RESEARCH REPORT POSTER DISPLAY

VALIDITY OF THREE PHYSICAL PERFORMANCE MEASURES IN INPATIENT GERIATRIC REHABILITATION. Brooks D^1-3,5, Aileen D^1,3, Gary N^1,3; ¹Graduate Department of Rehabilitation Science, University of Toronto, Toronto, Canada. ²Department of Physical Therapy, University of Toronto, Ontario. ³Toronto Rehabilitation Institute, Toronto, Canada. ⁴Baycrest Centre, Toronto, Ontario. ⁵West Park Healthcare Centre, Toronto, Ontario

PURPOSE: To evaluate the construct validity and the responsiveness of three measures of physical performance measures as outcome measures for frail older persons. RELEVANCE: Clinicians are often faced with the dilemma of which physical performance measure to use in different settings such as inpatient programs, outpatient clinics or in people’s homes. Physical performance measures can have different measurement properties in different patient populations and the literature on the use of these measures in frail elderly is conflicted. PARTICIPANTS: Fifty-two subjects (35 women, 17 men; age, 80 + 8 years). METHODS: Pre-post design was used with measures at admission and discharge from three inpatient geriatric rehabilitation programs. Physical performance measures were timed up and go (TUG), two-minute walk test and functional reach. Functional status was measured with the Functional Independence Measure (FIM) and the modified Barthel Index. ANALYSIS: To determine if performance measures define known groups of patients according to their ambulatory status at discharge, a one-way ANOVA was performed. To determine the relationship between performance and functional status measures, correlation analysis was performed using Pearson’s correlations, correlation analysis was performed using Pearson’s correlations. The standardized response mean was calculated by dividing the mean change by the standard deviation of the change score. RESULTS: The TUG and two-minute walk test scores were significantly different in groups of patients using different ambulatory aids (p = 0.006), while no such difference was observed for the functional reach (p = 0.4). The correlations between the TUG/two-minute walk test and the FIM were −0.59 and 0.59 (p < 0.0001), respectively, at admission, and −0.42 and 0.47 (p < 0.04), respectively, at admission and discharge. The correlations between functional reach and the FIM were not significant (p > 0.09). Standardized response means were 1.1 for the TUG, 0.7 for the two-minute walk test and 0.5 for functional reach. CONCLUSIONS: The TUG and two-minute walk test are valid and responsive outcome measures in older persons participating in geriatric rehabilitation. Functional reach was a moderately responsive outcome measure, but did not consistently reflect ambulatory or functional status. IMPLICATIONS: Physical performance measures appear to provide somewhat different information than functional status measures. The strong correlation between the TUG and two-minute walk test suggest a considerable overlap between these two measures. The TUG may offer some advantages over the two-minute walk as an outcome measure for this population. KEYWORDS: Geriatrics, physical performance, walk test, exercise, function, walking, ambulation. FUNDING ACKNOWLEDGEMENTS: Supported by grants from the Toronto Rehabilitation Institute and Collaborative Research Program, Rehabilitation and Long-Term Care. Drs. D. Brooks and A. Davis hold New Investigator Awards from the Canadian Institute for Health Research. Dr. G. Naglie is supported by the Mary Trimmer Chair in Geriatric Medicine Research, University of Toronto. CONTACT: dina.brooks@utoronto.ca

ETHICS COMMITTEE: Toronto Rehabilitation Institute, Baycrest and West Park Research Ethics Committees