Skip over navigation

Journal Issue: Children with Disabilities Volume 22 Number 1 Spring 2012

The Changing Landscape of Disability in Childhood
Neal Halfon Amy Houtrow Kandyce Larson Paul W. Newacheck

Surveying the Landscape and Moving Forward

This brief survey of childhood disability in the United States has revealed several important and interrelated findings that deserve additional consideration. First, the reported prevalence of disability has increased steadily since the 1960s, and at the same time the underlying health and environmental conditions associated with reports of disability have also substantially changed. Second, despite increases in prevalence, and the implementation and expansion of social programs to prevent and ameliorate childhood disability, social class disparities in disability have persisted at virtually the same level for the past fifty years. Third, the measures of disability used in national surveys lag behind current thinking about disability and its relationship to the environment. In addition, most available data on disability have come from cross-sectional surveys, that is, surveys that take a snapshot of the situation at a point in time and do not follow individual respondents over time, leaving gaps in our knowledge base about the dynamics and trajectories of disability as children with disabilities age. We take up each of these issues in turn, discussing their significance and implications for the future.

Increasing Prevalence over Time
Trend data from the NHIS demonstrate that the prevalence of activity limitations reported by parents across all condition categories has steadily increased over the past half century. There is no simple explanation for why rates of childhood disability have increased so dramatically. Indeed, the upward trend might be explained by a number of factors, including changes in survey procedures, increased exposures to risks and declines in resistance, changes in diagnoses, changes in access to health and educational services tailored to identify and treat children with disabilities, or other significant cultural shifts including expectations of what a typically developing child "should" be able to do.

Some of the increase in prevalence is attributable to changes in the NHIS data collection process. In particular, the questions used to identify activity limitations have changed over time. Significant changes in survey procedures and questionnaire items related to activity limitations occurred in 1967, 1969, 1982, and 1996. In some cases the questionnaire wording was broadened to be more inclusive (1967, 1969), and in other cases the emphasis shifted. For example, in 1996 the questions for school-age children shifted from an emphasis on limitations in ability to engage in school activities to enrollment in special education programs owing to health conditions. Conservatively assuming that all of the increases in prevalence occurring in those years were attributable solely to changes in survey procedures and questionnaire revisions, about one-third of the upward trend between 1960 and 2009 can be attributed to changes in the survey.

Large-scale improvements in access to health care also occurred during the study period through the enactment and implementation of programs such as Medicaid, the federal and state health care program for the poor, and the federal-state Children's Health Insurance Program (CHIP) for children in low- and moderate-income families. These programs made large numbers of children eligible for screening, diagnosis, and treatment of physical, developmental, and behavioral problems. Public assistance programs for families of children with disabilities also grew during this period through the enactment and subsequent broadening of the federal Supplemental Security Income program following the Supreme Court's Zebley ruling.49 Enrollment in SSI also confers automatic eligibility for Medicaid. As described earlier, landmark legislation for the education of school-age children with disabilities was enacted in 1975, followed in 1986 by legislation that provided states with funds to establish a comprehensive statewide, interagency, multidisciplinary approach for expanding evaluation, special education, and early intervention services to disabled children from birth to age three. Enactment dates for these programs and rulings are shown in figure 1. In reviewing the figure, readers should keep in mind that complete implementation of new programs often occurs years after the enactment of legislation, and the programs may never be fully funded.

Advances in medical care, treatment, and diagnosis are likely to have contributed to the rise in prevalence. For example, access to and improvements in the delivery of specialized care have resulted in a growing number of survivors of complex congenital disorders, prematurity, and cancer.50 Greater access to medical and dental screening through Medicaid's Early and Periodic Diagnosis and Treatment Program is likely to have led to increased identification of previously "hidden" conditions. Better tools to diagnose chronic conditions, particularly emotional and behavioral conditions, may also contribute to the upward trend in prevalence. Concepts of health and disease, and definitions of what constitutes a childhood chronic illness have also changed.51 As noted, the narrow medical and programmatic focus on physical defects during the "crippled children" era of the 1930s to the 1960s has evolved to include an expansive range of physical, developmental, emotional, and behavioral conditions.52

Besides improvements in access to health care services over the past half century, there have also been tectonic shifts in how the education system views and responds to children with mental, behavioral, and developmental problems. As a result of federal and state programs, more children have access to special education and early intervention services focused on mental, behavioral, and developmental conditions; thus, questions in the NHIS that identify the presence of a limitation in activity based on the provision of special education and early intervention services would be expected to mechanically increase reported prevalence.

Teasing out the contributions of each of these factors to the increase in prevalence is a daunting but critical task and beyond the scope of this brief article. However, considering trends in some of the major condition categories associated with the rising prevalence of limitation of activity can provide a useful starting point for a more careful and detailed examination.

A growing number of children are reported to suffer disability stemming from emotional, behavioral, and neural developmental conditions such as ADHD, autism, learning disabilities, and speech and language disorders. This trend could reflect a real change in the incidence of conditions caused by changing risk exposures during pregnancy and early childhood, or it could result from changes in recognition, screening, and diagnostic criteria. For example, autism, which was previously more narrowly defined and based on more severe symptoms, has been expanded to include a spectrum of pervasive developmental disorders that includes Asperger's Syndrome and other related disorders. The ongoing controversy regarding the validity of increased reports of childhood autism is indicative of the challenges inherent in attempting to understand the factors contributing to increased prevalence of childhood disability. Reported prevalence rates for autism have increased by several orders of magnitude in the past twenty years. Yet changes in recognition, diagnostic criteria, and incentives for early identification and intervention alone do not account for the increased prevalence.53 Several studies have also documented the relationship of higher rates of autism to age of parents at birth, birth spacing, breast feeding, and other social factors, as well as to exposures to a range of environmental toxins.54 Others have argued that increasing rates of autism, as well as of ADHD, are the result of a confluence of small but important changes in children's exposure to the combined influence of demographic changes, environmental toxins, toxic stress associated with poverty, and the instability associated with a lack of adequate resources for child rearing.55

The trend for childhood asthma is a special case in which the prevalence of disability owing to asthma increased dramatically over several decades only to apparently plateau since the turn of the century. Analysis by the Centers for Disease Control and Prevention showed that between 1980 and 1995 asthma rates in children were increasing at a rate of 5 percent a year.56 An analysis of NHIS data by two of the authors, Paul Newacheck and Neal Halfon, showed that between 1969–70 and 1994–95, the prevalence of asthma-related disability increased 232 percent, whereas disability in general from all other conditions increased by a much smaller 113 percent over the same period.57 These rapid increases in rates of disability related to childhood asthma were paralleled by rising hospitalizations and other indicators of the growing burden of asthma in children across the United States.58 Since 1997, however, the number of children with disabling asthma has leveled off and seems to be decreasing somewhat. The reasons for the spike in and subsequent plateauing of asthma rates are not clear, with hypotheses ranging from changes in infectious disease patterns to changes in activity levels, with more sedentary lifestyles that led to greater indoor exposures and less outdoor play time.59 The plateauing and declining rates of disabling childhood asthma also coincide with reductions in air pollution that have resulted from changes mandated by the Clean Air Act and its amendments. The absence of reliable longitudinal data makes this trend difficult to unravel.

Another trend of note is the rise in the number of preterm births. Long-term trends show that preterm rates peaked sometime in the mid-1960s and then steadily decreased over the next twenty years. From 1990 to 2006, however, rates of preterm birth increased from 10.6 per 1,000 live births to 12.8 per 1,000, with recent declines to 12.3 in 2008.60 While rates of preterm births for black mothers have remained persistently high (ranging between 18 and 19.5 per 1,000 for much of this period), rates for white mothers have increased from 8.0 to 10.2 per 1,000. Trends in most high-income countries are similar to those in the United States.61 Much of the overall increase for white mothers is attributable to an increase in late preterm deliveries by caesarean section. Between 1994 and 2007, the percent of children born via C-sections increased by 53 percent, from 21 percent of all births to 32 percent.62 Children born preterm are at risk for a number of short- and long-term neurodevelopmental disorders including cerebral palsy, intellectual impairment, sensory impairments, and ADHD.

Changes in perceptions, acceptance, and advantage associated with childhood disability may also help explain increased reporting of childhood disabilities. Over the past several decades, perceptions about the nature, impact, and mutability of behavioral, developmental, and emotional disorders such as ADHD have shifted significantly. Several factors, including the revolution in brain science and better understanding of the biological and developmental origins of behavioral, developmental, and emotional disorders, have served to reduce the stigma and encourage the acceptance of more aggressive early intervention, diagnosis, and treatment. ADHD, for example, once known as minimal brain dysfunction, was renamed and reframed after extensive natural history and epidemiologic research. That work now suggests that ADHD is not just a disorder of school-age children but a symptom complex that afflicts a growing number of individuals across the lifespan.63

Some commentators have questioned whether increased performance demands placed on children and young adults, particularly in school settings, are contributing to a growing reported prevalence of ADHD. A portion of the observed increase may be a response to a societal shift in perceptions and expectations of what is appropriate child behavior and greater acceptance of the ADHD diagnosis. While minimal brain dysfunction may not have been a condition that one would like to talk about at cocktail parties, ADHD has much less stigma and might also provide access to medications that can prove beneficial for achieving the next rung on the education or employment ladder. Greater demand for pharmacological interventions to enhance school performance is coincident with diminishing opportunities and fewer educational and career paths for children who do not succeed in gaining entry into traditional four-year colleges, creating growing pressure on families to do whatever they can to help their children succeed in school.

Other potential advantages come with a diagnosis of a specific developmental disability, including opportunities to receive early intervention and special education services. Children with a diagnosed learning disability also become eligible for "accommodations" when taking college admission tests. For those children attending private and parochial elementary and high schools, similar accommodations are often made both for admissions testing and even routine classroom quizzes and tests. Whether these accommodations also provide incentives for families to seek out such a diagnosis and designations is yet to be determined. In addition, the enactment and subsequent expansions of the SSI disability program for children and other public benefit programs have created other advantages for carrying diagnosed disability.

The Social Gradient
While the upward trend in childhood disabilities has shifted from physical and medical conditions to neurodevelopmental and behavioral conditions, the social gradient in prevalence of childhood disability is little changed. In other words, the country appears to have made no progress in reducing socioeconomic disparities in disability over the past half century. Lower socioeconomic status can adversely influence development, and severity of childhood disabilities as well as their potential for remediation. A growing body of literature documents not only the role that socioeconomic status plays in shaping preconditions of childhood disability, such as rates of prematurity, but also the likelihood that lower-income children will suffer from significant developmental delays, behavioral problems, and conditions such as ADHD.64

Research also demonstrates that the relationship between income and health gets stronger with age, and that children from low-income families often experience more health problems and have more difficulty recovering from those health shocks.65 Families with fewer resources also experience more barriers to receiving appropriate health and educational services and other necessary resources that can play a significant role in the capacity of the child and family to successfully adapt to or overcome the disability. The impact of socioeconomic status on the natural history of disabilities has been illustrated in cohort studies from England that compare cognitive and behavioral outcomes of low-birth-weight children and children born in the normal weight range. These longitudinal data show that low-birth-weight babies start life with greater rates of developmental delay, but that low-birth-weight babies from higher-income families surpass the function of even normal-birth-weight babies from low-income families.66 Better longitudinal data on the trajectory of developmental disabilities would go a long way toward understanding the role that social disadvantage plays in the origins, development, persistence, and resolution of disabilities in children.

Measurement of Disability and the Need for Better Data
Measures of disability are derived from conceptual models. Given the recent nature of the evolution of disability models from biomedical to those that are multidimensional and ecological, it is not hard to understand why there are gaps in how the origins, determinants, prevalence, trends, and trajectories of childhood disability are measured. The biomedical paradigm of disability measurement has dominated the field, and while the newer conceptual approach embodied in the WHO's international classification system and the UN Convention on the Rights of Persons with Disabilities is gaining ground, it has yet to be incorporated in national surveys in the United States. Moreover, the assessment of disability is plagued by the use of a hodgepodge of different definitions and measures. Current estimates thus tend to vary across surveys that are conducted by different organizations and agencies to fulfill very different measurement surveillance and administrative goals (see table 2).

Furthermore, most population-based data on disability are derived from cross-sectional surveys. To date, few longitudinal surveys, that is, surveys that follow the same individuals over time, have been conducted. Such surveys are needed to understand how disability develops, evolves, and potentially remits in U.S. children. The dynamics of disability in the child population is one of the areas where there has been a paucity of data. Without better data on the natural history of childhood disabilities, including changes in biological and clinical manifestations, as well as environmental adaptations and response to various types of interventions designed to treat and manage disability, health researchers will find it difficult to understand the difference between children who have severe and relatively static disabilities and those who might have moderate or mild disabilities that may be quite responsive to prevention, early intervention, and rehabilitation. For disabilities that arise as a result of major and often life-threatening health shocks (major illness, accidental or intentional injury) or through the cumulative impact of a series of smaller health shocks, the natural history of a disability will depend not only on disruptions to health and biological function but also on the capacity of the child, family, and their immediate and broader communities to respond. A young child with sufficient developmental plasticity or enough biological or behavioral reserves will respond quite differently to a health shock than will a child who has limited personal, familial, or community capacity or resources. For example, children who are born prematurely with neurodevelopmental vulnerabilities and respiratory dysfunction may follow a path of continued and increasing functional limitations, or they may be afforded the kinds of adaptive interventions that help them overcome their biological deficits and achieve normal developmental function. These alternative paths will depend on the nature of their biological deficit as well as on the ability of the family, the health system, and other community resources to rally in service of their adaptive developmental goals.

Going forward, a strategic measurement system that responds to the necessity for better and more actionable data is sorely needed. Such an approach must not only incorporate newer conceptualizations of disability but also respond to life-course models of how health develops so that better measurements of life history pathways, transitions, and turning points are routinely used. In creating this measurement strategy, it will be important to measure the origins, onset, and developmental trajectories of disabilities; to measure the risk factors that enhance the likelihood of developing chronic and disabling health conditions or factors that protect a child from experiencing those risks or actively promote optimal health development; and to include better measures of socioeconomic disparities and how those disparities develop over time. In addition, it is important to understand differences in individual disability development pathways as a basis for designing effective individual and population-based interventions aimed at preventing the development of disabilities in the first place and ameliorating the impact of those that do develop.

Improving Understanding of Childhood Disability
Different models of health, disease, and disability are influenced by and must respond to the dominant operating logic that organizes the structure and function of the larger health and health care system. The biomedical model of disability arose when health was narrowly defined by the absence of disease and the contributions of psychosocial factors were rarely considered or addressed. The ICF and UN convention focus on the role of environment and the importance of social participation reflects the ascendance of biopsychosocial models of human health and disability development that first emerged in the 1970s and have only recently gained prominence. This evolution in thinking has helped to broaden not only the conceptual understanding of disability but what constitutes appropriate measurement.

It has been argued that health systems are evolving toward a new era of health where the operating logic of the health and health care system will focus on optimizing the health of the entire population.67 As health researchers attempt to understand the causes of disability, explanatory models are becoming more sophisticated, shifting their focus from single and multiple risk factors to more dynamic, complex, and emergent factors organized around the development of health over the life course.68 Concepts of child health are also evolving beyond biopsychosocial constructs to embrace ideas from systems and complexity theory to describe the evolving qualities of health, disease, and disability as they develop dynamically. In 2004 the Institute of Medicine proposed a new definition of health in childhood: the extent to which children are able or enabled to "a) develop and realize their potential, b) satisfy their needs, and c) develop the capacities that allow them to interact successfully with their biological, physical, and social environments."69 This new definition describes health as a developmental capacity that enables individuals to achieve specific goals. This developmental definition, coupled with the UN convention definition, which highlights the interactional nature of disability, leads us to consider how best to define disability in children. We propose the following definition:

A disability is an environmentally contextualized health-related limitation in a child's existing or emergent capacity to perform developmentally appropriate activities and participate, as desired, in society.

This definition highlights the developmentally contingent relationship between health, functioning, and the environment; the gap in function between the child's abilities and the norm; and how this gap limits the child's ability to engage successfully with his or her world. Consistent with the ICF-CY and the UN convention, this more developmentally focused definition recognizes the dynamic nature of disability and how the experience of disability can be modified by factors in the child's environment. Also consistent with the newer framing of disability, our proposed definition does not require a specific diagnosis. It does require consideration of the continuum of health (from thriving to poor) as well as the continuum of disability (from enabled and flourishing to limited in all domains and functions). By incorporating notions of developmental potential and plasticity into considerations of disability, new strategies for intervention, remediation, adaptation, and accommodation can be considered even in the face of significant biological loss. Advancing this definition and conceptual approach will also provide new and better ways of understanding how children move along a health continuum from disability to flourishing, promoting ways for children to meaningfully engage in relationships, educational opportunities, and other activities that allow them to achieve happiness and life satisfaction.70