Predicting Parkinson’s Disease Motor Progression Using Digital and Clinical Data

This study explores the use of smartphone-based digital health technologies to identify and predict motor progression in patients with Parkinson’s disease. By applying data-driven clustering to clinical scores, researchers discovered that approximately one-quarter of participants were "fast progressors," a distinction traditional clinical categories failed to capture. Integrating digital biomarkers from smartphone tasks—such as gait and tremor—with standard clinical evaluations significantly improved the accuracy of long-term motor trajectory predictions. The findings demonstrate that high-frequency, objective data can effectively stratify patients at the individual level, addressing the challenge of disease heterogeneity. Furthermore, the approach showed high user acceptability, suggesting it is a feasible tool for enhancing the efficiency of future clinical trials. Ultimately, this framework supports the delivery of personalized medicine by identifying those at the highest risk for rapid decline