Summary of Workshop Presentations and Discussions
Alan Jette, chair of the Institute of Medicine (IOM) committee that planned the workshop, began by welcoming the participants. He noted that the workshop was the first step in a broader IOM project that would revisit the 1991 IOM report Disability in America and the 1997 report Enabling America. Dr. Jose Cordero (Centers for Disease Control and Prevention), Mr. Steven Tingus (National Institute on Disability and Rehabilitation Research), and Dr. Michael Weinrich (National Center on Medical Rehabilitation Research) provided additional welcomes. They also offered brief comments on the relevance of the workshop and the broader IOM project to the priorities and agendas of their organizations.
The morning session of the workshop was organized around two panels, the first of which focused on models and concepts of disability. The second panel provided overviews of disability trends in childhood, midlife, and late life. The first of the afternoon panels considered disability issues over the human life span, specifically, risk factors for disability late in life and transitions or transfers in care for adolescents and young adults. The other two afternoon panels provided perspectives on the nature and the prevention of secondary health conditions.
As noted in the introduction, this summary is based on the presentations and discussion during the workshop and does not necessarily reflect the views of the IOM committee that organized the meeting. The committee is preparing a comprehensive report that will cover a range of issues, including the topics discussed in the workshop. That report, which will include recommendations, should be released in early 2007.
MODELS AND CONCEPTS OF DISABILITY
Dr. Gale Whiteneck began with a brief history of models of disability. (The paper prepared by Dr. Whiteneck appears in Appendix B.) He noted in particular the seminal contribution of Saad Nagi in differentiating disability-related outcomes at three levels, namely, the organ, the person, and society. Dr. Whiteneck then focused on an assessment of the transition from the International Classification of Impairments, Disabilities and Handicaps (ICIDH) to the International Classification of Functioning, Disability and Health (ICF), which were published by the World Health Organization (WHO) in 1980 and 2001, respectively (WHO, 1980, 2001). ICF and its associated description offered one major conceptual step forward but also one step backward. Drawing on his experience with the classification revision process, Dr. Whiteneck proposed that several additional steps are needed to revise and apply the model.
The major step forward with ICF was the inclusion of physical, social, and other environmental factors that interact with an individual’s health conditions and other characteristics to produce outcomes, including activity (defined as the execution of a task or action by an individual) and participation (defined as an individual’s involvement in a life situation). This step was consistent with the recommendations of the IOM committees in 1991 and 1997 (IOM, 1991, 1997) and also with the provisions of the Americans with Disabilities Act (ADA; P.L. 101-336), which became law in 1990.
Other improvements in the classification involved the incorporation of neutral as well as negative ways of describing an individual’s status. For example, the ICF model uses the neutral phrase “body function and structure” as well as the term “impairment” and includes the term “activity” as well as “activity limitation.” The revised model also emphasized the complexity of possible interactions among its components (e.g., as shown by the two-way arrows rather than one-way “causal” arrows in the model’s graphic representation).
The step backward with ICF was the blurring of the conceptual distinctions between activity and participation (and between activity limitations and participation restrictions). Rather than identifying certain difficulties as activity limitations and other difficulties as participation restrictions, ICF grouped them together in a single list. Among the many differences between the two concepts, the foremost is that activity (e.g., walking) operates primarily at the person level, whereas participation (e.g., working) operates at the social and societal level. Activities are generally simpler than participation, and participation is more dependent on environmental factors than
activity. Participation also appears to be more relevant to the individual’s quality of life.
Dr. Whiteneck proposed that a first step in future revisions of ICF should be to distinguish between the elements of activity and the elements of participation. Such distinctions would be particularly useful in helping to understand and respond to environmental barriers to participation. Dr. Whiteneck emphasized the complexity of linkages between environmental factors and participation. For example, those who have higher levels of social participation may report higher levels of environmental barriers because they encounter and perceive more such barriers than those who participate less.
In addition to distinguishing between difficulties with activities and difficulties with participation, Dr. Whiteneck offered six additional steps to complete the ICF. These steps were to
add quality of life (an individual’s subjective assessment of his or her overall well-being) to the ICF model as recommended by the 1997 IOM report;
provide more theoretical and empirical specificity about the nature and complexity of environmental factors as they affect participation;
identify and assess personal factors, including psychological and behavioral characteristics, that affect activity and participation;
refine the graphic representation of the model so it would better help people to understand and use the model’s concepts and relationships;
develop a research strategy, including better measures and data resources, to test the model and its complex set of interrelations; and
design and test interventions that are consistent with the model to assess whether they improve the lives of people with disabilities.
In the second presentation, Dr. Rune Simeonsson noted his agreement with Gale Whiteneck’s discussion of ICF and his conclusions. He then highlighted current issues in defining and classifying disability among children and youth. (The paper prepared by Dr. Simeonsson appears in Appendix C.) He emphasized the need to move beyond classifications designed solely for health care to those applicable to educational and other service systems. Dr. Simeonsson also argued that we can learn from history, specifically, the evolution of models and concepts of disability across different disciplines.
Dr. Simeonsson reviewed historical models of child disability, emphasizing the transactional model of Sameroff and Chandler (1975). The premise of the transactional model is that outcomes for children with disabilities result not only from an initial cause—for example, having genetic syndrome—but also result from ongoing “transactions” in the child’s life,
including medical and educational interventions intended to improve outcomes. Both nature and nurture contribute.
Among the top current issues or challenges in conceptualizing, modeling, and measuring childhood disability are the changes in functional capacities that occur as children develop and the scientific complexity of differentiating developmental delay from impairment. Children are “moving targets.” What defines normal development in a two-year old is very different from that in a four-year old or a 14-year old. For very young children, their developmentally limited verbal and behavioral repertoires complicate assessments. The interviews and self-report tools that are frequently used with adults are not appropriate.
Dr. Simeonsson noted that additional challenges are presented by the inconsistency and arbitrariness of language used to describe disabilities and the lack of uniformity in how government agencies and service systems identify and categorize children with disabilities. He noted that some programs work more from a medical model and focus on diagnostic categories, whereas others—consistent with a social model of disability—focus on functional abilities and limitations, which is consistent with the ICF model.
Dr. Simeonsson suggested a number of priorities for researchers and policymakers. These include
adoption of the version of the ICF for children and youth (ICF-CY), when it is completed; and
development of measures of human functioning that operationalize all the components of the ICF for adults and children.
implementation of uniform concepts of child functioning and disability for health, education, and related services nationally and internationally; and
identification and refinement of developmentally relevant measures risk indicators for disability among children.
In her presentation, Dr. Julie Keysor tackled the question raised by both earlier speakers: how does the environment influence social participation and disability? In the absence of data on participation and participation restrictions, she focused on conceptual work and research related to mobility limitations and the environment. (The paper prepared by Dr. Keysor appears in Appendix D.)
Dr. Keysor reviewed several conceptual frameworks, beginning with ICF. The ICF explicitly mentions three general environmental domains: the physical, the social, and the political (WHO, 2001). It also identifies specific aspects of the environment that should be considered in research, policy, and other work: (1) products and technology; (2) natural and
human-made changes; (3) social support and relationships; (4) attitudes of other people; and (5) services, systems, and policies.
Dr. Keysor noted that the Quebec Group (Fougeyrollas and colleagues) has identified generally similar environmental factors ranging from the political-economic (e.g., public policies) to the socio-cultural (e.g., attitudes and norms) to the physical (e.g., natural or built features of the environment) (see, e.g., Fougeyrollas, 1995 and Fougeyrollas et al., 1997). Shumway-Cook, Patla and colleagues (2002) have elaborated specifically on aspects of the physical environment that influence mobility. They have identified eight dimensions related to distance, terrain, time (e.g., time to cross a street), physical load (e.g., carrying objects), need for postural transitions (e.g., change of direction), crowding or density, attentional demands (e.g., familiarity of surroundings), and weather and light levels.
Dr. Keysor also observed that she and her colleagues, including Alan Jette, have proposed a different way of classifying physical domains (Keysor et al., 2005). Their classification identifies home mobility barriers, community mobility barriers, mobility technology facilitators, communication technology facilitators, and transportation facilitators.
Related to these frameworks and others are several measurement instruments that can be used for epidemiological and observational studies. The research challenges are to assess which domains of the environment have the greatest influence on participation or disability and also to identify which elements act as participation barriers and which as facilitators.
Dr. Keysor described the Craig Hospital Inventory of Environmental Factors, which was developed by Gale Whiteneck and his colleagues (Whiteneck et al., 2004). This is a self-report instrument that asks people how often in the past 12 months an environmental barrier has been a problem and for an identified problem whether has it been a big problem or a little problem. Other self-report instruments, for example, the Measure of the Quality of the Environment developed by Fougeyrollas and the Quebec group, add a focus on environmental facilitators as well as barriers.
Based on her search of the research literature from 1991 to 2005 related to rehabilitation, stroke, spinal-cord injury, and arthritis, Dr. Keysor reviewed the handful of studies relating mobility limitations to environmental factors. In general, these studies found that people with mobility limitations do report facilitators and barriers in their environments and that they try to avoid the barriers. The research findings are, however, not that consistent or strong, and they raise many questions. For example, it is not clear from the research how long environmental effects or perceptions endure. Perhaps people adapt to the barriers they encounter, for example, by finding technologies, social supports, or other means to cope with the barriers, or perhaps they change their perceptions of their environment.
Although the growing number of measurement instruments is a plus,
the lack of consensus on domains of environment presents challenges. The paucity of studies argues for further investigations, especially ones that are prospective and experimental. In addition, researchers need to focus more on how people interact with their environments at work, school, and else-where. Finally, Dr. Keysor seconded Dr. Whiteneck’s call for more emphasis on assessing quality of life and life satisfaction.
One participant noted the lack of use of ICF in framing surveillance and research activities and asked if it is necessary to modify the conceptual framework or just move forward with the research. Dr. Simeonsson responded that one reason for this lack of use is that thus far most of the research relies on preexisting instruments and data sets, although new resources have been developed since the publication of ICF. Drs. Whiteneck and Keysor supported the need for feedback from research use of the models to refine or elaborate on the conceptual and classification model, but they said that it is not necessary to wait for the perfect model. ICF can be used now. In a similar vein, another participant wondered—given the different meanings of words to different people—whether or not the limits of communication have been reached and whether it should just be recognized that there is noise and uncertainty in the use of these concepts.
One participant probed further the proposal that models of disability incorporate quality of life. Dr. Whiteneck suggested that the best approach would be to keep quality of life in a separate domain and not try to integrate it into all domains of ICF. He argued that the essential aspect is a person’s subjective, overall assessment of well being rather than only health-related quality of life, which tends to dominate now. Dr. Simeonsson noted that quality of life is not isomorphic or perfectly correlated with activity or participation. That is, people with similar limitations in those areas may have very different assessments of their quality of life.
The remaining discussion focused on the relation to ICF of particular concepts that have been highlighted in the disability literature. One question involved the implications of all the two-way arrows in the ICF graphic model, including the implications for a life-course perspective and the malleability of health. Dr. Simeonsson responded that the concept of early intervention certainly existed before ICF.
Another participant asked where the family fits into ICF. All the panelists agreed that families are a critical part of the environment. Dr. Whiteneck added that involvment in family life is an important dimension of participation but that family structure or behavior can also be an environmental barrier or facilitator of participation. For example, people with traumatic brain injury have cited family attitudes as an important barrier. A final
question was whether ICF includes the concept of psychological accommodation, in which, over time, people with disabilities change their perspectives about the adequacy of their own functioning and develop a subjective sense of well being that may be quite different from what an outside observer might expect. Dr. Simeonsson responded that he was not sure where it fits into ICF, but there is ample evidence that individuals’ conceptions of what is important in life change as they live with disability.
TRENDS IN DISABILITY
Dr. Vicki Freedman, whose task was to discuss trends in late-life disability, reviewed perspectives on the implications of increased life expectancy for morbidity and disability in late life. (Dr. Freedman’s paper appears in Appendix E.) Life expectancy was 68 in 1950 and reached over 77 in 2002. This increase has raised the question: are these extra years of life associated with longer periods of morbidity and disability? Some have predicted increases in late-life morbidity and disability, some have predicted decreases, and others have hypothesized an increase in chronic conditions but a decrease in their progression to disability in late life.
Dr. Freedman stressed that disability in late life is a socially-defined concept that reflects the intersection of an older person’s capabilities, their environment, and the nature of the tasks that they wish to accomplish. In practice, most studies of trends in disability among the older population focus on self-reports of difficulty or assistance with activities of daily living (ADLs; e.g., bathing, dressing, and eating) and instrumental activities of daily living (IADLs; e.g., preparing meals and managing money). Ideally, measures and studies would focus not just on activities but also on participation in society (as conceptualized in the ICF) and the environmental factors that limit or assist participation. Dr. Freedman observed that no published epidemiologic studies, at this time, have consistently focused on these other dimensions of disability.
The earliest trend analyses for the 1960s and 1970s showed no increase in the levels of disability in old age, but for the 1980s and early 1990s, Manton and colleagues found declines in disability (Manton et al., 1993, 1997). Others, using different data sets, have found conflicting results (Crimmins et al., 1997). A review and evaluation of eight unique surveys that allow trend analysis reported that studies rated fair or good in their methods showed that IADL disability had declined substantially among older Americans (Freedman et al., 2002). Declines have been concentrated in activities central to living independently, for example, shopping, managing money and doing laundry. A technical working group found that there
were also declines in the number of older Americans who had difficulty with ADLs and who received help with ADLs, but these changes were relatively small. The group found that inconsistencies across previously published survey analyses of ADL trends could be attributed to differences in the wording of the questions, the period of analysis used, age standardization, and the inclusion of the institutional population.
Less evidence exists regarding demographic and socioeconomic disparities in trends in late-life disability, and at times the evidence has conflicted. One recent analysis of data from the National Health Interview Survey (NHIS) from 1982 to 2002 shows that racial and ethnic gaps in the need for help with ADLs and IADLs have persisted, and disparities related to education and income gaps appear to be growing (Schoeni et al., 2005). For example, ADL disability has been declining more rapidly for those with higher levels of education and income.
Also limited is the research on the causes of the trends in late-life disability. Given the strong negative relation between disability and education in the cross-sectional analyses, the dramatic increase in the educational attainment of the older population can explain some, but by no means all, of the change in rates of disability. One can hypothesize many ways in which education may influence functioning and disability, for example, by influencing the work-related risk of disease, injury, or impairment; the use of the healthcare system; the use of assistive technologies; adherence to medical regimens; and risk-taking behaviors.
Disability declines do not appear to be the result of declines in chronic disease. To the contrary, based on self-reports, the prevalence of many chronic diseases has increased, even as their disabling effects have declined. This pattern could reflect improved medical diagnosis and treatments, but studies thus far have yielded little insight. There has been a decline in physical functional limitations (such as the ability to stoop or climb, which are precursors to disability in the simplified linear models of the disablement process), but evidence about trends in cognitive and sensory functioning is mixed. Finally, recent data suggest that the increased use of assistive technology (e.g., for mobility and bathing) has contributed to the decline in people needing help with ADLS. But the role of other modern conveniences not specifically designed for people with disabilities (e.g., direct deposit banking and microwaves) is unclear.
Dr. Jay Bhattacharya discussed trends in disability among the U.S. working-age population. (The paper by Dr. Bhattacharya, coauthored by Kavita Choudhry and Darius Lakdawalla, appears in Appendix F.) Will the decreases in rates of disability in old age described by Vicki Freedman in the preceding presentation continue? The answer will depend on trends among the current working-age group, who are the elderly of the future. Dr. Bhattacharya discussed reasons for concern related to the recent worrying
increases in the rates of obesity, diabetes, and asthma among the working-age population. He also noted that a continuation of recent declines in disability in late life would have positive implications for Medicare’s solvency.
Dr. Bhattacharya observed that a challenge in measuring disability among those of working age is that reports of work disability (e.g., difficulty working because of a health problem) may be influenced by public policies, for example, changes in the generosity of disability insurance programs and in the application of the Americans with Disabilities Act. ADL-and IADL-based measures avoid this problem (although they have conceptual and other limitations as measures of disability).
The analysis that Dr. Bhattacharya presented used NHIS reports of the need for help with ADLs for 1984 to 1996 and 1997 to 2000. The division of the analysis into two parts reflects changes in the NHIS questions on disability after 1996. For the period from 1984 to 1996, disability rates for the working-age groups increased, but the rates were basically flat in the period from 1997 to 2000.
The next step in Dr. Bhattacharya’s analysis was to decompose or separate the trend in disability (i.e., needing assistance with ADLs) from 1984 to 1996 into three components: the change in the prevalence of chronic disease, the change in the probability of being disabled given that a chronic disease is reported, and the change in the probability of being disabled given no chronic disease. To illustrate the results, Dr. Bhattacharya presented data for three ages: 30, 45, and 60 years. (These data were developed using a “data smoothing” technique [described in Appendix F] that adjusts data on people not in these specific age groups, e.g., those one year older or younger, to provide more robust estimates.)
For all three age groups, there was a decline in disability among those without chronic diseases. For the 30- and 45-year-old age groups, the data showed an increase in the prevalence of chronic diseases and of disability due to chronic diseases. For these groups, the largest source of disability due to chronic disease was obesity. Those in the 60-year-old age group showed reductions in disability because a substantial decrease in the prevalence of disability among those with chronic illnesses offset a smaller increase in the prevalence of such illnesses. Although 60-year-olds with chronic disease showed decreased disability overall, the increase in obesity was a major countervailing influence in this group.
Beyond the increased prevalence of obesity in the two younger age groups, other highlights were as follows:
for 30-year-olds, increased disability related to heart disease and the combination of heart disease and obesity;
for 45-year-olds, increased disability among those with hyperten-
sion, chronic obstructive pulmonary disease (COPD), and obesity combined with diabetes; and
for 60-year-olds, an increased prevalence of asthma combined with COPD, heart disease combined with stroke, and obesity combined with diabetes (despite a decrease in obesity alone) and decreases in disability related to many chronic conditions but increased disability because of COPD and stroke combined with hypertension.
In concluding, Dr. Bhattacharya stressed the increase in obesity among the working-age population and the increase in disability rates among the younger segments of this population. Noting that he was speculating, Dr. Bhattacharya concluded by observing that the data suggest that the United States may not see a continuing decline in disability among those over age 65 as the current working-age population enters old age.
In the final presentation of the morning, Dr. Ruth Stein discussed trends in disability in early life. She began by reviewing demographic trends, including the growth and increasing racial and ethnic diversity of the child population in the United States. She also highlighted the strong association between poverty and poor health among children. (Dr. Stein’s paper appears in Appendix G.)
Dr. Stein argued that the two traditional measures of disability used for adults, namely, work disability and ADLs-IADLs, do not apply well to developing children, and to young children, in particular, for whom independence in activities is not expected. Instead of these measures, limitations in other activities, such as school and play, have been assessed, although very few children are not able to play.
Dr. Stein presented data on activity limitations among children from a variety of sources that have used different questions about disability or activity limitations. As reported by Newacheck and colleagues (1986), the prevalence of activity limitations among children under 18 years old rose from less than 2.0 percent in 1960 to about 3.5 percent in 1980. More recent data from NHIS show increases to greater than 6.0 percent in 2000 (Child Trends Data Bank, 2005). Most of this activity limitation is related to participation in special education, which is more common among boys than girls.
Supplemental Security Income enrollment for children, which primarily involves those in poverty, has increased. This increase reflects the addition of the mental health conditions list to the program’s eligibility categories and a 1990 Supreme Court decision (Sullivan v. Zebley) that broadened the eligibility criteria.
For childhood disability, major measurement challenges include the lack of an appropriate baseline measure of normal function against which functional deficits can be assessed. Across different parts of society, wide
variations exist in what is viewed as “normal” childhood physical and mental development. Another complexity is presented by the increasing recognition that the core tasks of childhood are development and maturation rather than playing or going to school per se. Dr. Stein cited a recent IOM report (joint with the National Research Council) on children’s health that stressed this point (NRC/IOM, 2004). She argued that it is important to investigate chronic conditions in early life that lead to disability in both childhood and adult life. It is likewise important to understand better the consequences of those conditions, such as a dependence on compensatory mechanisms or the use of services above the usual levels for the age group.
Dr. Stein reviewed the tools that have been developed to operationalize concepts of child health and disability. Based on one of those tools, the 1994–1995 NHIS Disability Supplement, 14.8 to 18.0 percent of children were identified as having special health care needs and about 50 percent of that group had functional limitations. The proportions of children affected increased with increases in age and poverty.
Dr. Stein also noted indirect evidence of overall trends determined from measures of particular child health problems. On the positive side, the rate of lead poisoning has declined substantially during the last 25 years, and the use of folic acid antenatally has been associated with declines in the numbers of children born with spina bifida and anencephaly since 1996–1997. On the negative side are increases in asthma rates and rates of the low birth weight and preterm births since the 1980s. Racial and ethnic disparities in infant mortality rates continue. Also troubling is the increase in the proportion of children ages 6 to 19 years who are obese from less than 5 percent in the mid-1960s to almost 15 percent in 1999–2000. Minority children are especially affected.
Dr. Stein closed by suggesting that if the goal is to minimize lifetime disability, then there is need for a conceptualization of disability for the young that is broader than the most severe degree of limitation. She pointed out that the genomic revolution is allowing the much earlier identification of children who are at biologic risk of later disability. It is, thus, providing an opportunity for early intervention to limit the progression or consequences of genetic conditions before they cause observable problems. If we enlarge the focus of disability policies beyond children and adults who are already severely disabled, the growing number of effective preventive strategies can be better applied.
Several questions during the discussion focused on data and methodology issues. One question was whether there were data sets that would allow investigation of disability trajectories for those with child-onset conditions
versus the trajectories for those with adult-onset conditions. Dr. Stein responded that such data were not available for the United States but might be for the United Kingdom. Another question was whether it would be possible to use current data sets to operationalize the two-way arrows that specify feedback from activity and participation to disease in the ICF graphic model. Dr. Bhattacharya indicated that it is not possible to capture such interactions in cross-sectional data sets and that useful longitudinal data are scarce.
One suggestion was that to predict future trends analysts need to understand the differences in environments and the rates of environmental change that different birth cohorts have experienced. Dr. Freedman responded that disentangling the large age, time period, and cohort effects is mathematically challenging. She argued that it would be helpful to focus somewhat less on describing what has happened in the past and somewhat more on using data to help guide society in acting to prevent disability in the future.
Responding to a question about the influence of the creation of Medicare on old-age disability trends, Dr. Freedman said that it was difficult to measure the influence of this key policy change. In general, studies that have focused on the influence of medical treatment on disability have suffered from serious data limitations. In addition, she suggested it would be useful to see more investigations about the impact of public policies in addition to Medicare.
When he was asked about any prevention messages that would follow from the morning’s presentations, Dr. Bhattacharya emphasized that successful steps to reduce the rates of obesity would have a very large effect on future disability rates and that reducing the prevalence of chronic diseases would also diminish disability. Dr. Stein agreed strongly. She added that the lack of universal health care for children in the United States particularly affects the uninsured working poor whose children have the worst health status.
Dr. Stein also mentioned a body weight and cohort concern related to the increasing survival of extremely low birth weight infants. Recent data suggest that almost half of such infants are experiencing health problems or functional limitations eight to ten years later.
Another participant pointed to the recent controversy about the relation between weight and mortality and wondered if there were statistical pitfalls that should be taken into account when emphasizing obesity reduction as means of preventing future disability. Dr. Bhattacharya noted that part of the difficulty in the relationship between weight and mortality is that overweight (rather than obese) people have mortality rates similar to those of normal-weight people. So far, his research has focused on obesity and not on overweight. For those who are obese, the link to disability is very persuasive.
One participant expressed concern that focusing more on children with less severe disabilities would leave behind those with the greatest disability because there are so many more children in the first group. Dr. Stein responded that there could indeed be a slippery slope, but from the perspective of the nation’s future, it is important to develop a more comprehensive understanding of disability in the population and its sources. We must continue to be concerned about children and others with the most severe problems, but we will miss opportunities if we focus too narrowly on this group.
DISABILITY ACROSS THE AGE SPECTRUM
Dr. Jack Guralnik began the afternoon panels with a review of research on risk factors for disability in old age. (Dr. Guralnik’s paper appears in Appendix H.) He credited the Nagi model of disability and its further articulation in the IOM reports for helping guide epidemiologic research and hoped that the new IOM study would produce equally valuable direction. Dr. Guralnik also cited the value of the more recent ICF model and the work of Verbrugge and Jette (1994). He encouraged the current IOM committee to continue with an approach or model that is practical and that can be operationalized in a valid and reliable way for population-based and other studies.
Dr. Guralnik explained that many of the data that he reviewed came from work performed at the National Institute on Aging that employed a set of population-based studies called the Established Populations for Epidemiologic Studies of the Elderly (Guralnik et al., 1993). By monitoring populations over time, it was possible to assess the developing incidence of mobility limitations and other problems and to identify chronic health conditions as risk factors for these problems. For purposes of their work, mobility problems were defined as the inability to walk a quarter mile and the inability to climb a set of stairs. By way of overview, the research group found that the odds ratio for the loss of mobility was in the range of about 1.2 to 1.5 for people who had baseline reports of heart attack, stroke, diabetes, dyspnea, or exertional leg pain. Other analyses have focused behavioral risk factors such as smoking, drinking, and physical inactivity.
Dr. Guralnik noted differences between chronic conditions as risk factors for women and men. He presented data showing that women reported arthritis as the main cause of problems for walking one-half mile, doing heavy housework, and bathing. The pattern for men differed. Men also reported arthritis as the top cause of difficulty walking one-half mile. However, for disability in doing heavy housework, the top cause was heart disease, and for bathing disability, stroke led by a small margin.
It is important to examine both the difficulty in performing an activity and the inability to perform the activity at all. For example, Leveille and colleagues (1999) have found that although pain has a significant impact on difficulty with climbing, lifting, and ADLs, it is not significantly related to whether people can or cannot perform these activities at all.
In identifying health risk factors for disability, it is also important to consider comorbid conditions. As the number of comorbid conditions increases, the risk of developing a new disability increases rather dramatically. Individuals with four or more conditions at the baseline were almost three times as likely as those with no conditions to have lost mobility at follow up. He noted that this kind of epidemiologic study is very difficult to do, even with datasets as large as the one he and his colleagues were using.
Dr. Guralnik discussed research that attempts to explain the mechanism underlying the association between diabetes and problems with lower-extremity function. When he and his colleagues examined specific diabetes-related conditions and impairments (e.g., peripheral neuropathy and visual impairment), they found that entering each condition into the group’s statistical model reduced the initial statistical association between diabetes and the measures of lower-extremity function. Collectively, the diabetes-related conditions explained about 80 percent of the statistical association between diabetes and (especially) mobility outcomes (Volpato et al., 2002).
Although most of his work has focused on health conditions as risk factors for disability, Dr. Guralnik reported that his group has examined some socioeconomic and demographic factors. He presented data for white and black women that showed more years of disability-free life expectancy for those with higher levels of education (Guralnik et al., 1993).
In closing, Dr. Guralnik reviewed data on the level of disability among people over age 65 in the years before death. For people in their 90s, disability rates are very high. Thus, although age-adjusted or age-specific rates of disability are declining in the United States, the overall number of people affected and the impact on society will grow because so many more people will be entering the very old age groups. Identification of the causes of late-life disability and interventions that can mitigate these causes or their effects will be increasingly important.
The presentation by Dr. John Reiss shifted the focus from the oldest segment of the U.S. population to the youngest and from risk factors for disability to certain difficulties that children with disabilities or special needs may experience as they move from adolescence into adulthood and from pediatric to adult health care providers. (Dr. Reiss’s paper, which is coauthored by Robert Gibson, is presented in Appendix I.) He distinguished between the narrow concept of an individual’s transfer from pediatric to adult health care services (an event) and the broad concept of transition, which refers to a planned process that should support the development of
the knowledge and the skills needed to support a young person’s full participation and decision making in the adult world.
Any discussion of disability among children needs to recognize the centrality of children’s physical and mental development, a process that begins with the normal dependency of infancy and that is, for most children, marked by increasing independence and by an emphasis on actions and policies to “enable” children’s greatest development. Dr. Reiss discussed a conceptual framework—represented graphically as a sort of kalei-doscope—that was proposed in a 2004 report, Children’s Health, the Nation’s Wealth: Assessing and Improving Child Health, from the National Research Council and the Institute of Medicine (NRC/IOM, 2004). That framework depicts the complex and dynamic nature of child health and the social and policy factors that affect a child’s physical and social environment and that interact with a child’s biological and other characteristics to affect the child’s health.
Dr. Reiss noted the support by the Maternal and Child Health Bureau for studies of health care transitions for children with special health care needs and for work to identify the policies and services that can help these children live more independently as adults. He also cited the consensus statement by the American Academy of Pediatrics, the American Academy of Family Physicians, and the American College of Physicians (2002) endorsing the provision of uninterrupted, high-quality, developmentally appropriate health care services to maximize individual functioning and support the transition of young people from pediatric to adult health care.
The reality appears to fall short of this goal. Results from the 2000 National Survey of Children with Special Health Care Needs found that only about half of the parents with children between the ages of 14 and 17 years said that their child’s pediatrician or family physician discussed how the child’s needs might change as he or she moved toward adulthood (Lotstein et al., 2005). Only 30 percent reported that they had a plan for addressing those changing needs, and just 20 percent had a plan for transferring their child to adult health care providers. These findings, combined with other information and experience, point to significant problems, including unplanned, abrupt transfers from pediatric to adult health care services. Data on what happens after unplanned transfers are, however, limited.
Dr. Reiss observed that transfers from pediatric to adult health care are typically based on age rather than the individual readiness of a young person. Policies for Medicaid and other health insurance and social service programs play a role, since many supportive services for children end at age 18.
Dr. Reiss suggested that internists and other physicians for adults tend to be less comfortable with a significant role for parents. Even young people without special needs often are still developing the decision-making and
other capacities and knowledge needed to navigate the adult health care system. An associated need for some degree of parental involvement and support may continue for many years after an individual reaches age 18.
Dr. Reiss proposed that more research is needed on what happens to individuals, especially those with special needs, after transfer to the adult health care system. Are there consequences for the utilization of resources, health outcomes, the development of secondary conditions, and the quality of life? Research should also focus on understanding what is expected of young adults in the adult health care system. What are the key competencies that youth need to move into the adult health care system, and what adjustments in that system may be necessary to provide developmentally appropriate services for young people with special needs?
Dr. Reiss concluded by suggesting that pediatric health care has changed over time to become much more child and family centered and that this change stems in part from successful advocacy by families and advocacy organizations. He argued that we need youth leaders to advocate for an adult health care system that is more responsive to youth and young adults with special health care needs.
The first question in the discussion focused on what the IOM committee should consider in terms of analysis and recommendations about making the transition to adult health care services work for children with disabilities or special health care needs. Dr. Reiss repeated his view that young adults are still developing their capacities, including their ability to assert themselves in encounters with health care professionals. Physicians and others need to recognize this. Particularly when they see young people with special health care needs, physicians need to become more accepting of a role for parents. He also suggested that the board requirements for internal medicine be revised to cover the transition needs of young adults.
Another participant questioned such broad requirements, arguing that few internists see enough young adults with conditions such as cerebral palsy to become experts in their management. Specialized requirements will go for naught if the actual practice volume is not high enough to reinforce and build on the internist’s initial training.
One of the next questions dealt with priorities for research to advance understanding of the epidemiology of disability in late life. Dr. Guralnik responded that researchers have investigated pieces of the pathway to disability for particular health conditions or limitations but that the overall picture is still very partial. The epidemiologic monitoring of people over the life course and the development of better and less burdensome instruments for assessing disability would help.
In response to another question, Dr. Guralnik added that it is particularly important to illuminate aspects of the pathway to disability that are amenable to interventions—for example, certain kinds of exercise—that will prevent or reduce disability. Suggestive observational data are available, but much still remains to be learned. For example, some data suggest that if a person has good balance, that person needs less strength to maintain mobility, so that might point to more of a focus on balance but also on the identification of the basic strength levels needed to maintain functional independence. The collection of some important data will require controlled trials of exercise and other interventions.
Dr. Reiss added the observation that young people who have had functional limitations from a very young age, for example, children who had needed a wheelchair for years, may find disability a foreign concept because they have developed special skills or have found assistive technologies or other means to do what they need to do. What they want and how they want to interact with health care providers and others may differ from the desires of someone who develops limitations as a result of an injury or a medical condition in later life. The perceptions of these young people may also differ from those of health care providers accustomed to working with older adults. Dr. Reiss also endorsed the focus on “enabling America” (as in the 1997 IOM report) as particularly appropriate for young people with special needs.
Dr. Guralnik observed that it is a challenge for those who do research on disability in old age to talk to people interested in childhood disability. Discussions can have an almost “apples and oranges” aspect. Among those doing research on aging, disability is viewed as a kind of final common pathway of many chronic diseases, and measures of disability are useful in summarizing health status. Young people, however, tend to be uncomfortable with and even insulted by data on how disability, for example, predicts institutionalization. Working and learning together are important but not easy.
SECONDARY HEALTH CONDITIONS: PART I
Dr. Margaret Turk led off the panel with an overview of secondary conditions. (Dr. Turk’s paper appears as Appendix J.) She proposed definitions of several concepts, suggesting that it would be helpful for researchers, analysts, and clinicians to be consistent and use the same terms or labels to refer to related but distinct health circumstances. Use of the same words to mean different things can impede communication.
Interest in secondary conditions seems to have been stimulated in the 1980s, when Dr. Michael Marge began to use the label “secondary condition” in his work with the National Council on Disability (Marge, 1988).
Use of the term in two IOM reports in 1991 and 1997 further focused attention on secondary conditions as subjects for study and intervention. Secondary conditions are important because they can create or intensify activity limitations, make participation in society more difficult, and decrease the quality of life.
Dr. Turk proposed several definitions and clarifications to distinguish secondary conditions from other conditions. A secondary condition is directly related to a primary disabling condition; the primary condition is a risk factor for the secondary condition. An example for people with cerebral palsy is contractures. Depending on the primary condition, other examples include pain, renal insufficiency, pressure ulcers, osteoporosis, and chronic edema in the lower limbs.
Dr. Turk argued for limiting the term “secondary condition” to health states (physical or mental) and not extending it to cover the social correlates or the potential societal consequences of having a primary disabling condition. Examples of such correlates include reduced access to employment or companionship. Such outcomes are important to our understanding of disability and quality of life, but lumping them together with secondary health conditions is not helpful. Doing so dilutes the concept and understanding of secondary health conditions. More important, perhaps, such a broad definition of secondary conditions blurs the focus on key elements in current models of disability, specifically, restrictions on participation in society and performance of social roles and the environmental and personal contributors to such restrictions.
The relationship between particular primary conditions and particular secondary conditions—including the likelihood and ways that an individual with a primary condition will experience a secondary condition—may be quite variable. Several general factors will influence the expression, prevention, or modification of a given secondary condition. These factors include (1) the state of scientific knowledge about the primary condition and its consequences, (2) the existence of proven clinical interventions or assistive technologies, (3) social and physical environments and public policies (including health insurance coverage for services and equipment), and (4) family and personal factors (including an individual’s age or stage of development and personality traits).
Dr. Turk said that it is important to recognize that a primary condition will have a number of features that are aspects of the basic pathology or nature of the condition, what we may call associated conditions. With cerebral palsy, for example, seizures, mental retardation, and spasticity are considered features—albeit variable features—of the primary condition. They are not secondary conditions that result from the primary condition. She recognized that there may sometimes be a fine line between
the basic pathology of a primary disabling condition and its secondary health consequences, and the new research can change perspectives.
A “comorbid condition” is one that is neither part of nor the result of a primary condition. For example, an individual who experiences a spinal cord injury may have skin cancer at the time that the injury occurs or may develop the disease later. Skin cancer is a comorbid condition in either case. Sometimes, increased scientific knowledge leads to an understanding that what was thought to be an unrelated comorbidity is actually a related secondary condition. For example, spinal cord injury is now recognized as a risk factor for insulin-resistant diabetes.
Dr. Turk also considered the relevance of aging to secondary health conditions. Aging is a normal developmental process. It is variable and modifiable to some degree across individuals, but manifestations of aging are an expected part of human life. People with disabling conditions, however, may have a smaller reserve capacity for performance and function as they age. In addition, some conditions associated with aging may be accelerated. For example, depending on a person’s primary condition, she might be at risk for pain from degenerative joint disease as part of normal aging or at risk for the early onset of symptoms.
If one thinks in terms of health and wellness of people with disabling conditions, then maintaining or improving health requires attention to the etiology, prevention, and management or modification of secondary conditions for which people are at risk because of their primary condition. Again, depending on the condition, a secondary condition can have serious consequences for a person’s survival, quality of life, and capacity to take advantage of participation-enhancing equipment or environments. Despite confusion and differences in the use of the term “secondary conditions,” the discussion overall has helped direct attention of physicians and others to issues and problems that are central to the lifelong health and well-being of many people with disabling conditions.
In the next presentation, June Kailes spoke primarily from a consumer perspective, describing herself as a “living aging-with-disability laboratory.” (Ms. Kailes’s paper appears as Appendix K.) As someone in midlife who has lived with cerebral palsy since birth, she observed that disability-specific health care programs treat many people with lifelong disabilities as if they disappear when they turn 21. People with disabilities between the ages of 21 and 65 find a dearth of experienced professionals and comprehensive services. Instead, they encounter access barriers, illogical bureaucracies, and professionals with little preparation to serve midlife adults with disabilities. They encounter a system focused on acute care rather than continuing and coordinated care and assistance for those living long term with disability and chronic conditions. Choices may be limited, if one is
lucky, to a pediatric-focused health care professionals with teddy bears on their business cards or geriatrics-focused health care providers.
Navigating the current health care system requires great energy, good “health literacy,” and strong skills in advocating for oneself. Not everyone can be expected to have these personal navigational resources.
Ms. Kailes noted that aging with disability is a particular concern. As they provide routine health care, medical professionals may not anticipate the early aging experienced by many people with disabling conditions such as cerebral palsy. Margaret Turk talked earlier about secondary conditions related to aging and living with disability. Progress in identifying these conditions has not yet translated into effective interventions.
Ms. Kailes reiterated that disability is not something one has or does not have. Instead, a continuum of activity limitations exists and affects most people to various degrees at some point in their lives, if not lifelong. Furthermore, for policy, service, and research purposes, disability should be defined on a functional basis and in a unified and comprehensive way rather than in fragmented and narrow categories. It is not helpful to have many different definitions of disability across government departments and programs.
With respect to research, a greater focus on functional limitations that are common across a number of disabling conditions makes sense, given the complexity and diversity of these conditions and the low prevalence of many discrete diagnoses. The science also needs to be coordinated across relevant federal agencies to infuse attention to disability content throughout the broad range of health research. Although people with disabilities have a wealth of experience to inform policy, they are just advocates with opinions if they do not have the support of sound research and data.
In addition, Ms. Kailes stressed that planning for services and research should be undertaken with people with disabilities rather than for but without them. People with disabilities bring important knowledge and perspectives to the planning process and also to implementation. They can contribute as managers, investigators, and collaborators. As an additional point, people with disabilities should not be routinely excluded from pharmaceutical and other clinical trials that are not related to their disabling condition.
Ms. Kailes contended that services that are initially denied because of antiquated and shortsighted policies, inaccessible medical facilities and equipment, and similar problems can lead to the subsequent downstream use of more health care and social services and increased costs. One specific problem is restrictive definitions of medical necessity in health insurance programs. Such definitions often exclude services and technologies, including durable medical equipment, that will prevent deterioration, injuries,
and decline in people with disabilities and thus avert the subsequent use of additional acute health care services and social services. Wheeled mobility is not a “convenience” item, as it has been described by one insurer. A motorized scooter is not a fashion statement.
Medicare specifically should change its policy that restricts coverage of mobility devices to those that are reasonable and necessary for use inside a person’s home. Many people with mobility limitations can get around the home by “wall walking” and “furniture surfing.” Outside the home, these methods are unsafe and unacceptable. The current Medicare policy reflects an outdated mind set that places minimal value on participation in society by people with disabling conditions.
Another problem identified by Ms. Kailes is that the implementation of ADA in health care settings lags far behind its implementation in many other areas. The implementation of ADA requires more attention to physical access to facilities and equipment (e.g., scales and examination tables) and also to accessible communications and other program features. Ms. Kailes closed by emphasizing that the work of the current IOM committee on disability is serious business for those who need better services and tools to live productive and fulfilling rather than unnecessarily confined lives.
Dr. James Rimmer began his presentation by endorsing the participatory model of disability research described by June Kailes. (Dr. Rimmer’s paper, coauthored by Swati Shenoy, appears as Appendix L.) With respect to the panel topic, he cited the impact of a conference held several years ago that focused on secondary conditions in individuals with spina bifida and cerebral palsy. The conference and publication of the conference proceedings were sponsored by the Spina Bifida Association and United Cerebral Palsy in conjunction with several federal agencies including the Centers for Disease Control and Prevention (Lollar, 1994).
Dr. Rimmer suggested that continued research on secondary conditions and their prevention and management is important because secondary conditions are a major contributor to health disparities among people with disabilities. In the literature on secondary conditions, exercise-related research is quite prominent. Most of this research is, however, descriptive, with few experimental studies that test exercise interventions to reduce secondary conditions.
Dr. Rimmer noted recent studies reporting that people with a limited exercise capacity were at higher risk of death than others in the population. As is often the case, this research excluded people with disabilities.
The literature reviewed by Dr. Rimmer and his colleagues covered the period from 1990 to 2005. For the literature search, key words included “exercise,” “physical activity,” “secondary conditions,” and “disability,” in addition to terms for specific disabling conditions, such as “multiple
sclerosis” and “spinal cord injury.” The search yielded 167 abstracts, of which 42 were exercise training studies.
Dr. Rimmer’s presentation focused on the literature examining the effects of exercise on fatigue, pain, and deconditioning. The review yielded few controlled trials. Several nonrandomized controlled studies reported that long-term aerobic and resistance exercise increased both physical and psychological well-being. Some additional studies reported that after the end of the exercise intervention, participants reported increased stress and a decreased quality of life without exercise.
A number of design limitations characterize this research. One limitation is small sample sizes, which is not surprising, because the recruitment of large numbers of research participants for studies of conditions with a low prevalence in the population is difficult. In addition, various exclusion criteria eliminated a large fraction of the target population. Dropout rates can be high, often because of transportation problems or the emergence of some kind of health complication. The implications of dropout rates for study validity must be assessed. Another limitation is that most interventions lasted 12 weeks or less and did not provide long-term follow-up. A different problem is that studies often include such a large number of outcome measures that some outcomes are likely to be found to be present at a statistically significant rate just by chance.
Among the important directions for research, one priority should be more efforts to conduct randomized controlled trials and longitudinal studies on exercise interventions for people with disabling conditions. Such studies will generate a stronger evidence base for guidelines and recommendations on exercise interventions for people with different disabling conditions.
Notwithstanding the value of randomized controlled trials, Dr. Rimmer observed that prospective and cross-sectional observational studies have a role to play. A larger study population can often be included in such studies, thereby providing more opportunity to control for demographic and other variables. Observational studies may allow assessments of exercise dose-response relationships that are expensive to assess in clinical trials.
New assessment technologies have real potential to reduce some of the burdens of research participation and make enrollment and study continuation easier. Examples include the use of web-based reporting by research participants, the use of more time-efficient assessment tools, and the use of global positioning systems to track certain aspects of physical activity.
Dr. Rimmer noted the existence of a large number of outcome assessment instruments that complicate comparisons across different studies. He suggested that a better understanding of the strengths and limitations of different assessment instruments would help researchers make better deci-
sions about which instruments they should use and would also allow them to weed out some of the instruments with low reliability and validity.
In addition to research on interventions and measurement tools, more basic work is also needed to identify the biological mechanisms associated with changes in conditions such as stress and fatigue. Such work can guide future interventions. Likewise, a better understanding of the effects of different elements of health promotion efforts that involve multiple components (e.g., exercise and peer support) would be helpful.
One participant observed that the discussion of secondary conditions needed to distinguish different disabling situations. One situation is a catastrophic disability in an otherwise young and healthy person; that individual has a primary condition—a medical diagnosis—that is a risk factor for specific secondary conditions or diagnoses. In contrast is the situation in much geriatric care, in which the problem is general frailty rather than a clear, primary disabling condition and in which frailty-associated activity limitations are risk factors for further deterioration. There needs to be a common framework for these contrasting situations that map onto the same longitudinal, enabling-disabling model of disability.
Dr. Turk agreed that the secondary conditions concept had been primarily applied to people who had lifelong primary disabling conditions such as cerebral palsy or people who experienced disabling injuries or illnesses in adolescence or midlife. How the concept should be applied to frail elderly people and also to people with chronic conditions such as obesity or hypertension needs more attention.
Another commenter returned to the issue raised by Dr. Reiss in the previous panel, specifically, the lack of professional education and research that focuses on people with disabilities who are too old for pediatric care and not old enough for geriatric care. The relative lack of service models approximating the pediatric and geriatric models is why families and adolescents feel that they drop into an abyss when a young person reaches age 18 or 20 or so. What programs exist are mostly diagnosis specific (e.g., for people with AIDS) and even then are generally not available to all people with the diagnosis. Dr. Turk pointed out that some model programs help adolescents with the transition from pediatric to adult services, providers, and programs. Widely available federal support may be needed for such programs to become more widely established.
Another issue for adolescents with disabilities is the high level of secondary mental health morbidity. This kind of secondary morbidity is not only a problem in itself but is also a behavioral risk factor for further problems
(including those resulting from risky or self-destructive behaviors). This point prompted mention of the Surgeon General’s new “call to action” on the health and wellness of persons with disabilities (Surgeon General of the United States, 2005). The document covers a broad range of topics, including some discussed in the workshop papers and presentations and other topics that need further consideration, for example, mental health problems.
In a comment about issues related to conducting research on exercise interventions to prevent or reduce disability, one participant cited Dr. Rimmer’s observation about high dropout rates in exercise studies. That participant observed that people in certain target populations see themselves as having a fixed budget of energy that requires them to make trade-offs each day, for example, between using energy to participate in an exercise study versus using energy to do laundry. In this context, might not effectiveness research be just as important as efficacy research? That is, researchers should look at people’s daily lives to determine whether something works in the real world, not just in a controlled trial situation. This participant asked whether researchers considered and investigated how people actually implement exercise in their daily lives and how they make trade-offs in spending their limited store of energy?
Dr. Rimmer responded that researchers bring people into laboratories. Although they may compare the characteristics of people who drop out versus the characteristics of those who continue with the research, they usually do not learn much about the actual reasons that people drop out of studies. Also, the studies he reviewed generally did not do “intention-to-treat” analyses and this limits the validity of the results because those who complete the exercise intervention may differ from those who drop out in ways that confound comparisons. How researchers can adapt their studies to accommodate people’s real-life problems with transportation, energy limitations, and other barriers to research participation and the long-term effectiveness of an intervention is an important issue.
Another question arose about the contribution of pain to disability (defined as a limitation in performing socially expected roles, such as work). Dr. Turk noted that data sets vary in the amount of information available for the assessment of how pain may limit activities or participation. She agreed that it is important to consider the role of pain, whether it is an inadequately treated aspect of a primary condition or an aspect of a related secondary condition.
A participant then returned the discussion to the issue of exercise as an effective intervention in people’s actual lives. It is important to understand dose-response relationships and to identify the critical thresholds (e.g., 20 versus 30 minutes of exercise) to achieve positive results for different kinds of interventions. Studies of this kind will help people with disabilities judge
how they should invest their limited energy in exercise. The identification of interventions for which underlying biological mechanisms of effect have been identified should also help people make decisions about intervention options.
More generally, bringing together what is known in the basic sciences with what is known in applied or clinical sciences and epidemiological sciences is important, and real opportunity exists in this area. One example involves research on the biology of muscle contractures among people with limited mobility, a secondary condition that is a major focus of clinical prevention and management efforts. If scientists could identify the molecular switch that causes fibrous ingrowth into muscle, then ways to control that switch and prevent contractures from forming in the first place might be found.
A further research area with considerable promise involves the interaction between exercise-patterned activity and pharmacological options. For example, investigators studying rats have found that amphetamine does nothing by itself to rehabilitate a rat and that exercise has some effect, but amphetamine in combination with exercise potentiates the effect of exercise dramatically. Although it is difficult to do these types of studies with humans, the potential of such pharmacological strategies to improve the outcomes of exercise interventions is significant.
One participant noted the importance of both social and personal responsibility for improving the health and well-being of people with disabilities. There is a social responsibility—which has been greatly inadequate in practice—to provide individuals with access to services and technology that allow them to function better and participate in the community. In addition, people with and without disabilities have a personal responsibility to lead the healthiest lives possible, for example, by looking after their diet, exercise, and similar elements important to health. Persuading people to adopt health-promoting practices and supporting them in that effort are, unfortunately, difficult tasks, whether the focus is on people with disabilities or not.
Another concern with health promotion is access by children with disabilities to school physical education activities. These activities provide opportunities for learning good health behaviors and improving physical fitness that can have both short-term and long-term benefits. Education and health personnel need to coordinate their attention to the benefits of school physical education, the barriers encountered by children with disabilities, and the options for overcoming these barriers.
The discussion for this panel ended with an expression of concern about an impending crisis with Medicare, Medicaid, and private insurance programs. The result could be potentially dangerous reductions in financial access to services and equipment for people with disabilities. Monitoring the allocation of health care dollars will be important.
SECONDARY HEALTH CONDITIONS: PART II
Dr. William Bauman discussed secondary conditions in the context of spinal cord injury. (Dr. Bauman’s paper appears as Appendix M.) His presentation covered pulmonary, cardiac, metabolic, gastrointestinal, and dermatologic secondary conditions. Dr. Bauman recalled that when he made his first presentation on this topic at a conference in 1990, the audience did not at that time appreciate the importance of understanding secondary conditions as problems. Now the topic is firmly established in the field of spinal cord medicine.
As in the broader population, heart disease and lung disease are common causes of mortality and morbidity for people with spinal cord injuries. Individuals with spinal cord injuries have restrictive ventilatory dysfunction, and those with higher cord lesions also show evidence of airflow obstruction. The higher the spinal cord lesion, the greater the effect on respiratory muscles and the more difficulty people have breathing and coughing effectively. The tubes in the lung can become blocked, causing infection. Research has helped identify the level of impaired expiratory function that begins to cause problems. Whatever the technique—mechanical, pharmacological, electrical, or other—used to help people breathe better, it is important to mitigate this important contributor to mortality and morbidity.
Initially, some of the secondary complications for people with chronic spinal cord injury were not recognized because patients often did not survive long term. With improvements in long-term survival has come recognition of excess heart disease deaths in this group.
One contributor to these excess deaths is the predisposition of individuals with spinal cord injuries to the metabolic syndrome. People tend to develop insulin resistance, become obese, and develop carbohydrate and lipid abnormalities—in particular, low levels of high-density lipoprotein (HDL) or “good” cholesterol. Those with lower cord lesions also tend to have elevations in blood pressure. Data from the Framingham (Massachusetts) Heart Study demonstrate the health risks of low levels of HDL. For every 1-milligram-per-deciliter rise in the HDL cholesterol level, the risk of a cardiac event decreases by about 2 or, possibly, 3 percent (Castelli et al., 1986). For those with spinal cord injuries, the role of low-density lipoprotein (LDL), or “bad” cholesterol, is more complex, but evidence suggests that these individuals have smaller, denser, more atherogenic LDL particles, which is a risk factor beyond the actual LDL level itself.
Another risk factor related to the metabolic syndrome involves changes in body composition and insulin resistance. In people with spinal cord injuries, such changes occur both immediately and chronically. In general,
the less active that a person is and the greater their level of body fat, the greater their extent of insulin resistance is. The mechanisms for these changes appear complex, possibly including (in addition to paralysis and immobilization) a reduction in anabolic forces and an elevation of catabolic hormones. In addition, those with the greatest neurologic impairment have the lowest frequency of normal glucose tolerance.
Management options for prevention of the development of diabetes and heart problems include increased activity, dietary changes, smoking cessation, and drug therapies. Some evidence suggests that even a small increase in the amount of exercise performed may be beneficial. Given the difficulty, cited earlier, in engaging people in exercise, this is encouraging.
Dr. Bauman went on to describe other secondary problems associated with spinal cord injury, one of which is difficulty with evacuation related to their injury. Bowel care is time-consuming, and shortfalls in such care increase morbidity. Cholinergic agents are effective but have adverse effects on the heart and on the lung. One promising area is combination drug therapy that selectively blocks these adverse effects.
Osteoporosis is another serious secondary problem. Losses in bone density occur very rapidly after injury and continue for years. It is particularly crucial to prevent bone loss immediately after the injury to avoid a compromised bone architecture that cannot be restored. Preventive strategies include drug and mechanical interventions. Vitamin D deficiency is also important in individuals with spinal cord injuries. More remains to be learned about vitamin supplementation strategies to prevent or limit osteoporosis in this group. Overall, more effective therapies to prevent or treat bone loss would improve employment prospects and other kinds of social participation.
Pressure ulcers are a tremendous source of morbidity in those with spinal cord injury and in the general population—probably costing about $8 billion to $10 billion a year nationwide for all care provided for this condition. Again, much remains to be learned about the development of pressure ulcers; rates of ulcer healing; and the roles of nutrition, inflammatory, and endocrine factors. Dr. Bauman noted that the Veterans Administration had funded a cooperative study to compare treatment with an oral anabolic agent against a placebo (each in combination with optimal clinical care) and assess the effects on the percentage of complete healing for pelvic region pressure ulcers. We will also learn more about the effects on healing rates of nutrition, inflammatory factors, and endocrine factors.
Dr. Bauman observed that most individuals with spinal cord injuries have learned to adapt to being paralyzed. In many respects, the secondary complications of the injury may have a greater impact on a person’s quality of life than the loss of ambulation itself. The knowledge gained from study-
ing the secondary consequences of spinal cord injuries can probably be transferred in large measure to other immobilizing conditions. Prevention and early intervention are important both to people’s quality of life and to the costs of health care.
For the next presentation, Dr. Bryan Kemp spoke about depression in adults with disabilities. (Dr. Kemp’s paper appears as Appendix N.) Whether depression should be categorized as a secondary condition or something else, it has important consequences for health and well-being. It is also a common and expensive problem for the population generally and affects every aspect of life: physical health, behavior, functioning, participation, interpersonal relations, and more.
Data indicate that depression is more common in people with disabilities and, in some respects, is more serious in this group. Compared with the rates of depression among people with no disability, the rates of depression among those with post-polio syndrome, rheumatoid arthritis, cerebral palsy, spinal cord injury, or stroke have been reported to be higher. Perhaps 1 person in 10 or 12 overall has a depressive disorder that warrants treatment. Among people with disabilities, the number is more like 1 person in 3 or 4.
Depression affects health and well-being both directly and indirectly because it reduces important brain transmitters as well as compliance with exercise and other health promotion activities and diminishes participation and social connections. One study of people with stroke reported that the 10-year survival rate was about 65 percent for those without depression but only about 30 percent for people with depression (Morris et al. (1993). Studies of change in functioning over time for people with disabilities report that depressed individuals show greater rates of decline than individuals without depression.
Dr. Kemp explained that the effects of aging present additional concerns for those with certain kinds of disabilities and increase the risk of depression. About 25 years after the onset of a disability (in a younger person) or at about 45 years of age (seemingly, whichever comes first), people begin to develop an inordinate number of new medical and function-related problems as well as pain, fatigue, and weakness. These changes are associated with higher rates of depression.
The literature supports some general statements about the causes of depression in people with disabilities. It does not appear to be directly related to the severity of the impairment; for example, more severe spinal cord injuries are not associated with higher rates of depression. Depression also does not appear to be related to the duration of impairment.
Dr. Kemp suggested that depression is most likely related to a person’s ability to cope with the changes and losses that are associated with a
disabling condition rather than to the existence of the condition per se. Studies also suggest that depression in people with disabilities is related to their financial status, level of social support, and other resources. For example, a study by Dickens and Creed (2001) reported a substantial increase in depression among people with disabilities who reported the loss of valued social or family activities.
Several factors seem to account for the inadequate treatment of depression in people in the community who have a disability. Lack of access to care is an important factor. In addition, symptoms of depression (such as fatigue) often overlap with symptoms of other health problems or disability effects. Some symptoms, such as apathy or irritability, may be missed by health care professionals who focus on feelings of sadness. Further, if professionals believe that depression is normal for people with disabilities, they will tend to undertreat it.
Dr. Kemp noted that long-term studies of treatment for depression among individuals living in the community are limited and that relatively few controlled treatment studies have been conducted. He reported that his group’s quasiexperimental studies of combined pharmacotherapy and psychotherapy suggest that treatment for depression can be effective in several respects (Kemp et al., 2004). The group found that treatment reduced the level of depression and increased community participation; subsequently, people reported increased life satisfaction.
Many important areas for research remain. Some involve the better measurement of depression and outcomes in people with disabilities. Others relate to the short-term and long-term effectiveness of different treatment strategies, for example, comparison of early and late interventions and of different durations of therapies.
Dr. Tom Seekins made the final presentation, which covered secondary conditions among people with intellectual or developmental disabilities. This group numbers about 2 million to 4 million people in the United States. Estimates indicate that more than 350,000 adults with such disabilities live in more than 110,000 supported arrangements around the country. (The paper of Dr. Seekins and his colleagues, which appears as Appendix O, includes an updated estimate that over 420,000 adults with developmental disabilities live in over 148,000 such arrangements nationwide.)
Dr. Seekins described a community-based intervention strategy—a wellness club—that was designed for the target population and community. The strategy employed a surveillance model and instrument to support the planning and assessment of wellness services within the larger context of community-based services for people with developmental disabilities. First steps were the initiation of annual surveillance of health status at the community level and the evaluation of surveillance results at the state level, which provided the basis for setting priorities and mobilizing resources for
changes at the local level. The surveillance instrument asked people to rate 45 secondary conditions. Using the responses, the investigators created a problem index based on the reported prevalence and the average reported severity. Data were also collected on 36 behavioral risk and protective factors and medical service utilization.
In the survey, the top three problems measured by the problem index were communication difficulties, physical conditioning problems, and weight problems. For those people reporting a problem, however, mobility problems had the highest average severity level. Among the risk factors for problems were stressful life changes, living arrangements, and turnover of personal assistance. Behavioral risk factors included inadequate physical activity and poor dietary habits.
Dr. Seekins also discussed the design of individual service plans for adults with developmental or intellectual disabilities who are living in supported arrangements in the community. Individual service plans, which are required by law, guide the provision of services for consumers and provide the blueprint for organizing the treatment environment. Dr. Seekins noted some positive preliminary findings that suggest that individual service plans that took a secondary condition into account were associated with more declines in problems than plans that did not, but no well-designed controlled studies have been conducted.
The individual service plan directs the activities of personal assistants, who are critical environmental elements. Again, preliminary findings suggest that turnover and change involving assistants are risk factors for secondary conditions. Also, the worse that the condition is initially, the larger the effect is.
Dr. Seekins’ group is also looking at the impact of nutrition with a study in group homes that analyzed the home’s menus and pantry contents for consistency with the formal nutrition plans for residents. They did not find a good match. An analysis of oral health arrangements in the homes showed a better match between these arrangements and the individual service plans.
Such analyses provide only snapshots of a person’s overall situation. As described by Dr. Seekins, the “wellness club” approach is designed to be comprehensive by integrating wellness goals into the individual planning process on an ongoing, annual basis. The approach covers healthy home environments, the individual’s health and lifestyle, assessment procedures for identification of the individual’s needs and priorities, staff instruction, and methods for program and progress evaluation. The club strategy is also designed to provide participants a sense of belonging to something that is fun. The approach is being pilot tested in Montana and Kansas. The formal evaluation process is just beginning.
The discussion began with a question about depression as a primary condition that can be disabling and that may also give rise to secondary conditions. Dr. Kemp agreed that depression can certainly be a primary condition as well as a secondary condition and that it may be complex to sort this out for an individual patient. The identification of coexisting or primary depression and the assessment of prevention and management needs in this context are important.
Another question arose about the risk factors for depression as a secondary condition. Part of Dr. Kemp’s response emphasized the importance of positive social relationships and, conversely, the potential for negative relationships to increase secondary depression. Also important are personal attitudes and appraisals; for example, self-blame for a disabling injury is associated with the development of depression.
One participant stressed the importance of considering behavioral health issues, noting that the primary disabling condition for 60 percent of Medicaid recipients is some kind of severe and persistent mental health problem. Also, for children and families, learning disabilities can be described as hidden disabilities because they may not be obvious. It may be difficult to convince people that a cognitive processing deficit or other learning condition exists, that it may not be susceptible to medical treatment, and that it will persist into adulthood with all sorts of possible negative consequences, including incarceration.
These observations prompted further discussion of the attitudes and expectations of health care professionals, for example, whether they consider depression to be a normal consequence of disability and then, to some extent, discount it or overlook it as a preventable and treatable source of distress and suffering that can complicate many aspects of care management. Dr. Kemp suggested that professionals who routinely care for people with disabilities are sensitive to depression as a potential problem and may evaluate an individual for depression when functional or other problems seem to be occurring or persisting to an extent that is not anticipated, given other aspects of the individual’s medical, treatment, and personal circumstances.
The chair asked if there were any general comments. One participant noted the changing profile of the American population and the increasing significance of immigrants. This development has implications and presents challenges for the provision of services to people with disabilities. Another concern is emerging disabilities or the transformation of diseases such as AIDS into chronic conditions for which people require rehabilitation or other assistance. The role of cognitive limitations was raised again with the observation that such limitations were far more common in connection with traumatic brain injury and other conditions than many of the prob-
lems discussed in the workshop. A related question was what to do about disabling conditions that “come out of left field,” with autism offered as an example.
One participant stated that the evidence base for understanding, preventing, and mitigating disabling conditions is getting stronger. She also cited a series of meetings that are being sponsored by the Agency for Healthcare Research and Quality and the Robert Wood Johnson Foundation to discuss alternative study designs to strengthen and promote research on these and other conditions. Given the practical and ethical challenges of conducting randomized, double-blind clinical trials, it is important to consider observational and other studies at the same time that efforts are made to improve the methods and tools for such studies. The capacity to undertake rehabilitation research is also relevant. Adequate funding for the support of such research is part of the picture, but more fundamental aspects of the research capacity or infrastructure may need attention.
A different research issue relates to the data used to describe and assess the delivery of rehabilitative services. The data in this area are much poorer than the data for emergency room services or physician office visits.
With respect to longitudinal data, one participant encouraged the committee to look at the National Children’s Study (sponsored by the National Institute on Child Health and Human Development and the Centers for Disease Control and Prevention, among other agencies).1 It may eventually generate long-term data relevant to secondary conditions and other problems discussed at the workshop.
Questions were raised about the extent to which an IOM study would examine problems that are not related primarily to health care, such as access to employment. Certainly, participation as a concept is itself not primarily medical. Likewise, the creation or modification of the built environment is not primarily a health care function, but that environment is very important not only to the participation of people in society but also to the maintenance of mobility and the prevention of mobility decline in people with chronic conditions. Schooling is another important arena that offers opportunities for prevention and health promotion.
The vulnerability of people with disabilities to physical or mental abuse by caregivers (which may range from the withholding of prostheses to sexual violence) is a further cross-cutting environmental issue. For example, inadequacies in adult protective services, inappropriate placements of people with certain kinds of disabilities, and a lack of treatment or protection in settings such as jails create conditions for abuse.
Information of this study can be found online at http://www.nationalchildrensstudy.gov/about/mission/index.cfm.
One participant cautioned against getting too preoccupied with theoretical or conceptual models to the extent that attention is diverted from significant practical and policy issues. The experiences of the two earlier IOM committees and those involved with WHO and ICF suggests that such a preoccupation is a likely—and difficult to avoid—hazard. In a final observation, another participant urged the IOM committee not to forget the personal dimension, the direct experiences of people living with disabilities.
American Academy of Pediatrics, American Academy of Family Physicians, and American College of Physicians-American Society of Internal Medicine. 2002. A consensus statement on health care transition for young adults with special health care needs. Pediatrics 110:1304–1306.
Castelli WP, Garrison RJ, Wilson PWF, Abbott RD, Kalousdian S, Kannel WB. 1986. Incidence of coronary heart disease and lipoprotein cholesterol levels: the Framingham Study. Journal of the American Medical Association 256:2835–2838.
Child Trends Data Bank. Available online at www.childtrendsdatabank.org. (Last accessed November 21, 2005.)
Crimmins EM, Saito Y, Reynolds SL. 1997. Further evidence on recent trends in the prevalence and incidence of disability among older Americans from two sources: the LSOA and the NHIS. Journals of Gerontology 52(2):S59–S71.
Dickens, C, Creed, F. 2001. The burden of depression in patients with rheumatoid arthritis. Rheumatology 40:1327–1330.
Fougeyrollas P. 1995. Documenting environmental factors for preventing the handicap creation process: Quebec contributions relating to ICIDH and social participation of people with functional differences. Disability and Rehabilitation 17(3-4):145-53.
Fougeyrollas P, Noreau L, St Michel G. 1997. Measure of the quality of the environment. ICIDH and Environmental Factors International Network 9(1):32-39.
Freedman VA, Martin LG, Schoeni RF. 2002. Recent trends in disability and functioning among older adults in the United States: a systematic review. Journal of the American Medical Association 288(24):3137–3146.
Guralnik JM, LaCroix AZ, Abbott RD, et al. 1993. Maintaining mobility in late life. I. Demographic characteristics and chronic conditions. American Journal of Epidemiology 137:845–857.
IOM (Institute of Medicine). 1991. Disability in America: Toward a National Agenda for Prevention. Washington, DC: National Press Academy.
IOM. 1997. Enabling America: Assessing the Role of Rehabilitation Science and Engineering. Washington, DC: National Academy Press.
Kemp BJ, Kahan JK, Krause JS, Adkins RH, Nava G. 2004. Treatment of major depression in individuals with spinal cord injury. Journal of Spinal Cord Medicine 27:22–28.
Keysor J, Jette A, et al. 2005. Development of the home and community environment (HACE) instrument. Journal of Rehabilitation Medicine 37(1):37–44.
Leveille SG, Guralnik JM, Hochberg M, Hirsch R, Ferrucci L, Langlois J, Rantanen T, Ling S . 1999. Low back pain and disability in older women: independent association with difficulty but not inability to perform daily activities. The Journals of Gerontology. Series A, Biological Sciences and Medical Sciences 54(10):M487–M493.
Lollar DJ. 1994 (February 17-19). Preventing Secondary Conditions Associated with Spina Bifida or Cerebral Palsy: Proceedings and Recommendations of a Symposium, Crystal
City, Virginia. Washington, DC: Spina Bifida Association of America. Available online at http://www.cdc.gov/ncbddd/dh/Publications/Conferences/1994aSB_CP/1994aTableofContents.htm. (Last accessed September 26, 2005.)
Lotstein D, McPherson M, Strickland B, Newacheck P. 2005. Transition planning for youth with special health care needs: results from the National Survey of Children with Special Health Care Needs. Pediatrics 115:1562–1568.
Manton KG, Corder LS, Stallard E. 1993. Estimates of change in chronic disability and institutional incidence and prevalence rates in the U.S. elderly population from the 1982, 1984, and 1989 National Long Term Care Survey. Journals of Gerontology 48(4):S153–166.
Manton KG, Corder L, Stallard E. 1997. Chronic disability trends in elderly United States populations: 1982–1994. Proceedings of the National Academy of Sciences 94(6):2593–2598.
Marge, M. 1988. Health promotion for persons with disabilities moving beyond rehabilitation. American Journal of Health Promotion 2(4):29–35.
Morris PLP, Robinson RG, Andrzejewski MS, Samuels J, Price TR. 1993. Association of depression with 10-year post-stroke mortality. American Journal of Psychiatry 150: 124–129.
Newacheck PW, Budetti PP, Halfon N. 1986. Trends in activity-limiting chronic conditions among children. American Journal of Public Health 76:178–184.
NRC/IOM (National Research Council and Institute of Medicine). 2004. Children’s Health, the Nation’s Wealth: Assessing and Improving Child Health. Washington, DC: The National Academies Press.
Sameroff AJ and Chandler MJ. 1975. Reproductive risk and the continuum of caretaking casualty. In Horowitz FD, Hetherington M, Scarr-Salapatek S, and Siegel G, eds. Review of Child Development Research, Vol 4, pp. 187–244. Chicago: University of Chicago Press.
Schoeni R, Martin L, Andreski P, Freedman V. 2005. Growing disparities in trends in old-age disability 1982–2001. American Journal of Public Health 95(11):2065-2070.
Shumway-Cook A, Patla AE, Stewart A, Ferrucci L, Ciol MA, Guralnik JM. 2002. Environmental demands associated with community mobility in older adults with and without mobility disabilities. Physical Therapy 82(7):670-81.
Surgeon General of the United States. 2005. The Surgeon General’s Call to Action to Improve the Health and Wellness of Persons with Disabilities. Washington, DC: U.S. Department of Health and Human Services. Available online at http://www.surgeongeneral.gov/library/disabilities/calltoaction/calltoaction.pdf. (Last accessed September 26, 2005.)
Verbrugge LM, Jette AM. 1994. The disablement process. Social Science Medicine 38(1): 1–14.
Volpato S, Blaum C, Resnick H, Ferrucci L, Fried LP, Guralnik JM. 2002. Women’s Health and Aging Study. Comorbidities and impairments explaining the association between diabetes and lower extremity disability: The Women’s Health and Aging Study. Diabetes Care 25(4):678–683.
Whiteneck G, Gerhart KA, et al. 2004. Identifying environmental factors that influence the outcomes of people with traumatic brain injury. Journal of Head Trauma Rehabilitation 19(3):191–204.
WHO (World Health Organization). 1980. International Classification of Impairments, Disabilities and Handicaps: a Manual of Classification Relating to the Consequences of Disease. Geneva, Switzerland: World Health Organization.
WHO. 2001. International Classification of Functioning, Disability and Health. Geneva, Switzerland: World Health Organization.