Abstract
Background and Purpose: Sit-to-stand tests measure a clinically relevant function and are widely used in older adult populations. The modified 30-second sit-to-stand test (m30STS) overcomes the floor effect of other sit-to-stand tests observed in physically challenged older adults. The purpose of this study was to examine interrater and test-retest intrarater reliability for the m30STS for older adults. In addition, convergent validity of the m30STS, as well as responsiveness to change, was examined in older adults undergoing rehabilitation.
Methods: In phase I, 7 older adult participants were filmed performing the m30STS. The m30STS was standardized to allow hand support during the rise to and descent from standing but required participants to let go of the armrests with each stand. Ten physical therapists and physical therapist assistants independently scored the filmed m30STS twice, with 21 days separating the scoring sessions. In phase II, 33 older adults with comorbidities admitted to physical therapy services at a skilled nursing facility were administered the m30STS, Berg Balance Scale, handheld dynamometry of knee extensors, and the modified Barthel Index at initial examination and discharge.
Results: In phase I, the m30STS was found to be reliable. Interrater reliability using absolute agreement was calculated as intraclass correlation coefficient (ICC)2,1 = 0.737 (P <= .001). Test-retest intrarater reliability using absolute agreement was calculated as ICC2,k = 0.987 (P <= .001). In phase II, concurrent validity was established for the m30STS for the initial (Spearman [rho] = 0.737, P = .01) and discharge (Spearman [rho] = 0.727, P = .01) Berg Balance Scale as well as total scores of the modified Barthel Index (initial total score Spearman [rho] = 0.711, P = .01; discharge total score Spearman [rho] = 0.824, P = .01). The initial m30STS predicted 31.5% of the variability in the discharge Berg Balance Scale. The m30STS did not demonstrate significant correlation with body weight-adjusted strength measures of knee extensors measured by handheld dynamometry. The minimal detectable change (MDC90) was calculated to be 0.70, meaning that an increase of 1 additional repetition in the m30STS is a change beyond error.
Conclusion: The m30STS is a reliable, feasible tool for use in a general geriatric population with a lower level of function. The m30STS demonstrated concurrent validity with the Berg Balance Scale and modified Barthel Index but not with knee extensor strength to body weight ratio. One repetition of the m30STS was established as the MDC90 as change beyond error.