Application of the Sports Technology Research Network Validation Framework to Evaluate Wearable Heart Rate Monitors for the Management of Parkinson's Disease

Parkinson’s disease (PD) affects over 6.1 million people worldwide and is characterized by progressive dopaminergic neurodegeneration in the substantia nigra, contributing to motor symptoms and reduced quality of life. Aerobic exercise at intensities near 80% HRmax promotes dopamine release and neuroplasticity, making heart rate (HR)-guided training a promising therapeutic strategy. Although the Rating of Perceived Exertion (RPE) is commonly used to estimate intensity, it lacks precision for monitoring real-time training load. Wearable HR monitors offer an objective alternative; however, limited validation exists in exercise contexts relevant to clinical populations such as PD. PURPOSE: To assess the accuracy of three commercial HR monitors (Polar, Garmin, WHOOP) compared to a gold-standard ECG (GE Dash 5000) during treadmill-based aerobic exercise in a healthy cohort, to inform future application in PD rehabilitation. METHODS: In this IRB-approved study (#2113291-4), 45 healthy adults completed a standardized treadmill protocol while HR was recorded every 10 seconds. Accuracy was evaluated using concordance correlation and Bland-Altman analysis. RESULTS: Polar showed the highest agreement with ECG (r² = 0.71), followed by Garmin (r² = 0.62) and WHOOP (r² = 0.26). WHOOP exhibited the greatest variability (SD = 19.86), especially at HRs above 125 BPM. Data loss rates were 11.5% (ECG), 2.5% (Polar), 5.3% (Garmin), and 19.6% (WHOOP). Polar and Garmin exhibited substantially lower variability compared to WHOOP (19.86), with SDs of 10.80 and 12.55, respectively. CONCLUSION: Among the devices tested, Polar provided the most accurate and consistent HR data relative to ECG. Validating wearable monitors in a healthy population performing structured aerobic exercise is a critical step toward safely integrating HR-guided training into PD rehabilitation, where precision is essential for driving neuroplastic adaptation.