The application of principal component analysis to characterize gait and its association with falls in multiple sclerosis

• Main Authors: Andrew S. Monaghan, Jessie M. Huisinga, Daniel S. Peterson
• Format: Article
• Language: English
• Published: Nature Publishing Group 2021-06-01
• Series: Scientific Reports
• Online Access: https://doi.org/10.1038/s41598-021-92353-2

Abstract People with multiple sclerosis (PwMS) demonstrate gait impairments that are related to falls. However, redundancy exists when reporting gait outcomes. This study aimed to develop an MS-specific model of gait and examine differences between fallers and non-fallers. 122 people with relapsing–remitting MS and 45 controls performed 3 timed up-and-go trials wearing inertial sensors. 21 gait parameters were entered into a principal component analysis (PCA). The PCA-derived gait domains were compared between MS fallers (MS-F) and MS non-fallers (MS-NF) and correlated to cognitive, clinical, and quality-of-life outcomes. Six distinct gait domains were identified: pace, rhythm, variability, asymmetry, anterior–posterior dynamic stability, and medial–lateral dynamic stability, explaining 79.15% of gait variance. PwMS exhibited a slower pace, larger variability, and increased medial–lateral trunk motion compared to controls (p < 0.05). The pace and asymmetry domains were significantly worse (i.e., slower and asymmetrical) in MS-F than MS-NF (p < 0.001 and p = 0.03, respectively). Fear of falling, cognitive performance, and functional mobility were associated with a slower gait (p < 0.05). This study identified a six-component, MS-specific gait model, demonstrating that PwMS, particularly fallers, exhibit deficits in pace and asymmetry. Findings may help reduce redundancy when reporting gait outcomes and inform interventions targeting specific gait domains.