For adults, the degree of functional limitation was based on ability to perform activities of daily living (ADLs),1 three of the standard instrumental activities of daily living (IADLs)2 (Katz, 1983), and six other sensory and physical functions. Individuals needing assistance with ADLs were the most severely limited, followed by those needing assistance with IADLs. A broader category of limitation—a "substantial" limitation—included all of those needing assistance with ADLs or IADLs plus people who were unable to perform one or more of the sensory or physical functions, or who had difficulty with two or more of those functions. For children, functional limitation was based on the presence of either a physical condition that limits the ability to walk, run, or play, or a mental or emotional condition that limits the ability to learn or do school work. These limitations in children were considered equivalent to "substantial" limitations in adults. Questions on work limitations were asked of persons ages 16-72. Unlike the NHIS, limitations due to acute conditions were not excluded. The data on receipt of disability benefits cover the noninstitutional resident population ages 18-64.
In 1982 and again in 1984, the Health Care Financing Administration conducted surveys of the Medicare-eligible population aged 65 and over to assess the characteristics of persons with chronic disabilities. Both surveys defined disability as a current or expected limitation of 90 days or more in the ability to perform one or more ADLs or IADLs. For the 1982 survey, interviews were conducted only with people living in the community; residents of nursing homes were excluded (Manton, 1989; Macken, 1986). The 1984 survey reinterviewed survivors from the first study, including those who had moved into nursing homes, and conducted first-time interviews with new respondents (Manton, 1989).
Each year the NHIS supplements its core questionnaire with additional questions on special topics. In 1984 the special topic portion of the NHIS, the Supplement on Aging (SOA), addressed the health status and living arrangements of people aged 55 and older. The SOA collected detailed data on subjects that included the respondents' ability to perform ADLs and