Screening older persons at risk of functional decline and high use of health care in a clinical setting may facilitate early intervention. The Established Populations for Epidemiological Study of the Elderly (EPESE), a large study that involved physical performance measures, was able to identify the risk of death and nursing home placement by administering a lower extremity performance battery. In this study, Studenski and colleagues assessed gait speed and the EPESE measures in elderly patients to determine whether these assessments could predict future health care use, health status, and functional status in a clinical setting.

Participants in the study were 65 years and older and received care from the Department of Veterans Affairs (VA) system (n = 140) or a Medicare health maintenance organization (HMO; n = 347). All of the participants had a score of 24 or higher on the Mini-Mental State Examination and a gait speed of 0.2 to 1.3 meters per second. The 487 eligible participants were given a series of health and depression assessments at baseline. Physical performance, health status, functional status, and symptoms were assessed at baseline and again at three, six, nine, and 12 months. Performance testing included gait speed and EPESE performance. Outcome domains were health care use (i.e., one or more hospitalizations, and one or more adverse events, including hospitalization, nursing home placement, and death); health status; and functional status.

The EPESE score predicted hospitalization in HMO and VA participants, whereas gait speed predicted hospitalization only in the HMO group. Global health declined in 12 percent of participants; both gait speed and EPESE scores were predictive of the decline in a pooled assessment. Functional decline in personal care occurred in 28 percent of participants, and physical function decline occurred in 27 percent of participants. Gait speed and EPESE score were strong predictors of personal care difficulty. Physical performance measures depended on baseline SF-36 scores (a physical function index). Patients who scored higher at baseline and who had poor performance scores had a slightly increased risk of decline.

The authors conclude that performance measures alone or in combination with the self-report measures included in this study are better predictors of outcomes than self-report alone. Incorporating simple physical performance measures in the office independently predicts use, change in health status, and decline in function. EPESE testing, which includes the chair rise and tandem stand tests, may have more predictive power than gait speed testing alone, but takes more time.