Lower extremity performance in nondisabled older persons as a predictor of subsequent hospitalization

Brenda W. Penninx, Luigi Ferrucci, Suzanne G. Leveille, Taina Rantanen, Marco Pahor, Jack M. Guralnik

Research output: Contribution to journalArticleAcademicpeer-review

319 Citations (Scopus)


Background. This study examines, in initially nondisabled older persons, the impact of reduced lower extremity performance on subsequent hospitalizations. Methods. A 4-year prospective cohort study was conducted among 3381 persons, aged 71 years and older, who initially reported no disability. At baseline, lower extremity performance was measured by an assessment of standing balance, a timed 2.4-m walk, and a timed test of rising from a chair five times. Data on subsequent hospital admissions and discharge diagnoses over 4 years were obtained from the Medicare database. Results. During the follow-up period, nondisabled persons with poor lower extremity performance spent significantly more days in the hospital (17.7 days) than those with intermediate and high performance (11.6 and 9.7 days, respectively). Poor lower extremity performance in nondisabled persons significantly predicted subsequent hospitalization over 4 years (relative risk for hospitalization in those with poor vs high performance: 1.78; 95% confidence interval, 1.45-2.17). This increased hospitalization risk could not be explained by several indicators of baseline health status. Increased hospitalization risks were especially found for geriatric conditions, such as dementia, decubitus ulcer, hip fractures, other fractures, pneumonia, dehydration, and acute infections. Conclusions. Even in persons who are currently nondisabled, a simple measure of lower extremity performance is predictive of subsequent hospitalization, especially for geriatric conditions.

Original languageEnglish
Pages (from-to)M691-M697
JournalJournals of Gerontology - Series A Biological Sciences and Medical Sciences
Issue number11
Publication statusPublished - 1 Jan 2000
Externally publishedYes

Cite this