Concurrent Validity, Test-Retest Reliability, and Sensitivity to Change of a Single Body-Fixed Sensor for Gait Analysis during Rollator-Assisted Walking in Acute Geriatric Patients

Body-fixed sensor (BFS) technology offers portable, low-cost and easy-to-use alternatives to laboratory-bound equipment for analyzing an individual’s gait. Psychometric properties of single BFS systems for gait analysis in older adults who require a rollator for walking are, however, unknown. The study’s aim was to evaluate the concurrent validity, test-retest-reliability, and sensitivity to change of a BFS (DynaPort MoveTest; McRoberts B.V., The Hague, The Netherlands) for measuring gait parameters during rollator-assisted walking. Fifty-eight acutely hospitalized older patients equipped with the BFS at the lower back completed a 10 m walkway using a rollator. Concurrent validity was assessed against the Mobility Lab (APDM Inc.; Portland, OR, USA), test-retest reliability over two trials within a 15 min period, and sensitivity to change in patients with improved, stable and worsened 4 m usual gait speed over hospital stay. Bland–Altman plots and intraclass correlation coefficients (ICC) for gait speed, cadence, step length, step time, and walk ratio indicate good to excellent agreement between the BFS and the Mobility Lab (ICC2,1 = 0.87–0.99) and the repeated trials (ICC2,1 = 0.83–0.92). Moderate to large standardized response means were observed in improved (gait speed, cadence, step length, walk ratio: 0.62–0.99) and worsened patients (gait speed, cadence, step time: −0.52 to −0.85), while those in stable patients were trivial to small (all gait parameters: −0.04–0.40). The BFS appears to be a valid, reliable and sensitive instrument for measuring spatio-temporal gait parameters during rollator-assisted walking in geriatric patients.