Wearable-sensor based walking and non-walking measures as progression markers in early to mid-stage Parkinson's disease.
Longitudinal wearable-sensor data identified multiple walking, non-walking, and composite digital measures that sensitively track progression in early- to mid-stage Parkinson's disease—some outperforming conventional clinical scales and detecting change within ~10 months.
What the AI sees
Longitudinal wearable-sensor data identified multiple walking, non-walking, and composite digital measures that sensitively track progression in early- to mid-stage Parkinson's disease—some outperforming conventional clinical scales and detecting change within ~10 months.
Research significance
These objective, remote digital biomarkers could increase sensitivity and temporal resolution in PD trials and monitoring, enabling smaller/shorter studies and earlier detection of treatment effects to accelerate therapeutic development.
Source abstract
Digital measures of walking and sedentariness may objectively quantify Parkinson's disease (PD) progression. We analyzed longitudinal wearable-sensor data to evaluate the sensitivity and specificity of digital walking and non-walking measures in ambulatory people with PD. We selected 26 individual and 6 composite measures with sufficient sensitivity and test-retest reliability in a development dataset (N = 171). Twenty measures showed significant within-participant changes, and 7 showed larger 2-year effect-size than gold-standard clinical measures in people with early-stage PD (N = 101, mean number of years since diagnosis [YSD], 2.2). Significant changes in non-walking and composite measures were detectable as early as 10 months. Twelve measures showed greater change in people with more advanced PD (N = 67; mean YSD 6.5) compared to matched non-PD individuals (N = 171). Sensitivity and specificity results indicate that measures capturing walking and especially non-walking behaviors hold promise as PD progression markers in early to mid-stage PD.