Test-retest reliability of kinematic and kinetic parameters during dual-task stair walking in the elderly.

This study aims to evaluate the test-retest reliability of kinematics and kinetics during single and dual-task stair walking in the elderly. Fifteen healthy elderly adults were recruited. Kinematic and kinetic parameters were measured using an infrared motion analysis system (Vicon, Oxford Metrics Ltd., Oxford, United Kingdom) and force platforms (Switzerland, Kistler 9287BA and 9281CA). Participants were tested under single-task and dual-task (serial 3 subtractions or carrying a cup of water) conditions. Each participant completed two sessions on two separate days with a 1-week interval. Intraclass correlation coefficients (ICC), Pearson correlation coefficient (r), and Bland-Altman plot were used to assess the reliability of stair walking. When ascending stairs, the ICC of kinematics and kinetics ranged from fair to excellent (ICC = 0.500-0.979) in the single and dual tasks, except for step length (ICC = 0.394) in the single task. The r value of kinematics and kinetics ranged from 0.704 to 0.999. When descending stairs, the ICC of kinematics and kinetics ranged from good to excellent (ICC = 0.661-0.963), except for min hip moment (ICC = 0.133) and min ankle moment (ICC = 0.057) in the manual task. The r value of kinematics and kinetics ranged from 0.773 to 0.960 in the single and dual tasks. In the Bland-Altman plots, all the zero values and most of the dots fell in the 95% confidence interval, and the mean difference was found to be close to zero for all the parameters during stair walking. These results obtained from this study show the good test-retest reliability of step cadence, step speed, and step width during single- and dual-task stair walking in the elderly, and the poor reliability of step length during ascending stairs. All the kinetic parameters, including min hip moment, max knee moment, and min ankle moment, had good test-retest reliability during single- and dual-task stair walking, but min hip moment and min ankle moment had poor reliability during manual-task descending stair. These results may help researchers in the assessment of biomechanics of dual-task stair walking in the elderly and to interpret the effect of interventions in this population.