AHA Scientific Sessions 2026 Abstract · Nov 2, 2026
Cardio Effort: A Heart-Rate-Based Wearable Metric for Characterizing Exercise Dose-Response Relationships at Scale
By Munaretto J, Lynch S, Rantanen A et al.
Background: Quantifying cumulative exercise stimulus and its relationship with biometric changes accounting for both intensity and duration across diverse activity types remains an open challenge in exercise science. This work introduces Cardio Effort (CE) designed to capture cumulative exercise stimulus and enable its estimation at scale from a smart ring. We explore the associations of year over year (YoY) changes in CE with cardiovascular health metrics. Methods: We analyzed de-identified data from 215,568 Oura Ring users meeting a wear compliance criteria (≥20 hours/day, ≥22 days/month, ≥2 activities/week) across Jan-Mar 2025 and 2026, yielding 9,536,457 workouts. CE is calculated as the cube of the workout heart rate (HR) minus 2 month average resting HR, integrated over time. CE was computed from ring HR time series during automatically detected workouts scaled so that 1.0 is approximately a 30 minute jog. YoY changes in CE and overnight HR, HR variability (HRV), estimated VO2max (eVO2max), and estimated pulse wave velocity (ePWV) were computed. Using raw data we compared stable (CE < 0.02) and large (CE > 0.15) effort change groups via Mann-Whitney U test across four age stratified groups. Results: Higher CE associated with lower HR and ePWV, and higher HRV and eVO2max. Within subject analyses revealed nonlinear dose response relationships. A YoY 0.1 unit increase in daily CE was associated with a YoY 0.5 bpm reduction in HR and 1.0 ms HRV increase; a 0.3 unit increase was associated with a 1.0 bpm reduction in HR and >2.0 ms HRV gain. At +0.2 and +0.5 CE units, eVO ₂ max increased 0.5 and 1.4 mL/kg/min YoY, respectively; ePWV decreased 0.1 m/s at +0.5 units. Benefits showed diminishing returns beyond roughly 0.5 daily CE units.Group differences (stable vs large CE change YoY) were significant (p < 0.001) for all metrics. The large increase group showed favorable YoY changes relative to stable: HRV +1.94 ms (Cohen’s d = 0.20), eVO2max +1.69 ml/kg/min (d = 0.51), HR –2.09 bpm (d = 0.39), and ePWV -0.01 m/s (d = 0.08). Effect sizes were small to medium, largest for eVO2max and smallest for ePWV. These differences were also significant in all age cohorts except for eVO2max and ePWV in the age 60+ cohort. Conclusion: CE captures dose response relationships between wearable derived exercise load and cardiovascular biomarkers across a large real world cohort. Modest daily increases (~15-30 min jogging equivalent) account for the majority of measurable benefits.