Correlation between outcomes of timed up and go (TUG) and five times sit to stand (FTSS) tests with trunk stability
Description
Raw data of completion time and trunk accelerometry for three versions of the TUG (conventional [TUG-C], dual-task [TUG-DT], and overline [TUG-OL] and the five times sit to stand (FTSS) tests. The first three columns (t1-t3) correspond to data of individual trials and the fourth column corresponds to means (M) for each participant (lines). In the current investigation, we assessed older individuals, with the following primary aims: (1) to evaluate the correlation of completion times observed in the FTSS and in different versions of the TUG test with a direct measurement of trunk stability given by accelerometry while performing these tests; (2) to compare completion time and trunk stability of a new version of the TUG test requiring increased dynamic balance with the versions being currently used of this test. As a secondary aim, we evaluated the correlation between tests for both completion time and trunk acceleration to estimate the extent to which performance in one test can predict performance in the others. The results indicated negative time-acceleration correlations for TUGC (rp = -.71, rp2 =.50, p <.01) and TUGDT (rp = -.77, rp2 =.59, p <.01) and a positive correlation for FTSS (rp =.73, rp2 =.53, p <.01). The TUGOL test failed to show significant time-acceleration correlations. In conclusion, our results suggest that completion time in the FTSS test importantly reflects dynamic balance stability in older individuals.
Files
Steps to reproduce
In all tests, a sequence of movements was to be performed in the shortest time, with the interpretation that short completion times indicate higher performance. Completion times were measured through a stopwatch, with visual detection of the onset and end of each trial. The following tests were evaluated: Five Times Sit to Stand (FTSS). The test was initiated with the participants sitting on a regular-sized chair (approximately 45 cm high), without armrests, keeping their feet hip-width apart fully supported on the floor. The test consisted of getting up and sitting down five times in the shortest time, refraining from discharging the whole body weight on the chair accent when sitting, while keeping their arms crossed over the chest. Timed Up and Go (TUG). For this test, three versions were analyzed. For the conventional version, participants started sitting on the chair, keeping both hands resting on the thighs and the feet hip-width apart fully supported on the floor. Following the examiner's verbal prompt, participants were to stand up, walk as quickly as possible toward a cone positioned 3 m away on the ground in front of the participant, circumvent the cone (180 degrees turning), return to the chair, and sit down. For the TUG dual-task version, participants performed the test as described for the conventional version while simultaneously performing a cognitive task. The cognitive component of this test consisted of speaking aloud names of colors, fruits or animals throughout the test duration, according to the initial letter spoken by the examiner immediately before trial onset. We also analyzed a new version of the TUG requiring increased dynamic balance. In this version, participants were to perform the gait component of the test by walking over a 5-cm width straight line, marked through a tape on the floor. The cone used in the other versions of this test was replaced by a transversal line crossing the end of the walking line. Participants were to cross this line with one foot before returning, stepping over the line to return to the chair. In all tests, a single familiarization trial was provided before the performance of three probing trials. In cases of failure to perform a trial, it was replaced immediately. The sequence of tests was randomized across participants. Sitting rest intervals of 1 minute between trials and 2 minutes between tests were provided to avoid fatigue. Data collection was completed in a single session of approximately 40 minutes. Accelerometer signals were sampled at a frequency of 1 kHz. After preliminary visual inspection of the signals, raw data were exported to a personal computer and processed offline by using MATLAB routines (Mathworks, Natick, MA). Raw signals were amplified with a gain of 1000 and filtered through a 10 Hz fourth-order double pass Butterworth filter.