Wearables & CGM.

The best-validated consumer device for sleep staging vs polysomnography. De Zambotti 2019 (PMID PubMed ID: the permanent number the US National Library of Medicine gives a published paper. Each one links to that paper's record on PubMed. 28323455) reported 79% epoch-by-epoch agreement vs PSG for total sleep time, the highest published number for a consumer wearable. Continuous skin temperature trend is uniquely useful for tracking circadian alignment, illness onset, and menstrual cycle phase. Heart-rate-derived HRV is solid; not as accurate as a chest strap.

• Best for: sleep quality, recovery readiness, circadian rhythm.
• Subscription: required ($5.99/mo) to unlock most features.
• Limitation: not real-time HRV during workouts; not a fitness tracker.

Strongest overnight HRV data of any consumer wearable, uses RMSSD with proper measurement window. Three claims get bundled together here and only two are validated. The wrist sensor has been assessed for heart rate and HRV against reference measures (PMID 34065516), with accuracy varying by movement during resistance exercise (PMID 36803578). Sleep measurement was validated against polysomnography (Miller 2020, PMID 32713257). The Strain and Recovery scores are proprietary algorithms and no independent validation of either is cited here; WHOOP describes the 0 to 21 strain scale as inspired by the Borg perceived-exertion scale, which is not the same as validated against it. No screen forces behavioral focus on the data. Subscription model means you don't own the device. Sleep-staging accuracy lags Oura.

• Best for: daily HRV trending, athletes, training load.
• Subscription: required ($30/mo): includes the strap.
• Limitation: no SpO₂ during sleep (was removed in 4.0); no screen.

FDA-cleared The FDA found a device substantially equivalent to one already on the market. Clearance is not approval and does not require trials showing benefit. single-lead ECG and AFib history are unique to Apple. Single-night SpO₂ readings, cardio fitness (VO₂max The most oxygen the body can use during all-out exercise, a standard measure of aerobic fitness. estimate) has been tested against laboratory measurement and does not agree with it: in the only validation study located (Caserman, JMIR Biomedical Engineering 2024, n=19), the Apple Watch significantly underestimated VO₂max against a metabolic gas analyser (41.4 vs 45.9 mL/kg/min, P=.01) with an intraclass correlation of 0.47, which is poor agreement. It overestimates in unfit people and underestimates in fit ones. The test used a cycle ergometer, not a treadmill. Treat the number as a trend line for yourself over time, not as a VO₂max. PMID 39083800 → Sleep tracking is the weakest of the three. The single highest-impact longevity feature is the AFib detection, which has rerouted clinical care for thousands of users.

• Best for: AFib detection, ECG, fall detection, cardio fitness estimate.
• Subscription: none.
• Limitation: daily battery; HRV measurement is sparse (single morning reading).

Continuous glucose monitoring in non-diabetic users now has FDA-cleared OTC sensors (Dexcom Stelo, Abbott Lingo). For longevity, the value isn't in chasing flat lines, it's identifying postprandial spikes > 160 mg/dL and nocturnal hypoglycemia that fasting glucose and A1c miss entirely. Most actionable wearable on the list for users with metabolic risk factors.

• Best for: postprandial spikes, meal-stack response, sleep glucose dips.
• Caution: use for 2–4 weeks of pattern detection, not perpetually.
• Limitation: 5–10 mg/dL accuracy gap vs venous; trends > absolute values.