Prospective value of self-report and performance-based tests of functional status for 18-month outcomes in elderly patients.

To determine predictors of death, nursing home placement and hospital admission, a prospective study was carried out in a university-affiliated geriatric hospital and a general practice. One hundred and thirty-five patients consecutively admitted from home to the geriatric hospital and discharged home again (site 1), and 144 patients, aged 70 years and over, of a general practice (site 2) were recruited for a comprehensive geriatric assessment program. At baseline, none was completely dependent on others, or severely demented. At follow-up after 18 months, 46 subjects (17%) had died, 20 subjects (7%) had been institutionalized, and 79 had been admitted to hospital (28%) in the interim. Univariate analysis demonstrated a significant association between Barthel Activities of Daily Living (ADL), Lawton-Brody Instrumental Activities of Daily Living (IADL), Mini-Mental State Examination (MMSE), Balance and Gait Evaluation, Timed "Up and Go", Timed Test of Money Counting. Grip-Strength and Williams Board Test, and death as well as nursing home placement. Barthel-ADL were significantly correlated with hospital admission. In a logistic regression analysis, Barthel-ADL were independent predictors for death; Barthel-ADL, Timed "Up and Go", Timed Test of Money Counting, and Williams Board Test were independent predictors for nursing home placement. No independent predictor was found for hospital admission. It is concluded that self-report and performance-based measures of functional capabilities are useful instruments to identify patients at risk for nursing home placement and death. Factors contributing to hospital admissions are other than those measured by the applied tests.