Heart failure & cardiomyopathiesCohort study
Remote monitoring of heart failure exacerbations using a smartwatch
- Prospective observational cohort study
- deep learning model
- held-out validation set
- external validation
Authors' conclusion
These results indicate that wearable-derived daily pVO2 provides earlier and improved risk discrimination compared with existing wearable fitness estimates and established clinical markers and offers a scalable and generalizable approach for longitudinal HF research and monitoring.
Study context
- Population
- Free-living patients with heart failure (HF) (Ted Rogers Understanding Exacerbations of HF (TRUE-HF) study) wearing Apple Watchmedian 94.5 d
- Exposure
- TRUE-HF vs All of Us
- Outcome
- Ability of Apple Watch data to predict pVO2 measured by in-clinic CPET
Main finding
As reported in the abstract
Wearable-derived daily pVO2 vs CPET-measured pVO2
Pearson's correlation, wearable-derived vs cardiopulmonary exercise testing (CPET)-measured pVO2
0.85
Observational data: associations, not effects.
Further findings
| Outcome | HR (95% CI) | P value | |
|---|---|---|---|
| Unplanned healthcare events per 10% drop in wearable-derived daily pVO2 · events at median 7.4 d after first 10% drop | 3.62 (1.37–9.55) | P < 0.01 | |
| 0.10.31310Hazard ratio (log scale) |
Unplanned healthcare utilization, external cohort (All of Us), reduced-sensor model
| Comparison | HR (95% CI) | P value | |
|---|---|---|---|
| Drops in wearable-derived daily pVO2 · events at median 21 d after first 10% drop · external validation | 1.32 (1.03–1.69) | P = 0.03 | |
| 0.512Hazard ratio (log scale) |
Original publication
Remote monitoring of heart failure exacerbations using a smartwatch. Nat Med. Published 20 March 2026. doi:10.1038/s41591-026-04247-3
Summarised from the published abstract; every number is checked against it