Quantity and quality of gait and turning in people with multiple sclerosis, Parkinson's disease and matched controls during daily living

Abstract

Clinical trials need to specify which specific gait characteristics to monitor as mobility measures for each neurological disorder. As a first step, this study aimed to investigate a set of measures from daily-life monitoring that best discriminate mobility between people with multiple sclerosis (MS) and age-matched healthy control subjects (MS-Ctl) and between people with Parkinson's disease (PD) and age-matched healthy control subjects (PD-Ctl). Further, we investigated how these discriminative measures relate to the disease severity of MS or PD. We recruited 13 people with MS, 21 MS-Ctl, 29 people with idiopathic PD, and 20 PD-Ctl. Subjects wore 3 inertial sensors on their feet and the lumbar back for a week. The Area Under Curves (AUC) from the receiver operator characteristic (ROC) plot was calculated for each measure to determine the objective measures that best separated the MS and PD groups from their respective control cohorts. Adherence wearing the sensors was similar among groups for 58-66 h of recording (p = 0.14). Quantity of mobility (activity measures, such as a median number of strides per gait bout, AUC = 0.93) best discriminated mobility impairments in MS from MS-Ctl. In contrast, quality of mobility (such as turn angle, AUC = 0.90) best discriminated mobility impairments in PD from PD-Ctl. Mobility measures with AUC > 0.80 were correlated with MS and PD clinical scores of disease severity. Thus, measures characterizing mobility impairments differ for MS versus PD during daily life suggesting that mobility measures for clinical trials and clinical practice need to be specific to each neurological disorder.

Keywords: Mobility; Multiple sclerosis; Neurological disorders; Parkinson’s disease.

Figures

Figure 1.

Comparison of group differences between people with MS verus MS-Ctl and between PD versus PD-Ctl for top three mobility measures discriminating MS and PD, respectively. P-values from t-tests between each neurological group and its control group are shown.

Figure 2.

AUC (in descending order) for each mobility measure discriminating people with MS and PD groups compared to their healthy control groups. The color-coding scheme is based on domains of mobility as shown in legend.

Figure 3.

Correlation of discriminative measures of daily mobility with clinical scores related to severity of MS and PD.