Field note
A consumer watch can read heart rate and movement from the wrist fairly well at rest, but it cannot see blood chemistry, genes, hormones, or most illness.
The sensor is looking from the outside in. A small green light watches the pulse under your skin. From that, and from how you move, the watch estimates almost everything else. At rest that picture is often genuinely good. It still never touches what is inside the blood itself.
Resting and overnight numbers are usually the strongest read. When you are still, the sensor sits quietly and the heart-rate signal is cleaner. Wang and colleagues (2017) tested wrist-worn heart-rate monitors and found accuracy that is useful for many everyday uses, with more error when motion rises. That is why resting heart rate and quiet overnight samples are the ones to trust most. During a hard workout the picture gets shakier. Motion and a looser fit jostle the sensor, so an activity reading is more of a guess than a measurement.
Zhang, Shen, and Qi (2016) meta-analyzed resting heart rate against all-cause and cardiovascular mortality in the general population. Higher resting rate tracked with higher risk at the group level. Kodama and colleagues (2009) meta-analyzed cardiorespiratory fitness against mortality and cardiovascular events. Those are directional signals from heart and fitness patterns. They are not a lab panel.
| Signal | Honest read |
|---|---|
| Resting / overnight heart rate | Strongest wrist band when you are still (Wang et al. 2017 context) |
| Fitness estimate from pace + heart rate | Useful direction; not a mask VO2 lab result (Kodama et al. 2009 is about fitness, not wrist math) |
| Sleep stages | Softest guess; duration and regularity hold up better (Schyvens et al. 2025) |
| Skin temp, SpO2, breathing rate | Most honest as change from your own baseline |
How fit your heart and lungs are is one of the most useful things a watch tracks. The watch usually works it out from heart rate and pace. It does not measure the oxygen you actually breathe. Read it as a direction your fitness is heading, not a figure from a lab. What that estimate looks like on a consumer device is in What is VO2max on a watch?
Sleep stages are the softest guess. Schyvens and colleagues (2025) validated six commercial wrist sleep trackers against polysomnography for stage scoring. The watch is better at sleep versus awake than at labeling deep, REM, and light on a single night. Trust total sleep and how regular your nights are before one night's slice of deep sleep. That stage limit is spelled out in Are Apple Watch sleep stages accurate?
Skin temperature, blood oxygen, and breathing rate are most honest as a change from your own baseline, not as a number to compare against a doctor's instrument.
A watch cannot see inside your blood. Cholesterol, blood sugar, inflammation, hormones, and vitamins are invisible to it. So are your genes, what you eat and drink, and the medications you take. Most illness stays hidden too. A great-looking score can sit right on top of something only a blood test or a doctor would catch.
Wearable biological-age style numbers sit in that same honesty gap. They lean on heart, fitness, and sleep signals that have population links. They do not replace a clinical aging test. What those years displays are trying to say is in What is biological age?
The number is a direction to nudge, never a clean bill of health.
I write these because I build Aera, an iPhone app that reads Apple Health against your own baseline and puts the research and the error bars underneath every number it shows you, including this one. You do not need it to use anything above.