Are performance-based measures sufficiently reliable for use in multicenter trials? Musculoskeletal Impairment (MSI)...

Area of Science:

• Gerontology
• Rehabilitation Science
• Clinical Measurement

Background:

• Limited research exists on the test-retest reliability of performance-based measures.
• Assessing functional limitations in frail older adults often relies on these measures.

Purpose of the Study:

• To determine the test-retest reliability of seven timed, performance-based measures.
• To evaluate their utility in assessing functional limitations in frail, older adults.

Main Methods:

• One hundred and five frail, elderly subjects underwent two test administrations approximately two weeks apart.
• A battery of timed tests included the 8-foot walk, get-up-and-go, stair climb, chair stands, and 10-pound lifts.
• Agreement between mean times for each task across administrations was assessed using intraclass correlation coefficients.

Main Results:

• Intraclass correlation coefficients ranged from 0.25 (single chair stand) to 0.79 (8-foot walk).
• The 8-foot walk demonstrated the highest reliability among the tested measures.
• Only the single 10-pound lift showed a significant difference in time between the two administrations.

Conclusions:

• Timed performance-based measures exhibit a broad spectrum of test-retest reliability.
• Protocols involving familiar tasks with discrete start and end points tend to yield higher reliability.
• Unfamiliar tasks or those with ambiguous elements may result in lower reliability.