Sleep HRV Variability Reveals Lifestyle Habits and Aging Patterns in 21,000 Users
A new digital biomarker derived from nightly HRV fluctuations links alcohol, inactivity, and poor sleep to cardiac autonomic disruption across age and sex.
Summary
Researchers analyzed nearly 2 million nocturnal heart rate variability (HRV) readings from over 21,000 wearable device users to validate HRV coefficient of variation (HRV-CV) as a digital biomarker. HRV-CV captures day-to-day fluctuations in cardiac autonomic function during sleep. The study found that just five nights of data are sufficient to reliably estimate a full 7-day HRV-CV score. Higher HRV-CV correlated with worse health behaviors—including greater alcohol intake, lower physical activity, shorter and irregular sleep—and with older age and higher BMI. Age and sex patterns differed notably: males showed rising HRV-CV after age 40, while females showed a U-shaped curve, declining through midlife and rising again after 50. These findings position HRV-CV as a scalable, behavior-sensitive tool for personalized health monitoring.
Detailed Summary
Heart rate variability (HRV) is a well-established marker of autonomic nervous system health, but single-point HRV measurements can be noisy. HRV coefficient of variation (HRV-CV)—tracking how much HRV fluctuates day to day—may offer a more stable and informative window into a person's physiological resilience and lifestyle habits. This study is the first large-scale characterization of sleep-derived HRV-CV as a digital biomarker.
The research team analyzed approximately 2 million nocturnal HRV readings collected via wearable devices from more than 21,000 adults, stratified by age and biological sex. They used simulation modeling to determine the minimum number of sleep nights required for reliable HRV-CV estimates, and tested associations with alcohol consumption, physical activity, sleep duration, sleep consistency, and behavioral variability.
Key results showed that five out of seven nights of sleep data are sufficient to achieve reliable 7-day HRV-CV estimates (ICC ≥ 0.80). Higher HRV-CV was significantly associated with greater alcohol use, lower physical activity, shorter sleep, irregular sleep patterns, and higher behavioral variability overall. Notably, alcohol and sleep consistency showed stronger associations with HRV-CV than with standard HRV alone, suggesting HRV-CV may be more sensitive to behavioral inputs.
Age and sex differences were striking. In males, HRV-CV rose sharply after approximately age 40. In females, it followed a U-shaped trajectory—declining through midlife and rebounding after age 50, potentially reflecting hormonal transitions around menopause. BMI was positively associated with HRV-CV in both sexes.
These findings support HRV-CV as a scalable, passive, behavior-sensitive biomarker captured during routine sleep. For longevity-focused clinicians and individuals, it offers a practical lens for identifying autonomic dysregulation tied to modifiable lifestyle factors and age-related cardiovascular risk.
Key Findings
- Five nights of sleep data reliably estimate 7-day HRV-CV (ICC ≥ 0.80), enabling practical tracking.
- Higher HRV-CV associates with greater alcohol use, lower activity, and shorter, irregular sleep.
- Males show rising HRV-CV after age 40; females show a U-shaped pattern, rising again after age 50.
- HRV-CV correlates more strongly with alcohol and sleep behavior than standard HRV metrics.
- Higher BMI links to elevated HRV-CV in both sexes, indicating metabolic-autonomic overlap.
Methodology
Observational analysis of ~2 million nocturnal HRV readings from >21,000 WHOOP wearable users, stratified by age and sex. Simulation modeling determined minimum data requirements; regression models assessed behavioral and demographic associations with 7-day HRV-CV.
Study Limitations
Data derive exclusively from WHOOP device users, who skew toward health-conscious, active populations, limiting generalizability. The study is observational, precluding causal inference between behaviors and HRV-CV. Only biological sex was recorded, not gender identity or hormonal status, which may affect interpretation of female-specific patterns.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe: