Home > Research > Publications & Outputs > Intraday reliability and sensitivity of four fu...
View graph of relations

Intraday reliability and sensitivity of four functional ability tests in older women

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published
<mark>Journal publication date</mark>1/01/2014
<mark>Journal</mark>American Journal of Physical Medicine and Rehabilitation
Issue number8
Volume93
Number of pages5
Pages (from-to)703-707
Publication StatusPublished
<mark>Original language</mark>English

Abstract

Functional tests are commonly used to evaluate the functional ability of older individuals; however, intraday reliability and sensitivity are required to enable informed decisions on whether repeated trials are necessary and to ensure that the values obtained from a single session are a patient's true score. This study aimed to investigate the intraday reliability and sensitivity of four commonly used functional tests in older individuals. Seventy-one healthy older women (mean [SD]: age, 71.7 [7.3] yrs; body mass, 64.8 [10.2] kg; stature, 1.58 [0.07] m) performed the 6-m maximum walking speed, timed 8-foot up-and-go, chair sit-and-reach, and back scratch tests three times in one single session, with 1 min between trials. Reliability between all trials was examined using intraclass correlation coefficient, and sensitivity was examined using typical error. All tests were highly reliable (intraclass correlation coefficient range, 0.89-0.99), indicating no need for a familiarization trial. Typical error between trials 2 and 1 were 0.06 m·s, 0.42 s, 1.13 cm, 0.92 cm for the 6-m maximum walking speed, timed 8-foot up-and-go, chair sit-and-reach, and back scratch tests, respectively. Practitioners should perform two trials to examine whether the difference between them is less than the typical error reported here. These results should help practitioners ensure that scores obtained from an individual from these functional tests are a true reflection of their functional ability rather than measurement error.