Using wearable sensors, we aimed to quantify differences in gait in people with multiple sclerosis (MS)

with different levels of disability compared to healthy controls. These measures could allow automated, objective and reliable quantification of disability and could be used in clinical monitoring and as biomarkers of disability progression in MS trials.

Quantification of disability in patients with MS using existing clinical tools, including the EDSS, has significant shortcomings, including the requirement for significant clinician time,  poor inter- and intra-rater reliability and poor responsiveness. 

59 people with secondary progressive MS and 24 healthy controls were asked to perform a detailed gait analysis including a six-minute walking test during routine clinical visits. People with MS were dichotomised into moderate disability (MSm, EDSS 3.0-5.0) or severe disability (MSs, EDSS 5-6.5). Three commercially available sensors were attached to lumbar spine and ankles. Twenty gait measures were calculated and grouped into four conceptual domains (rhythm, variability, asymmetry and, balance and coordination).

Between session reliability was calculated over two sessions (7-14 days apart). Differences between controls, MSm, MSs were tested using a non-parametric test with post hoc analysis with Bonferroni correction.

The automated analysis indicated that people with MS walked with significantly longer stride time, increased step variability and asymmetry, and reduced speed, stride length, regularity and symmetry. These measures were all significantly worse in people with MSs compared to MSm. The measures were highly reproducible with overall Intra Class Correlation Coefficient of 0.92 (above 0.9 is regarded as excellent). 

Automated gait analysis using wearable sensors is practical in the clinical setting, reproducible, and can accurately and reliably quantify walking differences between groups of people with MS with differing levels of disability. 

Longitudinal data are now being collected to verify sensitivity to disease progression.