A systematic review of real-world gait-related digital mobility outcomes in Parkinson’s disease
Cameron Kirk, Emma Packer, Ashley Polhemus, Mhairi K. MacLean, Harry E. Bailey, Felix Kluge, Heiko Gaßner, Lynn Rochester, Silvia Del Din, Alison J. Yarnall
Abstract
Clinical assessments of Parkinson's disease (PD) focus on structured motor tasks, potentially underestimating real-world mobility impairment. Digital mobility outcomes (DMOs) from wearable devices, offer a promising alternative, but their clinical value remains unclear. This study reviewed the clinical utility of real-world DMOs in PD. Eight databases were searched (2000-2024). Eligibility included reports assessing real-world gait-based DMOs in PD. Study selection and data extraction were conducted independently by four reviewers. No meta-analysis was performed. Twenty-seven reports met inclusion criteria, reporting 21 DMOs. Walking speed was most common (92%), followed by step length (59%) and stride time (48%). Devices, assessment durations and walking bout definitions varied. DMOs differentiated PD from controls and real-world from supervised conditions. PD Subgroup analyses and associations with motor severity were inconsistent. Real-world DMOs offer valuable insight into PD mobility. Standardised protocols are needed for comparability. Further research should explore predictive utility and complex or composite DMOs. Trial Registration: PROSPERO 2025 CRD42021281213.