Simple counts of the number of basic ADL dependencies for long-term care research and practice.

General acceptance of a patterned progression of dependency in activities of daily living has led to the widespread practice of simply counting the individual's basic ADL dependencies to reflect his or her self-care needs and level of impairment. This method is convenient, and it is practical to the extent that individuals do fit a scaled pattern of dependency that allows some meaningful comparison among individuals and between groups to be made. This research, based on 3,611 Medicaid cases in Virginia, reports that 36 percent of those individuals screened for nursing home admission do not match a commonly accepted pattern of dependency. The analyses include a logistic regression procedure to explain the characteristics of the "ADL divergent" cases and a Guttman scaling procedure on the ADL data for the sample. Results of the analyses indicate that a Guttman scaling procedure does as well as, but not better than, the original Katz ADL scale, with both scales describing approximately two-thirds of the cases in the sample.