Opportunities to Improve Survey Measures of Late-Life Disability: Part I - Workshop Overview. Advances in Measuring the Underlying Components of Disability


Disability is defined at the intersection of several key concepts: an individual’s capacity to physically and cognitively function, the physical and social environment, the tasks that individuals choose to carry out, and the accommodations they make to carry them out (e.g., use of assistive devices, use of personal care from another person, or changes in their behavior). Advances in the understanding and measurement of each of these components have taken place over the last decade.

Physical functioning. There are two basic approaches to measuring physical functioning--that is the ability to function without help or the use of assistive devices. Subjective, self-reported measures of functioning and objective, performance-based measures appear to measure different, although related, constructs (Reuben et al. 1995); however, when combined they appear to provide more precise predictors of mortality and future hospital costs (Reuben et al. 2004a, 2004b).

Physical performance measures involve an individual performing a movement or task according to a standardized protocol and a trained observer rating the performance using objective, predetermined criteria (Guralnik et al. 1989, 1996; Halter and Reuben 2000). Batteries have been developed to measure the basic components of functioning (e.g., strength, balance, coordination, flexibility, endurance) as well as physical movements (e.g., walking) and goal-oriented functions (e.g., ADLs and IADLs). NHANES, for example, includes performance measures of balance and strength in its mobile exam center protocol. Further work is needed to promote standardization of performance measures in research settings (Guralnik and Ferrucci 2003).

Self-reports of functioning include Nagi’s functional limitations items and other self-assessments of the ability to function without assistance. More recently, Sayers et al. (2004) report high correlations between performance-based measures and new self-reported measures of functioning from the Late-Life Function and Disability Instrument (LLFDI) (Haley et al. 2002). The LLFDI subscale asks about difficulty without help or special devices (none, a little, some, quite a lot, cannot do) with 32 functions (including, for example, unscrewing the lid off a previously unopened jar without using any devices; running half a mile or more; making a bed). They conclude that the functioning component of the LLFDI is an acceptable substitute for performance-based measures when self-report is a preferred mode of data collection format.

There are a number of challenges in incorporating measures of physical functioning into national surveys. In December 2003, the National Institute on Aging sponsored a meeting on physical functioning to review existing protocols and discuss issues related to their use in population-based studies (Suthers and Seeman 2004). Methodological challenges were outlined for both self-reports and performance-based measures. For example, both types of measures would benefit from enhanced discriminatory capability of measures. Other issues raised specifically about performance measures related to the feasibility of administering tests in the home, the need for systematic training of observers, and need to consider possible bias associated with differential refusal rates.

Measures of time use and participation. Although ADLs and IADLs have been used widely in research and policy making activities and have advanced our understanding of functioning in late-life, it is unclear the extent to which these measures--and IADLs in particular--reflect meaningful activities to older Americans today.

Information on time use can help identify how older Americans are spending their time. Time use information is typically collected either using “stylized” questions about time (e.g., last week how many hours did you spend _____) or using time use diaries (e.g., At 5 AM what were you doing? For how long?). The Bureau of Labor Statistics has recently revised the American Time Use Survey (Horrigan and Herz 2004), which takes advantage of computer-assisted telephone interviewing technology and time use diaries.

The reliability and validity of various strategies for collecting time use data has been explored. Juster et al. (2003), for example, find that similar estimates of labor market work hours and of historical trends are produced by diaries and stylized questions. Most often information on time use is reported in broad categories (e.g., unpaid work, paid work, self-care, active and passive leisure activities). For example, research in Canada suggests that older adults are active and engaged (Fast and Frederick 2004) and cross-national comparisons find similarities in the age patterns of activities (Gauthier and Smeeding 2003).

Measures of the environment and technological accommodations. Over the last decade awareness has grown that disability cannot be assessed outside of the physical and social environment and the accommodations individuals make to carry out purposeful tasks and roles (Brandt and Pope 1997; Satariano 1997). Current measures of disability rarely acknowledge the environment in which tasks take place. For example, questions about difficulty bathing generally do not address whether an individual bathes in a stall shower or a bathtub or whether they hold on to a grab bar or use a shower stool.

A recent project funded by the Office of the Assistant Secretary for Planning and Evaluation in HHS seeks to begin to remedy this gap by developing and evaluating new measures of assistive technology use and the home environment (Freedman and Agree 2005). The purpose of this project was to develop, pilot, and disseminate an instrument to be used in national surveys to collect information on assistive technology and environmental modifications used in both the home and workplace. The instrument has been extensively cognitively tested and then pilot tested under the direction of the National Center for Health Statistics by telephone with 360 older adults (ages 50 and older) living in the community. The final instrument consists of brief modules that may be adopted in whole or in part by national surveys. Concepts that have been measured include the use of assistive technology, intensity of use, the home environment, difficulty with ADLs residual to technology use, and the effectiveness of assistive devices on several quality of life dimensions.

Other instruments have been developed to assess barriers in the community that may impede mobility. For example, Shumway-Cook et al. (2003) have developed an instrument to measure environmentally specific mobility disability. Their approach involves self-reports of frequency of encounter and avoidance of 24 features of the physical environment, grouped into eight dimensions: distance, temporal, ambient, terrain, load, postural transitions, attention, and density. Initial evaluations of validity and reliability with a small sample suggest that the approach appears promising (Shumway-Cook et al. 2005).

TABLE 2. Anchoring Vignettes to Measure Self-Care: 2002 World Health Study*
  1. [Helena] pays a lot of attention to the way she looks. She requires no assistance with cleanliness, dressing and eating.
  2. [Anne] takes twice as long as others to put on and take off clothes, but needs no help with this. Although it requires an effort, she is able to bathe and groom herself, though less frequently than before. She does not require help with feeding.
  3. [Victor] usually requires no assistance with cleanliness, dressing and eating. He occasionally suffers from back pain and when this happens he needs help with bathing and dressing.
  4. [Sandra] lives on her own and has no relatives or friends nearby. Because of her arthritis, she is house-bound. She often stays all day in the same clothes that she has slept in as changing clothes is too painful. A neighbor helps her wash herself.
  5. [Sue] is quadriplegic and must be washed, groomed, dressed and fed by somebody else.
  1. Overall in the last 30 days, how much difficulty did [name of person/you] have with self-care, such as washing or dressing [yourself/himself/herself]?
  2. In the last 30 days, how much difficulty did [name of person/you] have in taking care of and maintaining [your/his/her] general appearance (e.g., grooming, looking neat and tidy, etc.)
Response Categories
  1. None
  2. Mild
  3. Moderate
  4. Severe
  5. Extreme/Cannot Do
* A list of vignettes, including those listed here, can be found at http://gking.harvard.edu/vign/.

View full report


"survmeasI.pdf" (pdf, 2.19Mb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®