Reliability and minimal detectable change for inertial measurement units. Derived stability, symmetry, and smoothness indexes of gait in people with multiple sclerosis

Background: people with multiple sclerosis (PwMS) experience loss of gait stability, smoothness, and symmetry, which affects quality of life and requires accurate evaluation for effective rehabilitation. Inertial measurement units (IMUs) offer a promising approach to monitor gait quality by quantifying indices reflecting stability, symmetry, and smoothness whose minimal detectable changes (MDCs) are not defined in people with MS (PwMS). Research question: to assess the within-day test-retest reliability and MDCs of IMUs – derived gait stability, symmetry, and smoothness metrics in PwMS during the 10 m walking test (10MWT). Methods: 58 PwMS wore five IMUs and performed the10MWT twice with a 10 min rest between each trial. Log dimensionless jerk (LDLJ) and improved Harmonic Ratio (iHR) were calculated for each gait trial based on the signals from the pelvis – mounted IMU, normalized Root Mean Square (nRMS) were calculated also from the head and sternum-mounted IMUs. Intraclass correlation coefficient (ICC) were calculated between the results of the two 10MWT to assess test -retest reliability, and minimal detectable change scores were calculated. Results: Reliability of the investigated parameters ranged from moderate to excellent values, with ICC ranging from 0.64 to 0.98. MDC values ranged from 0.09 to 0.53 for the nRMS, from 7.54 to 11.36 for the iHR and from 0.15 to 0.20 for the LDLJ. Significance: This study showed moderate to excellent reliability for the investigated indexes when calculated based on 10MWT, with the LDLJ showing the highest reliability, thus providing a reliable smoothness metric in pwMS. Also, nRMS showed good reliability, but caution is warranted with iHR due to its lower reliability and higher MDCs.