Disability (NHIS-D; 1994–1995) and produced estimates of rates of disability in children between 14.8 and 18 percent from the NHIS-D by the use of
different analytic algorithms.33,34 As in the case of the estimates obtained by using traditional measures of activity limitations, the numbers increased with the ages of the children, were higher for males, and were considerably higher among those whose family incomes were at or near the poverty level and whose parents had low levels of education.34
More recently, the CSHCN screener has been used in the State and Local Area Immunization Telephone Survey Children on Special Health Care Needs, funded by the Maternal and Child Health Bureau, and produced an estimate of disability among children of 12.9 percent.35 Later national estimates of the rates of disability among children based on the CSHCN screener from the Medical Expenditure Survey suggest a rate of 17.6 percent. It is unclear whether this represents a true increase in prevalence or is related to differences in the methodologies related to the implementation of the CSHCN screener. The 2005 National Survey of Children with Special Health Care Needs is in the field and is again using the CSHCN screener.
Before leaving discussion of this approach to identification, it is worth noting that significant numbers of children have conditions that cause considerable consequences and are not identified by the functional limitations questions, even in the longest instrument. In fact, only 49 to 66 percent of the children identified as having disabilities by QuICCC are identified by 16 functional limitations items alone.30,33,36 Most importantly, among the more than 50 diagnoses inventoried in the validation study, children with the same diagnoses were found to have different types of consequences as a result of their disabling conditions. Except for the fact that children with more types of consequences were more likely to have multiple diagnoses than those with only one type of consequence, the type of consequence was not useful in identifying classes of children with disabilities.36 For example, a child with asthma might have only functional limitations or, when he or she was properly treated, might have been identified only by dependence on medication or might have intermittent inadequately treated asthma and experience only increased service use. This underscores the deficits that occur from the use of only functional limitations for the identification of children with disabilites.36
TRENDS IN DISABILITY
To evaluate trends in disability, it might be useful to look at a few examples for which data on trends in childhood disability are available. These trends have gone in different directions and make it hard to provide a single message about what is happening with childhood disability.
Blood Lead Levels
From 1976 to 1980, 88.2 percent of children 1 to 5 years of age had blood lead levels that were more than 10 micrograms of lead per deciliter. By 1988 to 1991, that figure had fallen to only 8.6 percent of children in that age group. It continued to fall so that in 2003 only 1.6 percent of children ages 1 to 5 years had blood lead levels of 10 micrograms of lead per deciliter.37 This is a result of major changes in the environment, especially as a result of the elimination of the use of lead paint and lead-containing gasoline.8
Although this particular threshold of the blood lead level is not associated with measurable disability, it is a good marker for the level of population exposure to environmental lead, which at high levels of body burden is associated with significant cognitive as well as behavioral effects in children. High blood lead levels cause significant disability and even at low levels cause some impairment of the intellectual quotient.
Failure of the neural tube to fuse during fetal development is a cause of major childhood disability, regardless of the measure used. Another example of a success in lowering disability among children has resulted from the unraveling of the complicated interaction of genetics and environmental folic acid deficiency during pregnancy in a vulnerable subset of the population.8 The inclusion of folic acid supplements to women of childbearing age has led to marked declines in the rates of both spina bifida and anencephaly. The Neural Tube Defect Ascertainment Project reported a 31 percent decline in the rate of spina bifida after fortification of the mother’s diet with folic acid and a 19 percent decline in the rate of anencephaly.38 Thus, the rate of the two conditions combined decreased from 7.6 to 5.4 per 10,000 births in a 3-year period (1997 through 1999).
Asthma presents a very different pattern, one of increasing disability among children. There has been a steady increase in the proportion of U.S. children with asthma. A dramatic rise in the prevalence of asthma has been seen since the 1980s, with the rate almost doubling from the 1980s to the present.39 Although many cases are mild and do not cause the significant disability seen in children with neural tube defects, the prevalence is much higher. More than 5 percent of children have asthma, which makes the prevalence of asthma 100 times greater than that of spina bifida, so that even if only a small percentage of children experience disability because of
asthma, the increase in their numbers significantly overshadows the decline in the numbers of children with spina bifida.
Low Birth Weight
One of the successes of modern medicine has been the survival of low-birth-weight infants. As the rate of mortality has declined and the rates of low birth weight and preterm birth have increased somewhat, more children in the low-birth-weight population are surviving.40 Although most low-birth-weight children do well, these survivors still experience far higher rates of morbidity and more significant rates of disabilities than children born at term with higher birth weights.41 Another concern is that the infant mortality rate is not decreasing at the same rate among all the racial and ethnic groups. As a result, there are real disparities in the rate of infant mortality by race and ethnicity. The trends in low birth weight are also not evenly distributed, so that the rate of low birth weight among blacks is higher than that among the rest of the population.42 A rapid rise in multiple births is also being seen, which is another reason for the increases in the rates of low birth weight and disabilities associated with low birth weight.41
Members of the IOM committee have probably seen information from the Centers for Disease Control and Prevention depicting the rise in the proportion of the adult population experiencing obesity from less than 10 percent in most states to more than 25 percent in many states. This is an issue that has not been limited to adults.
In the 1960s the rates of obesity among children were 4 to 5 percent. Now the rates of obesity, considered a body mass index (BMI) more than 2 standard deviations above the mean for age and gender, are consistently close to 15 percent among adolescents and late-school-age children. This is an epidemic with great consequences for the future. Again, people in minority groups are the most affected. The rates among black and Mexican-American children have been increasing over time, and the rates of obesity are increasing among males as well as females. Among older Hispanic children, one in four meets or exceeds the criterion for being over-weight.43
Although many people do not consider obesity per se to be a disability, this dramatic rise in the rate of obesity among children has major implications for the rates of disability for the children now and later in their lives. Excess BMI is associated with a variety of the conditions that cause disability, including cardiovascular problems, type II diabetes mellitus, hyperten-
sion, and hypercholesterolemia. The incidence of type II diabetes among children and adolescents has increased dramatically over time and at the Children’s Hospital at Montefiore, the incidence equals that of type I diabetes, something that was previously unheard of.44
Finally, two major omissions in this summary are noteworthy. First, the presenters were asked to exclude from discussion children who are institutionalized, who are not surveyed in any of the household-based surveys, and mental health morbidity, which is a major growing cause of disability among children and youth. Nevertheless, it would be inappropriate not to mention that institutionalized children are among those with the most severe disabilities. In addition, mental health conditions are among the leading causes of disability among children and youth and have extremely important implications for long-term well-being and functioning.
So what does the information presented here mean for the IOM committee? Current approaches to childhood disability that focus only on the severely disabled show low rates compared with those of adults, with a modest but consistent upward trend among children and youth. These disabilities have major implications for the young people who experience them, for their families, and for society as a whole. However, if the purpose of disability policies is to minimize impairment and lifetime disability, then I believe that the current emphasis on those with the most severe childhood disability is misguided.
I think that a broader conceptualization for the assessment of milder disabilities and conditions among the young is needed so that the precursors of disability that would allow the minimization of future impairment and disability can be tracked. Furthermore, this mismatch of agendas, in which the focus is on those who already are experiencing severe impairment, is going to increase dramatically as the genomic revolution allows children who are at biologic risk to be identified much earlier. This has real implications for the specialized preventive services that those children are going to need to prevent or postpone the onset of their disabilities.
There is an urgent need to address these issues and to redefine how society identifies and thinks about disability among the young, both for the children themselves and their current health and for their future health and their ability to be productive members of society in the long run.
1. U.S. Census Bureau Population Estimates and Projections. http://childstats.gov/americaschildren/pop1/asp. Accessed July 21, 2005.
2. U.S. Census Bureau Population Estimates and Projections. http://childstats.gov/americaschildren/pop3/asp. Accessed July 21, 2005.
3. U.S. Census Bureau Population Estimates and Projections. http://childstats.gov/americaschildren/pop3/asp. Accessed July 21, 2005.
4. Federal Interagency Forum on Child and Family Statistics. America’s Children: Key Indicators of Well-Being. Washington, DC: U.S. Government Printing Office, 2005.
5. National Center for Health Statistics. Health, United States, 2004, with Chartbook on Trends in the Health of Americans. Hyattsville, MD: National Center for Health Statistics, 2004.
6. Newacheck PW and Benjamin AE. Intergenerational equity and public spending. Health Affairs 23:142–146, 2004.
7. U.S. Census Bureau Population Estimates and Projections. http://childstats.gov/americaschildren/hea1/asp. Accessed July 21, 2005.
8. National Research Council and Institute of Medicine. Children’s Health, the Nation’s Wealth: Assessing and Improving Child Health. Committee on the Evaluation of Children’s Health. Board of Children, Youth, and Families, Division of Behavioral and Social Sciences and Education. Washington, DC: National Academies Press, 2004.
9. Chatterji S, Ustun B, Sadana R, Salamon JA, Mathers CD, and Murray CJL. The conceptual basis for measuring and reporting on health. Global Programme on Evidence for Health Policy Discussion Paper 45. Geneva, Switzerland: World Health Organization, 2002.
10. U.S. Congress, Office of Technology Assessment. Technology-Dependent Children: Hospital v. Home Care: A Technical Memorandum. OTA-TM-H-38. Washington, DC: U.S. Government Printing Office, May 1987.
11. Stein REK and Jessop DJ. Functional status II(R): a measure of child health status. Medical Care 28:1041–1055, 1990.
12. Newacheck PW, Budetti PP, and Halfon N. Trends in activity-limiting chronic conditions among children. American Journal of Public Health 76:178–184, 1986.
13. Newacheck PN, Stein REK, Bauman LJ, et al. Disparities in the prevalence of disability between black and white children. Archives of Pediatrics and Adolescent Medicine 157:244–248, 2003.
14. Newacheck PW and Halfon N. Prevalence and impact of disabling chronic conditions in childhood. American Journal of Public Health 88:610–617, 1998.
15. Child Trends Data Bank. www.childtrendsdatabank.org. Accessed July 21, 2005.
16. Perrin JM and Stein REK: Reinterpreting disability: changes in SSI for children. Pediatrics 88:1047–1051, 1992.
17. Social Security Administration, Office of Policy, Office of Research, Evaluation and Statistics. Children Receiving SSI, December 2003. SSA Publication 13-11830. Washington, DC: U.S. Government Printing Office, 2005.
18. Sullivan v. Zebley, 88–1377. U.S. Supreme Court, February 20, 1990.
19. Perrin JM, Kuhlthau K, McLaughlin TJ, Etner SL, and Gortmaker SL. Changing patterns of conditions among children receiving Supplemental Security Income disability benefits. Archives of Pediatric and Adolescent Medicine 153:80–84, 1999.
20. Stein REK. Measurement of children’s health. Ambulatory Pediatrics 4:365–370, 2004.
21. Gortmaker S and Sappenfield W. Chronic childhood disorders: prevalence and impact. Pediatric Clinics of North America 31:3–18, 1984.
22. Jessop DJ and Stein REK. Consistent but not the same: effects of method on chronic conditions rates . Archives of Pediatrics and Adolescent Medicine 149:1105–1110, 1995.
23. Newacheck PW and Taylor WR. Childhood chronic illness: prevalence, severity, and impact. American Journal of Public Health 82:364–371, 1992.
24. Pless IB and Pinkerton P. Chronic Childhood Disorder: Promoting Patterns of Adjustment. Chicago, IL: Year Book Medical Publishers, 1975.
25. Hobbs N and Perrin JM. Issues in the Care of Children with Chronic Illness: A Sourcebook on Problems, Services, and Policies. San Francisco, CA: Jossey-Bass, 1985.
26. Stein REK and Jessop DJ. What diagnosis does not tell? The case for a noncategorical approach to chronic illness in childhood. Social Science and Medicine 29:769–778, 1989.
27. Stein REK, Bauman LJ, Westbrook LE, Coupey SM, and Ireys HT. Framework for identifying children who have chronic conditions: the case for a new definition. The Journal of Pediatrics 122:342–347, 1993.
28. McPherson M, Arango P, Fox H, et al. A new definition of children with special health care needs. Pediatrics 102:137–140, 1998.
29. Perrin EC, Newacheck P, Pless IB, Drotar D, Gortmaker SL, Leventhal J, Perrin JM, Stein REK, Walker DK, and Weitzman M. Issues involved in the definition and classification of chronic health conditions. Pediatrics 91:787–793, 1993.
30. Stein REK, Westbrook LE, and Bauman LJ. The questionnaire for identifying children with chronic conditions (QuICCC): a measure based on a noncategorical approach. Pediatrics 99:513–521, 1997.
31. Stein REK, Silver EJ, and Bauman LJ. Shortening the questionnaire for identifying children with chronic conditions (QuICCC): what is the consequence? Pediatrics 107(4):pe61, 2001.
32. Bethell C, Read D, Stein REK, Blumberg SJ, Wells N, and Newacheck PW. Identifying children with special health care needs: development and evaluation of a short screening instrument. Ambulatory Pediatrics 2:38–48, 2002.
33. Stein REK and Silver EJ. Operationalizing a conceptually-based noncategorical definition: a first look at US children with chronic conditions. Archives of Pediatrics and Adolescent Medicine 153:68–74, 1999.
34. Stein REK and Silver EJ. Comparing different definitions of children with chronic conditions in a national data set. Ambulatory Pediatrics 2:63–70, 2002.
35. van Dyck PC, Kogan MD, McPherson MG, Weissman GR, and Newacheck PW. Prevalence and characteristics of children with special health care needs. Archives of Pediatrics and Adolescent Medicine 158(9):88:4–90, 2004.
36. Westbrook LE, Silver EJ, and Stein REK. Implications for estimates of disability in children: a comparison of definitional components. Pediatrics 101:1025–1030, 1998.
37. Centers for Disease Control and Prevention, National Center for Environmental Health. Lead research, as quoted by Trends Data Bank. www.childtrendsdatabank.org. Accessed July 21, 2005.
38. March of Dimes, personal communication. Data from National Birth Defects Prevention Network-Centers for Disease Prevention and Control.
39. Centers for Disease Control and Prevention. Childhood asthma prevalence before and after the 1997 redesign of the National Health Interview Survey—United States. Morbidity and Mortality Weekly Report 49:908–911, 2000.
40. Martin JA, Kochanek K, Strobino DM, Guyer B, and MacDorman MF. Annual summary of vital statistics—2003 Pediatrics 115:619–634, 2005.
41. Hack M, Taylor HG, Drotar D, Schluchter M, Cartar L, Andreias L, Wilson-Costello D, and Klein N. Chronic conditions, functional limitations, and special health care needs of school-age children born extremely low birth weight in the 1990s. Journal of the American Medical Association 294:318-325, 2005.
42. U.S. Census Bureau Population Estimates and Projections. http://childstats.gov/americaschildren/hea5/asp. Accessed July 21, 2005.
43. U.S. Census Bureau Population Estimates and Projections. http://childstats.gov/americaschildren/hea3/asp. Accessed July 21, 2005.
44. Hannon TS, Rao G, and Arslanian SA. Childhood obesity and type 2 diabetes mellitus. Pediatrics 16:473–480, 2005.