Total: 1
The observational GaitLab study (ISRCTN15993728) enrolled adult people with multiple sclerosis (PwMS) with Expanded Disability Status Scale (EDSS) <=6.5. PwMS performed the U-Turn Test (UTT), a smartphone-based assessment of dynamic balance, in a gait laboratory (supervised setting) using 6 smartphones at different body locations and daily during a 2-week remote period (unsupervised setting) using 1 smartphone. In the supervised setting, the accuracy of detecting turns with smartphones was compared against turns detected with a motion capture system (mocap) using F1 scores. Agreement between turn speed measured with smartphones and mocap was assessed by intraclass correlation coefficient (ICC[3,1]) and bias. In the unsupervised setting, test-retest reliability was assessed by ICC(2,1), and correlations with clinical and patient-reported measures by Spearman rank correlation. Ninety-six PwMS were included. In the supervised setting, turns were detected with high accuracy (F1 scores >95% across smartphone wear locations). Smartphone-derived turn speed was comparable across the supervised (1.44 rad/s) and unsupervised settings (1.47 rad/s), and with mocap-derived turn speed (1.47 rad/s). ICC(3,1) revealed high agreement between smartphone- and mocap-derived turn speed (ICC[3,1]: 0.87-0.92 across smartphone wear locations). Bias was minimal (-0.04 to 0.11 rad/s). In the unsupervised setting, test-retest reliability (ICC[2,1]) was >0.90 when aggregating >=2 tests. The UTT correlated with Timed 25-Foot Walk gait speed, EDSS, Ambulation score, 12-item Multiple Sclerosis Walking Scale, and Activities-specific Balance Confidence scale (r=-0.79 to -0.61). The UTT measures turn speed accurately and reproducibly irrespective of smartphone wear location and settings. These findings affirm its potential as a valuable tool in multiple sclerosis trials.