Wearable-derived cardiovascular fitness age and its lifestyle correlates in 442 adults

This study of 442 adults demonstrates that wearable-derived cardiovascular fitness age is a valid metric that significantly correlates with independent lifestyle factors like sleep quality and physical activity, while also tracking physiological improvements over time.

The Gist

Imagine your body is a high-performance car. Usually, when you check the car's health, a mechanic looks at the engine, the tires, and the oil to give you a complex report full of technical jargon. But what if, instead of a mechanic's report, you got a simple number: "Your car is 45 years old," even though you bought it when it was only 30?

That is essentially what this study is about. It looks at a new feature called "Cardio Age" found on a smart ring (the Ultrahuman Ring). This feature tries to tell you how "old" your heart and lungs really are, based on how well they are working, rather than just how many years you have been alive.

Here is the breakdown of the study using simple analogies:

1. The Big Idea: The "Heart Age" vs. Your "Birthday Age"

Most of us know our "Birthday Age" (Chronological Age). But your heart might feel younger or older than that.

• The Study: Researchers looked at 442 people who wore these smart rings for a year.
• The Result: On average, these people had a "Heart Age" that was 1.8 years younger than their actual age. About 8 out of 10 people had a "younger" heart.
• Why? The study suggests this is because people who buy fitness rings are usually already health-conscious, like drivers who always keep their cars tuned up.

2. The Mystery: Does the Ring Just "Guess" or Does It "Know"?

The researchers wanted to know: Is this "Heart Age" number just a magic trick based on the data the ring already uses, or does it actually reflect real-life habits?

Think of the algorithm (the computer brain inside the ring) like a chef.

• The Ingredients (Direct Inputs): The chef uses specific ingredients to make the dish: your resting heart rate, how much your heart rate varies (HRV), and your estimated fitness level (VO2 max).
• The Dish (Cardio Age): This is the final meal served to you.

The worry was: If the chef uses "sleep" as an ingredient, of course the dish tastes like sleep. But the researchers wanted to see if the dish tasted like sleep even if "sleep" wasn't explicitly listed as an ingredient.

The Discovery: They found that even though the ring didn't directly feed "sleep" or "steps" into the Heart Age formula, people with better sleep and more steps still had younger Heart Ages.

• The Analogy: It's like tasting a soup and saying, "This tastes like fresh basil," even though you didn't see the chef add basil. It means the soup (Heart Age) is capturing the essence of a healthy lifestyle, not just the raw data.

3. The Lifestyle Clues (What Makes Your Heart Feel Younger?)

The study found four main "life hacks" that were linked to having a younger Heart Age:

1. Sleep Efficiency: Waking up less often during the night. (Think of it as a car engine that runs smoothly without sputtering).
2. REM Sleep: The deep, dreamy part of sleep. People with younger hearts got about 22 more minutes of this per night.
3. Total Sleep Time: Sleeping longer overall.
4. Daily Steps: Walking more.

The "Extreme" Comparison:
The researchers compared the "Super Fit" group (youngest hearts) to the "Struggling" group (oldest hearts).

• The Super Fit group slept 37 minutes longer per night.
• They got 22 more minutes of deep REM sleep.
• They took 1,600 more steps per day.
• They also had a lower Body Mass Index (BMI).

4. The Weight Connection

There was a clear "dose-response" relationship with weight.

• Underweight/Normal: Their hearts felt 3.7 years younger than their birthday age.
• Obese: Their hearts felt only 0.5 years younger.
• Metaphor: Imagine carrying a heavy backpack. The heavier the backpack (weight), the harder your heart has to work, making it feel "older" and more tired.

5. The Time Machine: Can You Actually Get Younger?

The most exciting part of the study is that it wasn't just a snapshot; it watched people over 12 months.

• They found a group of people who consistently improved their habits.
• The Result: Over a year, these people didn't just feel better; their "Cardio Age" actually dropped by 3.2 years.
• The Proof: As their Heart Age dropped, their resting heart rate slowed down (like a car idling smoothly) and their fitness capacity went up.

This proves that the "Cardio Age" number isn't static. It's a living dashboard. If you change your lifestyle, the number changes with you.

Why Does This Matter?

Imagine trying to lose weight. If you just look at a scale, you see a number. If you look at a chart of your heart rate, it's just a line. But if you see a number that says, "Your heart is now 3 years younger than it was last year," that is a powerful, easy-to-understand motivator.

In Summary:
This study shows that a simple number on a smart ring can accurately reflect your real-life habits (sleep, walking, weight). It's not just a computer glitch; it's a mirror showing that when you sleep better and move more, your heart literally feels and functions like it's younger. It turns complex medical data into a simple "age" that anyone can understand and use to improve their health.

Technical Summary

1. Problem Statement

Cardiovascular age (or "heart age") is a valuable tool for communicating complex physiological risk factors into an intuitive metric. However, traditional estimates rely on clinical exercise testing or laboratory-grade $VO_2$ max measurements, limiting scalability and preventing continuous monitoring. While consumer wearables can now estimate cardiorespiratory fitness (CRF) and derive a "fitness age" using photoplethysmography (PPG) heart rate data, two critical gaps remain:

1. Validity of Lifestyle Correlates: It is unclear whether these wearable-derived estimates capture substantive, independent lifestyle variations (e.g., sleep quality, physical activity) or merely reflect algorithmic artifacts.
2. Algorithmic Circularity: Many metrics correlated with fitness age (e.g., resting heart rate, BMI) are direct or indirect inputs to the algorithm itself. Studies often fail to distinguish between "constituent" metrics (inputs) and "independent" metrics (outputs), leading to tautological associations.
3. Longitudinal Tracking: No prior study has examined whether within-person changes in wearable-derived cardiovascular age track genuine physiological improvements over time.

2. Methodology

• Study Design: Retrospective observational cohort study.
• Population: 442 adult users of the Ultrahuman Ring AIR (median age 31, 65.4% female).
• Observation Window: 12 months (March 1, 2025 – February 28, 2026).
• Primary Metric: Cardio Age Gap (CA gap), defined as the mean estimated Cardio Age minus chronological age.
• Algorithmic Transparency: The authors explicitly categorized metrics into three tiers to avoid circularity:
  • Direct Constituents (Inputs): Nighttime resting heart rate, HRV (RMSSD), and estimated $VO_2$ max.
  • Indirect Constituents: Recovery score, stress rhythm score, BMI, and weight (partly mediated through the algorithm).
  • Independent Lifestyle Correlates: Sleep duration, sleep efficiency, REM sleep, deep sleep, daily steps, active hours, and movement index (no direct algorithmic link).
• Analytical Approach:
  • Extreme Group Analysis: Compared the "Youngest Hearts" (bottom 10th percentile of CA gap) vs. "Oldest Hearts" (top 10th percentile) using Mann-Whitney U tests.
  • Correlation Analysis: Spearman rank correlations across the full cohort, with Bonferroni correction for multiple testing.
  • Longitudinal Trajectory: Classified users into "Sustained Improvers" and "Worseners" based on the change in CA gap between the first and last 30 days of the 12-month window (threshold: $|\Delta| > 2.0$ years).

3. Key Contributions

• Decoupling Inputs from Outputs: The study provides a rigorous framework for separating algorithmic inputs from independent lifestyle factors, proving that wearable-derived fitness age captures variation beyond its constituent inputs.
• Longitudinal Validation: It demonstrates that within-person changes in Cardio Age correspond to measurable physiological shifts (decreased resting heart rate, increased $VO_2$ max) over a 12-month period without structured intervention.
• Sleep and Activity Correlates: It identifies specific sleep architecture metrics (REM, efficiency, duration) and daily steps as significant independent predictors of cardiovascular fitness age.

4. Key Results

• Distribution: The mean CA gap was −1.84 ± 2.97 years, with 82.6% of participants having a "younger" estimated cardiovascular age. The distribution was left-skewed, likely due to self-selection bias among health-conscious ring users.
• Independent Lifestyle Correlates:
  • Significant negative correlations were found between CA gap and REM sleep ($r = -0.203$), sleep duration ($r = -0.200$), sleep efficiency ($r = -0.194$), and daily steps ($r = -0.145$).
  • These associations survived Bonferroni correction ($p < 0.004$).
  • Extreme Group Comparison: Users with the youngest hearts slept 37 minutes longer, achieved 22 more minutes of REM sleep, and had 1.8% higher sleep efficiency than those with the oldest hearts.
• BMI Dose-Response: A monotonic relationship was observed where higher BMI correlated with older cardiovascular age. Underweight participants had a mean CA gap of −3.73 years, compared to −0.52 years for obese participants.
• Longitudinal Stability & Change:
  • Within-person CA gap was highly consistent over 12 months (median SD = 1.21 years).
  • Sustained Improvers (N=52) showed a mean CA reduction of 3.24 years, accompanied by a −0.8 bpm decrease in resting heart rate and a +1.3 mL/kg/min increase in estimated $VO_2$ max.
  • Sustained Worseners (N=59) showed a mean CA increase of 3.49 years with opposite physiological trends.
  • Divergence between improver and worsener groups became visible by months 3–4.

5. Significance and Implications

• Scientific Credibility: The study validates that consumer wearable fitness age metrics are not merely algorithmic artifacts but reflect genuine physiological and lifestyle variations.
• Actionable Feedback: By translating abstract physiological data (e.g., $VO_2$ max, HRV) into a single "age" number, the metric makes health data actionable. Users can intuitively understand that a 3-year reduction in cardiovascular age is a tangible goal.
• Behavioral Motivation: The data suggests that passive, daily feedback on fitness age may sustain user engagement. The "sustained improver" cohort achieved significant physiological improvements without structured intervention, implying that the metric itself acts as a motivator.
• Clinical Utility: While not a diagnostic tool, Cardio Age serves as a practical, continuous monitoring tool for cardiovascular conditioning, bridging the gap between episodic clinical assessments and daily wellness tracking.
• Limitations: The study acknowledges algorithmic circularity (though explicitly managed), self-selection bias in the cohort, and the lack of external validation against laboratory-grade $VO_2$ max measurements.

In conclusion, this paper establishes that wearable-derived Cardio Age is a robust, longitudinal metric that effectively tracks physiological adaptation to lifestyle changes, particularly sleep quality and physical activity, offering a new paradigm for population-level cardiovascular health monitoring.