NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Research Council (US) Committee on National Statistics; National Research Council (US) Committee on Population. Improving the Measurement of Late-Life Disability in Population Surveys: Beyond ADLs and IADLs, Summary of a Workshop. Washington (DC): National Academies Press (US); 2009.

Cover of Improving the Measurement of Late-Life Disability in Population Surveys

Improving the Measurement of Late-Life Disability in Population Surveys: Beyond ADLs and IADLs, Summary of a Workshop.

Show details


This report is a summary of a workshop organized to draw upon recent advances to improve the measurement of physical and cognitive disability in population surveys of the elderly population. Are the measures of activities of daily living (ADLs) and instrumental activities of daily living (IADLs) used in many population surveys sufficient as the primary survey-based indicators of late-life disability? If not, should they be refined or should they be supplemented by other measures of disability in surveys? If yes, in what ways should disability measures be changed or modified to produce population estimates of late-life disability and to monitor trends. What further research is needed to advance this effort?


People in the United States today are living longer and healthier lives than ever before. The average life expectancy at birth increased from 54.8 years in 1915 to 68.2 years in 1950 to an estimated 77.8 years in 2005. The average life expectancy at age 65 has increased from 77.8 years in 1950 to 83.7 years in 2005 (He et al., 2005; National Center for Health Statistics, 2009). While chronic conditions increase with age (about 80 percent of people aged 65 years and older have at least one chronic condition and 50 percent have at least two), and so do disability and functional limitations, there has been a downward trend in the estimated prevalence of disability among the older population.

The most commonly used measures for estimating trends in disability from population surveys have been ADLs and IADLs. These measures were developed several decades ago in response to the recognized need to assess functional status and disability rather than relying solely on measures of self-perceived health or data on previous health histories. ADLs measure people’s abilities to perform basic tasks of daily life without assistance, including eating, dressing, bathing, using the toilet, and getting in and out of bed. IADLs measure people’s abilities to function independently in carrying out such activities as housework, preparing meals, shopping, managing money, and using the telephone. Questions about ADLs and IADLs are easy to administer to obtain self-reports or reports from caregivers. Estimates of the number and characteristics of older adults with difficulties in performing ADLs have important implications for health policies and programs and associated costs of health care for the elderly population.

Data on ADLs from the National Long-Term Care Survey (NLTCS) showed a marked decline in prevalence rates of disability among the elderly population from 25 to 19 percent over the period 1984 to 1999, an annual decline of nearly 2 percent per year from 1984 to 1999 (Manton et al., 1997; Manton and Gu, 2001). Researchers have identified some of the factors contributing to the decline: improvements in medical treatment including new and improved medicines and technology, improvements in health behavior, increased use of assistive devices to help people cope with impairments, environmental changes, changes in socioeconomic status and exposure to disease, and increase in educational attainment and related changes in occupational composition among older people.

Other research has broadly supported those findings, although with variations in the estimates of prevalence of ADLs or IADLs. However, not all studies have found declines for all measures (see, e.g., Crimmins et al., 1997; Waidman and Liu, 2000; Schoeni et al., 2001; Freedman et al., 2002; Freedman, 2006). Although the focus of this workshop is disability measurement in late life, note that trends in reported disability rates among younger cohorts, especially those 30–50 years of age, appear to be rising even as the elderly are becoming less disabled (see, e.g., Lakdawalla et al., 2004; Soldo et al., 2006).

Because of these inconsistencies, several efforts have been undertaken to examine the estimated prevalence rates and trends in disability. In 1988 the Federal Interagency Forum on Aging Related Statistics appointed a committee to systematically examine the different estimates from the various surveys measuring ADLs. The committee found substantial differences in the prevalence of disabilities for the elderly population across the various national surveys. Sources of variation included diversity in conceptual definitions of disability, the nature of reference periods used for recall, the disability measures used, the wording of questions, the sampling frame, the reporting rules for answering ADL questions, the order of the questions on the questionnaire, the modes of data collection (in person, by telephone, etc.), the age composition of the respondents, and the threshold levels used for construction of disability statistics and other features of the analysis (U.S. Department of Health and Human Services, 1989; Wiener et al., 1990).

About a decade later, the report of a workshop (National Institute on Aging and National Bureau of Economic Research, 2001) suggested that work was needed to reconcile different disability measures, such as chronic illness rates, functional limitations, ADLs, IADLs, uses of support services, and physical versus cognitive health; to craft new or revised survey questions to increase comparability across surveys; to test actual physical performance (e.g., climbing stairs) in addition to answering questions; and to develop measures that assess people’s ability to live independently, considering objectively measured physical and cognitive abilities and assistance from devices and services that foster independent living even without improvements in underlying health.

In September 2002, a technical working group, funded by the National Institute on Aging (NIA), considered how to resolve inconsistencies in old-age trends in ADLs across national surveys (Freedman et al., 2004). That was followed by the report of a workshop convened by the Office of the Assistant Secretary for Planning and Evaluation in the U.S. Department of Health and Human Services on improving survey measures of late-life disability, which also discussed needed work in this area (Freedman et al., 2005).

In addition to the issue of the inconsistencies of findings based on ADLs and IADLs is the question of whether ADLs and IADLs continue to be sufficient as survey-based measures of disability. Should they be refined or augmented with other measures that could better track progressive declines in functional status? A related question is how responses to ADLs and IADLs may have changed over time because of changes in the environment, such as availability of assistive devices, health care, and other social services that affect respondents’ abilities or perceptions of their abilities.

Other possible methods are now being developed for policy-relevant disability measures in addition to ADLs and IADLs that are feasible to administer in population surveys. One way is afforded by time-use survey methodology to refine or augment existing measures of ADLs and IADLs, ranging from 24-hour diary surveys to more intensive measurements of activity. Time-use surveys can capture the frequency and time spent on activities, such as getting dressed, exercising, driving, and shopping. Such methods offer ways to validate and improve other measures and may provide additional measures for standardization across surveys. Performance measures of functional limitations, such as chair stands, gait speed, and grip strength, could be included in surveys, and their relationship to current self-reports of ADLs and IADLs could be analyzed.

In addition to new methods to measure disability, other changes in survey methods offer opportunities for measuring and understanding disabilities. One such change is in the mode of surveys: from personal interviews using paper-and-pencil questionnaires to telephone interviews to computer-assisted personal interviews and, most recently, to the Internet, which may well be the wave of the future in terms of speed, flexibility, and low cost. Another possibility is the use of easily collected biomarkers of chronic diseases that may track decline in functionality over the life course to supplement ADL and IADL measures that capture changes in functionality across thresholds.

Another approach comes from the Patient-Reported Outcomes Measurement Information System (PROMIS) initiative at the National Institutes of Health, an effort to improve assessment of self-reported symptoms and health-related quality-of-life outcomes using contemporary outcome measurement methodology (Hambleton, 2005). PROMIS is developing and testing a large bank of questionnaire items, some of which may be appropriate for population surveys.

NIA has cofunded and assisted a wide array of surveys across the world, somewhat comparable to the Health and Retirement Study, to learn about the prevalence and the trends of disability in those countries. Cross-cultural and cross-national comparability of measures and data becomes a difficult issue.

Vignettes, as used in the World Health Survey of 2000–2001, is a promising approach to obtain comparability in self-reports of disability in population surveys when a lot of subjectivity is involved across important cross-national population groups. However, this method needs further research; how well it works and in what circumstances has not yet been established.

Given this history and recent developments, a reassessment is timely; NIA asked the Committee on National Statistics (CNSTAT), in collaboration with the Committee on Population (CPOP), to conduct a public workshop on the potential of using methodological advances to improve the measurement of physical and cognitive limitations in population surveys of older adults. The workshop was to focus on disability measures that are suitable for population surveys and that are most relevant for monitoring trends and policy-oriented research and evaluation.


An estimated 40 to 50 million Americans currently live with disability (Institute of Medicine, 2007). Regardless of whether the prevalance of disability continues to decline or increases among the older population, the number of people with disability is going to increase substantially as the U.S. population ages over the next several decades. The estimates of disability derived from population surveys are therefore increasingly important in order to monitor trends and to track societal progress in the coming years in addressing late-life disability in the population.

The lack of consistency in the concept and definition of disability, and anaysis of trends in the prevalence of disability in the population, has been a real barrier to moving forward in the area of measurement of disability. Because of different views about what is meant by the term disability, over the years researchers have used a variety of measures. Defining disability has occupied the attention of many individuals and groups in the United States for many years. As Jette and Badley (2002, pp. 183–184) explain

The field of disability research is in need of uniform concepts and a common language to guide scholarly discussion, to advance theoretical work on the disablement process, to facilitate future survey and epidemiological research, and to enhance understanding of disability on the part of professionals as well as the general public. A commonly understood language can also influence the development of public policy. . . . The current lack of a uniform language and commonly understood definition of the concept of “disability” is a serious obstacle to all these endeavors. . . . A common understanding of the term “disability” is an essential first step . . . and is the foundation for a fruitful discussion of improving survey research in the general area of disability. . . . Understanding the source of contemporary conceptual confusion requires a review of the major disability frameworks found in the literature. The goal of bringing together the several different schools of thought on disability and the disablement process remains elusive. Achieving a commonly accepted conceptual language is one of the primary challenges facing the field of disability research. . . .

How disability is defined is profoundly important before grappling with the measurement challenges of how to improve measures of disability, not only within a survey but also comparable across surveys both in the United States and in other countries. If the goal is to devise strategies on how best to “go beyond ADLs and IADLs,” the starting point is defining “disability.”

The lack of a common definition that can be considered the gold standard shows up throughout the field. The 2003 update of the 1995 paper prepared for the National Institute on Disability and Rehabilitation Research of the U.S. Department of Education shows about 67 federal statutes in which disability is defined (Cherry Engineering Support Services, Inc., 2003). There is overlap among definitions, some statutes have more than one definition, and others refer to definitions contained in other legislation. Broadly, of the 67 statutes or programs that define disability, 35 have self-contained definitions, 26 use definitions from another statute or section, and 6 use definitions from more than one statute or section (Cherry Engineeering Support Services, Inc., 2003).

How disability is defined will, explicitly or implicitly, set the goals for how society chooses to deal with disability. Different definitions will suggest different kinds of solutions and indicate different targets for intervention and action. It is important to reach some clarification on what is meant by the term “disability” (Iezzoni and Freedman, 2008).


To understand the source of contemporary conceptual confusion calls for a review of the major disability frameworks found in the literature. Since the 1960s, survey measures of disability have been rooted in ADLs and IADLs. One of the early large-scale studies was the Established Populations for Epidemiologic Studies of the Elderly (EPESE), which began in the 1980s (see Cornoni-Huntley et al., 1991). EPESE approached physical disability from an operational point of view, focusing on limitation in mobility, ADLs, and IADLs. It used what have become classic batteries, such as a modified Katz ADL scale (dependence in ADLs: walk, bathe, groom, dress, eat, get from bed to a chair, use a toilet) (see Katz and Akpom, 1976), Nagi’s functional limitation items from a survey conducted in early l970s (difficulty to push large objects, stop, crouch or kneel; reach above shoulder level; write or finger or handle small objects) (see Nagi, 1976), and Rosow and Breslau functional health scale (1966) (inability to perform heavy housework, climb stairs, walk a half mile)—all of which focused on very specific ADL- or IADL-type functions. The NLTCS has a heavy emphasis on ADL- and IADL-type metrics with 27 measures of disability that include the Katz ADL index, IADL items from Lawton and Brody (1969), and Nagi’s functional limitation items. It also includes some items on vision impairment.

Historically, in survey approaches to assessing disability the definition reflected a medical approach. In the medical view, disability emerges from specific diseases or pathologies. Disability is as an attribute of the person: it is a problem that the person has that is caused by various diseases, trauma, or other health conditions. The management of disability focuses on preventing the condition, curing it, or helping individuals adjust, or change, their behavior to better adapt to it. For example, the Social Security Administration (2003) defines disability as one’s inability to engage in substantial gainful activity and this inability is by reason of any medically determinable physical or mental impairment directly tied to an underlying condition. The American Medical Association in 2001 defined disability as an alteration of an individual’s capacity to meet personal, social, or occupational demands because of an impairment (Cocchiarella and Andersson, 2001). Disability is directly tied to underlying conditions.

A striking contrast to a medical view is a more social view of disability. This view has not been widely discussed in the aging literature. However, it is prevalent in the rehabilitation literature and the rehabilitation field, in which disability is not seen as an attribute of the individual; rather, disability is viewed primarily as a socially created issue. The crux of this definition is whether or not an individual is or can be fully integrated into roles in society. The inability to be fully integrated is what is defined as disability. For example, Disabled Peoples’ International1 defines disability from the social perspective as the loss or limitation of opportunities to take part in the normal life of the community on an equal level with others due to physical or social barriers.

In the social view, disability is the product of the physical, organizational, and attitudinal barriers in society, which lead to discrimination. It is very different from the medical view of disability. Perhaps the key difference is the assumption that disability is a socially created problem; it is not an attribute of the individual. The defining issue is attitudinal or ideological and requires social or environmental change. Disability is viewed as a human rights issue, not a medical issue. This view is illustrated in the independent living movement and in deinstitutionalization efforts. It is also reflected in the Americans with Disabilities Act.

There have been several efforts to develop a synthesis of both the medical and the social views. For example, there is Nagi’s work (1976) and models from the Institute of Medicine (1991, 1997), which defined disability as a phenomenon that results from the interaction between health conditions and contextual factors. They tried to take into account both the medical and social issues and made the argument that disability results from the interaction of both factors.

More recently, the World Health Organization ([WHO] 2001) developed a coherent synthesis view of what constitutes disability. The International Classification of Functioning, Disability and Health (ICF) defines disability as an umbrella term that encompasses differing components, referred to as impairments, limitations in one’s activity, and restrictions in one’s ability to participate in social roles. Unfortunately, this is not a precise use of the term disability. The more precise terms in the ICF are the impairments, the limitations in activity, and restrictions in participation. The WHO’s umbrella term for this collection of concepts is disability.

At the level of body functions and body structures, the WHO document refers to such conditions as hemiparesis (muscle weakness affecting only one side of the body) secondary to restricted blood flow, for example, to the brain—restrictions occurring at the level of organs or body systems. It also discusses limitations in activities or restrictions in participation, such as, functionally, an inability to walk, severe difficulty in dressing oneself, or difficulty in preparing meals. These limitations are similar to ADLs and IADLs, but they also include much broader social role behaviors, particularly under the concept of participation. Then very explicitly, the model tries to incorporate factors of the environment, such as social, attitudinal, and physical barriers and facilitators that affect either the body functions and structures or activities and participation. That is how the WHO has tried to pull together a synthesis of some of the different views of what constitutes disability.


The purpose of the workshop, as noted earlier, was to draw on the most current understanding of methodological advances to improve the measurement of physical and cognitive disabilities in population surveys of older adults, focusing on disability measures that are suitable for population surveys and that are most relevant for monitoring of trends and policy-oriented research and evaluation. To respond to these issues, CNSTAT, in collaboration with CPOP, appointed an expert workshop steering committee for the task. The committee was charged to plan the workshop, decide on commissioning technical background paper(s) as needed, develop the agenda and identify researchers for the presentations, and convene and serve as moderators for the workshop.

The steering committee believed that a focused discussion among a wide range of disability researchers and survey methodologists could identify a framework for a long-term research agenda in this area for NIA and others in the field. To contribute to that framework, the committee commissioned a background paper Population Survey Measures of Functioning: Strengths and Weaknesses by Barbara Altman, which is included as Appendix A in this volume. The workshop drew people from a wide variety of disciplines and perspectives from federal agencies, academia, nongovernmental organizations, and the public. The workshop agenda and presenters and biographical sketches of steering committee members are shown in Appendixes B and C, respectively.

The workshop was structured to combine formal presentations and discussion among the participants. This report is a summary of the presentations and the group discussions flowing from the presentations during the sessions outlined in the agenda. The next chapter opens with an overview of the technical background paper, then discusses the challenges on improving current measurement of late-life function and disability. Chapter 3 focuses on potential methods for refining or augmenting current measures. Chapter 4 addresses issues related to improving the validity of cross-population comparisons, and Chapter 5 addresses the need to measure functioning and disability in context. The final chapter focuses discussion on the future in terms of suggesting research areas to advance the current efforts from the perspective of the participants attending the workshop.

This report is a summary of what transpired at the workshop and therefore limited to the views and opinions of those participating in the workshop. It reflects the concerns and areas of expertise of the workshop participants. As such, it does not provide a comprehensive review of the research and current status of survey measurement of disability.



Disabled Peoples’ International is a network of national organizations or assemblies of disabled people, established to promote their human rights through full participation, equalization of opportunity, and development.

Copyright © 2009, National Academy of Sciences.
Bookshelf ID: NBK28468


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (1.2M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...