單任務與雙任務計時起走測試之信度與最小可偵測變化值

Reliability and Minimal Detectable Change of Single- and Dual-Tasking Timed Up & Go Tests

目的：本研究旨在探討單雙任務計時起走測試之達成穩定表現所需測試數、再測信度、最小可偵測變化值（MDC）、以及建構效度。方法與結果：本橫斷式研究收錄66位50歲以上之社區活躍中老年人。主要變項為執行單任務計時起走測試（TUG_(single)）、搭配一拿水杯任務之雙任務計時起走測試（TUG_(manual)）、搭配一連續減3任務之雙任務計時起走測試（TUG_(cognitive)）所需時間，每種任務在練習1次後皆測量3次。此外記錄受試者過去跌倒經驗，並測量衰弱指標。受試者中7位在1周後返回再測。根據這7位受試者（平均年齡69.5 ± 7.7歲）的資料，從第一次測量的單次資料到前兩次測試的平均，標準測量誤(SEM)和MDC顯著下降，尤其在TUG_(manual)特別明顯。使用三次測試的平均並沒有進一步大幅下降SEM和MDC，在TUG_(cognitive)反而增加了。使用前兩次測試的平均，再測信度（組內相關係數；ICC）和MDC分別為0.994-0.965和0.37-1.53秒。根據66位受試者（平均年齡71.6 ± 8.1歲）的資料，TUG_(single)和TUG_(manual)都和衰弱指標數和跌倒次數顯著相關，TUG_(manual)在控制年齡後的部分相關係數仍然維持顯著（部分相關係數分別為0.269和0.263）。結論：和TUG_(single)以及TUG_(cognitive)相比，在社區中老年人中TUG_(manual)具有建構效度和絕佳再測信度。測量兩次之後取平均值可以最佳化TUG_(manual)的再測信度（ICC 0.965）和MDC（1.53秒）。然而，上述結果是根據於小樣本，未來還需要更多研究來加以測試。

Purpose: This study aimed to determine the number of trials necessary for stable performance measuring-retest reliability, the minimal detectable change (MDC), and the construct validity of singleand dual-tasking Timed Up & Go Tests (TUG). Methods and Results: This cross-sectional study recruited 66 adults aged 50 years and over who actively participated in local community programs. Time taken to complete single-tasking TUG (TUG_(single)) and dual-tasking TUG, carrying a cup of water (TUG_(manual)), or performing serial-3 subtraction (TUG_(cognitive)) while executing TUG, was measured three times each after one practice trial. Participants were interviewed for their previous history of falls. Frailty status was defined based on Fried’s phenotypic definition. A subgroup of 7 participants was tested again after 1 week. Of the 7 participants (mean age= 69.5 ± 7.7 years), both the standard error of measurement (SEM) and MDC significantly dropped from the first trial to the mean of the first two trials, especially so for TUG_(manual). Using the mean of three trials did not largely decrease SEM and MDC further, and it increased TUG_(cognitive) values. Retest reliability (intraclass correlation coefficients; ICC) and MDC by using the mean of the first two trials were 0.994-0.965 and 0.37-1.53 seconds respectively. In the 66 participants (mean age= 71.6 ± 8.1 years), both TUG_(single) and TUG_(manual) were significantly correlated with degree of frailty and number of previous falls, and the coefficient remained significant after controlling for age in TUG_(manual) (0.269 and 0.263 respectively). Conclusion: TUG_(manual) was found to have construct validity and high retest reliability, compared to TUG_(single) and TUG_(cognitive) in community-dwelling middle-aged and older adults. It was found that by conducting two trials and taking the average of TUG_(manual), retest reliability (ICC 0.965) could be optimized with MDC 1.53 seconds. These results, however, were based on a small sample and should be tested in further studies.