What Apple Watch actually gives you

The list is impressive when you sit down and write it. As of watchOS 10 + the current Health app, your Watch collects (or can be made to collect):

  • Cardio fitness — VO2 max estimate updated continuously, with a "low / below average / above average / high" categorical band against your age + sex peer group [1]
  • Resting heart rate — daily, with multi-year trend
  • Heart rate variability (RMSSD) — sporadic samples, not continuous, but enough for a weekly trend
  • Sleep stages (since watchOS 9) — REM / Core / Deep / Awake, classified algorithmically with roughly 60-75% epoch-level agreement vs polysomnography [2]
  • Sleep duration + efficiency + time-in-bed
  • ECG — single-lead, FDA-cleared for atrial fibrillation detection
  • Walking steadiness — fall-risk classification (only meaningful for older adults)
  • Time in daylight — passive ambient-light exposure tracking
  • Mindful minutes — meditation app pulls + manual entry
  • Exercise minutes + workout zones + zone-2 estimates (since watchOS 10)
  • Wrist temperature deviation (Series 8+) — nightly skin-temp shift from personal baseline
  • SpO2 spot checks (Series 6+, when supported in your region)
  • Cycle tracking with temperature-based ovulation estimates
  • Fall detection + crash detection + irregular-rhythm notifications

That's an extraordinary dataset to have collected passively by a single piece of jewellery. For most people, it's more health-relevant signal than they've ever generated in their lives.

What Apple Watch doesn't give you

Read the list above again and notice what's not there:

  • A biological age estimate. No single number that says "your cardiovascular system looks ~5 years younger than chronological" or "your metabolic signal looks older than your peers". The Cardio Fitness category is the closest, and it's deliberately one domain only.
  • A multi-domain rollup. Sleep is in one app section. Cardio fitness is in another. HRV is buried four taps deep. There's no surface that says "your sleep is drifting younger and your cardio is drifting older this month, here's the net effect."
  • A trajectory. The Health app shows the current value and a chart, but doesn't model "your bio-age is moving at X years per chronological year of aging — better or worse than expected".
  • A daily protocol or recommendation. "Today's signal is light on cardio + heavy on sleep debt — the highest-leverage action is a 30-minute zone-2 walk before lunch." Apple won't write that sentence for you.
  • A confidence layer. "The bio-age estimate is high-confidence on cardio + sleep, low-confidence on body composition because you haven't logged weight + body fat in 6 weeks." Apple Health doesn't surface what your model is uncertain about.
  • An AI / coach layer. A chat that knows your bio-age trend, your last 14 days of sleep, your VO2 max trajectory, and can answer "is my HRV drop this week real or noise?" with the actual numbers in context.
  • Cross-modal integration. If you also wear Oura at night for finer sleep tracking, use a Withings scale for body comp, and ride with a Garmin for power-based zone 2, Apple Health is a fine receiver but won't synthesise across all of them.
  • Lab biomarker integration. When you get a lipid panel or fasting insulin or hsCRP test done, Apple Health stores the value but doesn't fold it into a coherent bio-age picture.

Why Apple doesn't ship any of this

Three structural reasons, in descending order of importance:

Regulatory. A single biological-age number could be construed as a medical-device claim under FDA / MDR rules. Apple has been extremely conservative about anything that approaches that line — the ECG took years of FDA clearance work to ship; the SpO2 feature has been on-and-off in regulated markets. A bio-age score that synthesises across 13 domains would trigger an even more complex regulatory review without the obvious commercial upside that justified ECG.

Audience scale. Apple Health serves 1+ billion users. A bio-age score requires opinionated coefficient choices (does sleep get 12% of the weight or 18%? does body fat penalise more than cardio fitness rewards?). Those decisions can't be made for everyone — different evidence weightings suit different populations. Apple's platform strategy is to ship the most-conservative reading (e.g. categorical "below average / above average") and let third-party apps make the opinionated rollups.

Strategy. Apple's positioning is the data + sensor layer. HealthKit exists precisely so third-party apps can build interpretation layers on top. Apple makes the iPhone + Watch hardware revenue + the OS platform stickiness; opinionated health software is third-party turf. This is the same shape as why Apple Maps doesn't try to be the AI travel assistant that Google Maps experiments with — the platform vendor leaves the opinionated layer to apps.

The upshot: Apple Health is unlikely to ever ship a biological-age estimate. Not because they can't, but because it's not where their margins or risk-tolerance live.

What it takes to fill the gap

A useful bio-age estimate on top of Apple Health needs five things the Watch doesn't ship:

1. A published 13-domain model with peer-reviewed coefficients for each. Sleep duration / efficiency / regularity. Cardio fitness (VO2 max + zone-2 minutes). Cardiovascular markers (RHR + HRV + BP). Body composition (BMI + body fat % + waist-hip). Cognitive function (reaction time + working memory). Clinical labs (if uploaded). Mental-health questionnaires (PHQ-9 + GAD-7). Family history. Genetic risk (if uploaded). Each domain needs a defensible threshold table and a citation.

2. A smoothing layer. Wearable data is noisy. A single-night sleep estimate from Apple Watch can be off by 60 minutes vs polysomnography for a given individual. The trend across 7-14 nights is much more reliable. Any bio-age model that runs off raw daily data is too jittery to act on; it needs rolling averages + outlier rejection + freshness weighting. The smoothing math has to be public so users can understand why their score moved (or didn't).

3. Manual-input fields for what the Watch can't measure. Blood pressure (cuff at home or pharmacy). Body fat % (DEXA or smart scale). Lab values (lipid panel + HbA1c + hsCRP + fasting insulin + vitamin D). Family history. Cognitive test results from a structured battery. The app has to make manual entry frictionless without requiring it.

4. A daily protocol layer. "Today's lever, based on what's drifting in your data." Without this, the bio-age number is interesting but doesn't change behaviour. The protocol has to be ranked by which domain is currently drifting (sleep down 12% this week → today's lever is sleep), not a static checklist.

5. An AI doctor / coach with your full context. A chat that knows your bio-age trend, your last 14 days of sleep, your VO2 max trajectory, and can answer "is my HRV drop this week real or noise?" with the actual numbers in context — not generic "drink more water" answers.

Where Thier sits in this picture

Thier is the layer that does the five things above on top of your existing Apple Watch / HealthKit data. The app reads from HealthKit on launch, computes a 13-domain bio-age via a transparent published-coefficient engine (every threshold + paper on the Science page), smooths across 14 days, supports manual entry for the lab + composition fields, ranks today's protocol by which domain is drifting, and ships a longevity-coach chat that knows your full context.

This is the natural way to think about Apple Watch in the longevity stack:

  • Apple Watch + HealthKit = the sensor + storage layer. Unchanged.
  • Apple Health app = the categorical metric viewer. Still useful for "what's my exact VO2 max today" or "show me my ECG."
  • Thier (or an equivalent third-party) = the opinionated bio-age model + daily protocol + AI layer on top. The thing that turns sensor data into "here's what to do today."

The mistake people make is asking Apple Health to be the opinionated layer. It won't be, by structural design choice. The right answer is to stack: keep Apple Watch + Health for what they do brilliantly, layer a bio-age + protocol app on top for what they don't.

What you can do right now without spending anything

Two free wins to extract more from your existing Watch setup before adding any third-party software:

1. Set up the Health app's Highlights + Favorites correctly. Open Health → Browse → tap a metric → Add to Favorites. Build a Favorites strip with: VO2 max, RHR, HRV, Sleep Hours, Walking Heart Rate Average, Cardio Fitness. Open the Health app's Summary tab daily for 30 seconds. That alone gives you the trend signal you weren't getting before. The data was always there.

2. Enable sleep tracking in the Sleep app + set a consistent Sleep Schedule. Most people leave the Sleep app un-set; the data quality is dramatically better with a proper Sleep Schedule than with ad-hoc charge-on / charge-off detection. 15 minutes in Settings → Sleep one evening earns you 18 months of cleaner data.

Once you've extracted the maximum free signal from the Watch, the bio-age + protocol layer is the next-step lever. See the Thier bio-age calculator for a no-signup taste of what 13-domain modelling looks like on top of your data.

The takeaway

Apple Watch is the most powerful consumer health-data collector ever built. Apple Health is a high-quality data-viewer. Neither is, or will be, a bio-age model or a daily-protocol coach — that's a deliberate platform choice, not a technical limit. The bio-age estimate + multi-domain rollup + adaptive protocol + AI doctor layer is third-party turf by design. Stack a transparent bio-age app on top of your existing Watch setup; don't wait for Apple to ship the opinionated layer (they won't).

If you want a 13-domain bio-age estimate built on 100+ peer-reviewed coefficients with every threshold transparent — running on the data your Apple Watch already collects — join the Thier waitlist for first access.

Frequently asked questions

Does Apple Watch estimate bio age?

No. Apple Watch and the Apple Health app collect a lot of biometric signal (resting heart rate, HRV, VO2 max estimate, sleep stages, walking steadiness, cardiac fitness category) but neither produces a single biological-age estimate. The closest Apple ships is the Cardio Fitness category, which is a relative VO2 max bracket vs your age + sex peer group — useful but narrow. Apple has historically resisted shipping single-number health scores.

What does Apple Watch measure better than other wearables?

Apple Watch is the highest-fidelity consumer wearable for ECG, fall detection, irregular rhythm notification, and Cardio Fitness (VO2 max estimation). The ECG sensor is FDA-cleared for atrial fibrillation detection. The Sleep app since watchOS 9 has reasonable sleep-stage classification. Where it lags: continuous nighttime SpO2 (introduced late, less consistent than dedicated rings), readiness scoring (Apple does not ship one), and any prescriptive "what should I do today?" layer.

Why doesn't Apple ship a biological age score?

Three reasons. (1) Regulatory: a single biological-age number could be construed as a medical device claim under FDA/MDR rules; Apple has historically been conservative about anything that crosses that line. (2) Audience: Apple Health serves 1+ billion users; a number that aggregates 13 domains into one score requires opinionated coefficient choices that no consumer-platform vendor wants to defend. (3) Strategy: Apple's positioning is the data layer; third-party apps build the interpretation layer on top. Apple Health's HealthKit API exists precisely so apps like Thier can read the signal and model it.

Is HealthKit data accurate enough to build a bio-age model on?

For trending — yes, very. For single-night absolute precision — partly. Apple Watch sleep-stage classification has roughly 60-75% epoch-level accuracy vs polysomnography (see Chinoy 2021). VO2 max estimates are within 10-15% of treadmill-tested true values for most adults. Resting heart rate and HRV trends are excellent. The right way to use the data is as a smoothed weekly trend across multiple domains — that's the abstraction Thier's bio-age model is built on, not a single-day snapshot.

Do I still need other wearables if I have Apple Watch?

Probably not for the bio-age picture. Apple Watch + HealthKit + manual entry for the lab + body-composition values it can't measure (BP, body fat %, lipid panels) is sufficient input for a strong 13-domain bio-age estimate. Where additional wearables add real signal: Oura for dedicated sleep + temperature tracking, Whoop for recovery-focused strain modelling, Garmin/Polar for serious endurance training. Stack what matches your goals; don't stack for completeness.

References

  1. Mandsager K, Harb S, Cremer P, et al. Association of Cardiorespiratory Fitness With Long-term Mortality Among Adults Undergoing Exercise Treadmill Testing. JAMA Network Open. 2018;1(6):e183605. PubMed
  2. Chinoy ED, Cuellar JA, Huwa KE, et al. Performance of seven consumer sleep-tracking devices compared with polysomnography. Sleep. 2021;44(5):zsaa291. PubMed