Wearables & Home Vitals, Ranked by What They Measure

Across prospective cohorts and meta-analyses, home/self-measured blood pressure predicts cardiovascular events and mortality at least as well as, and often independently of, office measurement. When paired with support such as clinician-guided medication…

What the evidence actually says: Trials and an individual-patient meta-analysis show home monitoring lowers blood pressure only when paired with clinician-guided titration or similar support (about 3.5-6 mmHg systolic at 12 months); the device alone does little. Accuracy depends on buying an independently validated monitor and using the right cuff and technique, and most monitors sold online are unvalidated.

Full evidence page → · Check price on Amazon ↗ · More Air, Water & Sleep Systems · Blood pressure

Across large prospective cohorts and meta-analyses, lower grip strength is consistently associated with higher all-cause and cardiovascular mortality, incident cardiovascular disease, disability, and, in some studies, cognitive decline and depression, often…

Dose used in studies: Three squeezes per hand, seated, elbow at 90 degrees, best value recorded - and the same chair, same arm position, same time of day every time. Monthly is plenty. What matters is your own trend over years, not where you sit against a population chart.

What the evidence actually says: Grip strength is one of the most durable prognostic markers in epidemiology: in PURE across 17 countries and in half a million UK Biobank participants, lower grip tracked all-cause and cardiovascular mortality, and it did so after adjustment and across very different populations. This specific device also clears the bar that most consumer instruments fail - the Camry digital dynamometer has been tested head-to-head against the Jamar, the reference instrument, with excellent reliability and correlations of 0.97-0.99, so a cheap unit is genuinely measuring the thing. Two honest caveats - grip strength is a marker of overall muscle and health rather than a target in itself, and squeezing a dynamometer more often will not move it, since the intervention that moves it is resistance training, which is graded separately and is where the actual benefit lives. The second is that this is the site's second grip dynamometer: it measures exactly what the other one measures, the other is graded strong, and nothing in the evidence says a 200 kg-capacity unit tells a middle-aged adult anything a 90 kg one does not - buy one, and let price decide which.

Full evidence page → · Check price on Amazon ↗ · More Diagnostics & Epigenetic Tests · Grip strength · Blood pressure

The biomarkers a body-composition scale tracks are each linked to healthy aging: higher body-fat percentage and visceral adiposity are associated with greater cardiovascular and metabolic risk and, in general adult populations, higher all-cause mortality,…

Dose used in studies: Weigh at the same time each morning, after voiding and before eating, and read the trend line over weeks rather than any single day. Use the weight; treat the body-fat, segmental and vascular readings as rough estimates that drift with hydration.

What the evidence actually says: The FDA clearance on this scale is narrow and specific: it covers detection of atrial fibrillation from the 6-lead ECG, and nothing else. The bioimpedance body composition is the part most people buy it for, and it is the part that does not hold up at an individual level: against DXA, bioimpedance underestimates body-fat percentage by roughly 5 to 6 percentage points and overestimates fat-free mass by around 4 kg, with limits of agreement wide enough that published comparisons conclude the two methods are not interchangeable for one person, only for group averages. I could not find any published validation of the vascular-age or nerve-health scores against a reference standard. Self-weighing does genuinely help, adding about 1.7 kg of weight loss when built into a behavioural programme, but that benefit comes from stepping on a scale regularly and works on a $25 one; no trial shows that watching your body-fat percentage or vascular age changes any outcome.

Full evidence page → · Check price on Amazon ↗ · More Biometric Wearables & Sensors · Fiber

Large prospective cohorts and pooled analyses of more than 10,000 people show that ambulatory blood pressure, especially the nighttime and 24-hour average, predicts cardiovascular events and all-cause and cardiovascular mortality more strongly than clinic…

What the evidence actually says: Large international cohorts and a 2025 patient-level meta-analysis consistently show 24-hour and nighttime ambulatory readings predict death and cardiovascular events better than office measurements, and US guidelines recommend confirming hypertension out-of-office. No randomized trial has shown that ABPM-guided treatment itself improves outcomes — the value is in better management, and one influential 2018 cohort study in this field was retracted in 2020. Consumer devices should be clinically validated with a properly fitted cuff.

Full evidence page → · Check price on Amazon ↗ · More Air, Water & Sleep Systems · Blood pressure

Across brands, independent validations find these devices agree closely with polysomnography for sleep-versus-wake detection and with ECG for heart rate and time-domain HRV (RMSSD), while four-stage sleep staging is only moderate (roughly 50-75% agreement)…

Dose used in studies: Wear it nightly for at least two weeks before reading anything into a number; single nights are noise. The defensible outputs are sleep duration, nocturnal resting heart rate and the HRV trend. Readiness is a composite, not a measurement.

What the evidence actually says: For duration and cardiac metrics this is the best-validated device on this list: against an ECG reference across 536 nights the Gen 4 gave nocturnal resting heart rate within about 2% and HRV within about 6%, the best of five wearables tested, and against multi-night polysomnography the Gen 3 did not differ significantly from PSG for total sleep time, wake after sleep onset, light sleep or deep sleep. Sleep staging is still the soft spot everywhere, including here: per-stage sensitivity ran about 76-80% in a laboratory comparison, wake-detection specificity was 0.41 in a week of unrestricted home sleep, and the most favourable of those studies has an author who sits on Oura's medical advisory board. Readiness and Cardiovascular Age are proprietary composites with no published validation against any reference standard and no FDA clearance for any feature. Nobody has shown that acting on a Readiness score changes sleep, fitness or illness risk, and for a subset of people nightly scoring makes sleep worse rather than better - the pattern sleep physicians named orthosomnia.

Full evidence page → · Check price on Amazon ↗ · More Biometric Wearables & Sensors · HRV · Sleep

Across brands, independent validations find these devices agree closely with polysomnography for sleep-versus-wake detection and with ECG for heart rate and time-domain HRV (RMSSD), while four-stage sleep staging is only moderate (roughly 50-75% agreement)…

Dose used in studies: Wear it continuously and read Recovery as a rolling trend across weeks. If you are going to act on HRV, act on the direction of a multi-week average, not on a single morning's score.

What the evidence actually says: WHOOP is good at the thing it does not advertise and weaker at the thing it does. Against polysomnography it got total sleep time almost exactly right - an 8-minute non-significant overestimate, and in a later systematic review the smallest total-sleep-time disagreement of the devices compared - but its specificity for detecting wake was only 51%, four-stage agreement was 64% with a kappa of 0.47, and the same review found it had the largest REM disagreement of the group at 21 minutes. Be precise about the regulatory picture: the FDA-cleared ECG feature sits on the more expensive WHOOP MG, not on the 5.0, and Blood Pressure Insights has never been FDA-cleared - the FDA issued a warning letter over it in July 2025 and closed that letter in June 2026 after WHOOP changed the display, leaving the feature as an unregulated wellness output rather than a validated blood pressure measurement. Strain and Recovery are proprietary composites with no external validation; HRV-guided training has modest meta-analytic support as a training principle in endurance athletes, but that is not evidence for WHOOP's particular score, and this is the only device on the list that also requires an ongoing subscription.

Full evidence page → · Check price on Amazon ↗ · More Biometric Wearables & Sensors · HRV · Sleep

Across brands, independent validations find these devices agree closely with polysomnography for sleep-versus-wake detection and with ECG for heart rate and time-domain HRV (RMSSD), while four-stage sleep staging is only moderate (roughly 50-75% agreement)…

Dose used in studies: If the cardiac features are why you are buying it, turn on irregular rhythm notifications and AFib History and wear it overnight. A notification is a reason to see a clinician, not a diagnosis; the ECG app only tells you about the moment you held your finger on the crown.

What the evidence actually says: This is the most rigorously evaluated device on the list, and the regulatory picture is real: the ECG app, irregular rhythm notifications, AFib History, sleep apnea notifications and hypertension notifications are all FDA-cleared. Cleared means the feature performs as described, not that using it makes you healthier - in the Apple Heart Study only 0.52% of 419,297 people were ever notified, and among those who returned a usable ECG patch just 34% actually had atrial fibrillation, while a head-to-head against physician-read 12-lead ECG put the watch at 85% sensitivity and 75% specificity with 18% of tracings inconclusive. Sleep staging is far weaker than sleep duration: against polysomnography the watch underestimated deep sleep by 43 minutes and overestimated light sleep by 45. The USPSTF still finds the evidence insufficient to say that screening asymptomatic adults for atrial fibrillation improves outcomes, and among people already diagnosed with AF, wearable users had more testing, more procedures and more symptom preoccupation than matched non-users - which is the honest reason not to buy this to reassure yourself.

Full evidence page → · Check price on Amazon ↗ · More Biometric Wearables & Sensors · HRV · Sleep

Across brands, independent validations find these devices agree closely with polysomnography for sleep-versus-wake detection and with ECG for heart rate and time-domain HRV (RMSSD), while four-stage sleep staging is only moderate (roughly 50-75% agreement)…

Dose used in studies: Use it for pace, distance and steps; pair a chest strap for interval and threshold work, because wrist optical heart rate lags at high intensity. Treat the VO2max estimate as a rough training marker and re-check it against an actual field test rather than reading it as a fitness score.

What the evidence actually says: Garmin's optical heart rate is among the better performers at rest and steady state and its GPS-derived pace and distance hold up, but no brand measures energy expenditure accurately, and a Garmin watch running the same submaximal-run algorithm underestimated laboratory gas-analysis VO2max by about 4.7 mL/kg/min, with the error roughly twice as large in the fittest runners. Sleep is the weak point: Garmin devices scored worse than research actigraphy on sleep/wake detection, and against polysomnography a Garmin multisport watch overestimated total sleep time by roughly 85 minutes and significantly misassigned sleep stages, overcalling light sleep and undercalling deep. Worth being blunt about: the Forerunner 265 does not have Garmin's FDA-cleared ECG app, which requires hardware only present on certain other models, so this watch has no rhythm-detection feature at all. The metric here with genuinely strong outcome evidence is steps, and that evidence is about steps rather than about any particular watch.

Full evidence page → · Check price on Amazon ↗ · More Biometric Wearables & Sensors · HRV · Sleep

Across brands, independent validations find these devices agree closely with polysomnography for sleep-versus-wake detection and with ECG for heart rate and time-domain HRV (RMSSD), while four-stage sleep staging is only moderate (roughly 50-75% agreement)…

Dose used in studies: Use it for steps, pace and everyday heart rate; add a chest strap if you train by heart-rate zones. Read the sleep score as a duration estimate and ignore the stage breakdown.

What the evidence actually says: This is the same measurement platform as the more expensive Garmins with fewer running features, and it inherits the same profile: steps are accurate enough to act on, heart rate is good at rest and steady state and degrades with wrist motion, and energy expenditure is not accurate in any consumer device from any brand. Sleep is where it is weakest - a systematic review of a comparable Garmin against polysomnography found only moderate accuracy for sleep stages and total sleep time, and in a laboratory comparison Garmin devices detected wake worse than research-grade actigraphy did. Like the Forerunner 265, the Vivoactive 6 does not carry Garmin's FDA-cleared ECG app, so there is no rhythm-detection function here regardless of what the heart-rate alerts feel like. The one number on this watch with strong outcome evidence behind it is the step count, where mortality falls steeply between about 2,000 and 7,000 steps a day - and a $90 tracker counts those steps about as well.

Full evidence page → · Check price on Amazon ↗ · More Biometric Wearables & Sensors · HRV · Sleep

Across brands, independent validations find these devices agree closely with polysomnography for sleep-versus-wake detection and with ECG for heart rate and time-domain HRV (RMSSD), while four-stage sleep staging is only moderate (roughly 50-75% agreement)…

Dose used in studies: Leave irregular rhythm notifications on and wear it overnight for the sleep and resting heart rate trend. If you get a notification, use the ECG app to capture a tracing at the time, and take that to a clinician - it is a prompt to be checked, not an answer.

What the evidence actually says: Two features here are genuinely FDA-cleared: the ECG app, cleared in 2020, and the photoplethysmography-based irregular rhythm notifications, cleared in 2022. The Fitbit Heart Study is the honest reading of the second - of 455,699 participants only 1% ever received a detection, and among those who wore a confirmatory ECG patch, 32% had atrial fibrillation on it; separately, head-to-head against a physician-read 12-lead ECG the Sense's ECG app was 66% sensitive and 79% specific with 21% of tracings inconclusive, the weakest of the smartwatches in that comparison. Sleep duration is reasonable and staging is not: against polysomnography the Sense 2 overestimated light sleep by 18 minutes and underestimated deep sleep by 15, and the EDA-based stress score has no reference standard to be validated against at all. The USPSTF finds the evidence insufficient to say that screening asymptomatic adults for atrial fibrillation improves outcomes, and people who already have AF and wear these devices end up with more testing and more preoccupation, not demonstrably better health.

Full evidence page → · Check price on Amazon ↗ · More Biometric Wearables & Sensors · HRV · Sleep

Across brands, independent validations find these devices agree closely with polysomnography for sleep-versus-wake detection and with ECG for heart rate and time-domain HRV (RMSSD), while four-stage sleep staging is only moderate (roughly 50-75% agreement)…

Dose used in studies: Wear it every day and use one number: steps. Aim for a personal increase of 1,000-2,000 steps a day over where you currently are, rather than for 10,000, and ignore the calorie estimate entirely.

What the evidence actually says: This is the honest value on the list, and the reason is unflattering to everything above it: for the single output with strong outcome evidence behind it - the step count - a cheap tracker performs about as well as a watch costing eight times more. The step evidence itself is genuinely strong: pooled across 15 cohorts and 47,471 adults, all-cause mortality fell progressively with steps per day up to about 6,000-8,000 in people over 60 and 8,000-10,000 in people under 60, and a 2025 dose-response meta-analysis put 7,000 steps a day at 47% lower all-cause mortality than 2,000. What the device does less well is everything else: energy expenditure is inaccurate, sleep duration is decent but staging is mediocre, and the irregular rhythm notifications run Fitbit's FDA-cleared photoplethysmography algorithm but there is no ECG app on this model to capture a tracing when one fires. And a tracker is not a substitute for the behaviour - in the IDEA trial, adding a wearable to a supervised weight-loss programme produced less weight loss at 24 months than the programme alone, 3.5 kg versus 5.9 kg.

Full evidence page → · Check price on Amazon ↗ · More Biometric Wearables & Sensors · HRV · Sleep

Across brands, independent validations find these devices agree closely with polysomnography for sleep-versus-wake detection and with ECG for heart rate and time-domain HRV (RMSSD), while four-stage sleep staging is only moderate (roughly 50-75% agreement)…

Dose used in studies: Wear it daily and treat the step count as the one output with real outcome evidence behind it. Read sleep stages, Nightly Recharge and training-load numbers as week-to-week trends rather than nightly verdicts, and confirm any heart-rate-driven training decision with a chest strap.

What the evidence actually says: Polar's step counts and sleep duration track research instruments closely enough to be useful, but the systematic review of wrist-worn Polar devices concluded their accuracy remains unclear because the individual studies conflict, and energy-expenditure error is large in all of them. Against polysomnography in sleep-clinic patients a Polar multisport watch overestimated total sleep time by about 45 minutes, overestimated light sleep, underestimated deep sleep, and failed to detect daytime sleep at all; against an ECG reference across 536 nights, Polar had the poorest nocturnal HRV agreement of five wearables tested, with a mean absolute percentage error of 16%. The wrist ECG on the Vantage V3 exists to run Polar's orthostatic and recovery tests and carries no FDA clearance to detect an irregular rhythm, so unlike an Apple Watch, a Fitbit or a Withings device this watch is not a rhythm check. Trackers as a class do get people moving, roughly 1,800 extra steps a day in pooled trials, but that effect belongs to any step counter, and nothing in the literature says the extra several hundred dollars over a basic tracker buys a different outcome.

Full evidence page → · Check price on Amazon ↗ · More Biometric Wearables & Sensors · HRV · Sleep

In randomized trials and meta-analyses, continuous and flash glucose monitoring modestly lowers HbA1c (about 0.2–0.3 percentage points on average) and, more strikingly, cuts time spent in hypoglycemia—by 38% in type 1 diabetes and by roughly half in…

Dose used in studies: Treat it as a two-week experiment, not a permanent feed. Test specific questions — how your usual breakfast compares with a higher-protein version, what a 15-minute post-meal walk does to the curve — then take the sensor off. Do not chase a flat line: post-meal rises are normal physiology, and there is no validated target range for people without diabetes.

What the evidence actually says: Two questions matter for a CGM worn without diabetes, and both answers are uncomfortable. On accuracy: in healthy adolescents the Libre's mean absolute relative difference against fingerstick plasma glucose was 13.1%, with only 68% of paired readings meeting the ISO 15197:2013 criterion, and in children without diabetes it showed magnitude bias at fasting levels and correlated poorly with true fasting glucose (r-squared 0.06) — it is built to track the large swings of diabetes, not to resolve the narrow band a normoglycemic person lives in. On whether measuring changes anything: the meta-analysis of 25 randomized trials of CGM as a behaviour-change tool found a modest HbA1c benefit driven by participants who had diabetes, no significant effect on weight or BMI, and only three trials in people without diabetes. Healthy people already spend roughly 15% of the time in the prediabetic range on CGM, and when 18 expert clinicians were handed non-diabetic CGM reports they could not agree on who needed follow-up — so expect to see spikes, and expect that nobody can currently tell you what they mean.

Full evidence page → · Check price on Amazon ↗ · More Biometric Wearables & Sensors · HbA1c · Glucose

Research on the biomarker this device tracks - blood-oxygen saturation and nocturnal hypoxemia - links greater overnight oxygen desaturation (quantified as the sleep apnea-specific 'hypoxic burden') to higher risks of hypertension, cardiovascular events,…

Dose used in studies: If you use one, sit still, warm your hands, remove nail polish, and take the reading at rest - cold fingers and motion are the commonest causes of a wrong number. Track the trend rather than a single value, and never let a normal reading override how you actually feel.

What the evidence actually says: The accuracy problem with pulse oximetry is a safety problem rather than a nuance. These devices read high in people with darker skin: pooled bias is about 1.1% for high skin pigmentation and 1.5% in people described as Black or African American, and in a meta-analysis of 23 studies the accuracy root-mean-square exceeded the FDA's 3% threshold in every pigmentation group tested. That matters because occult hypoxemia - a reassuring-looking reading sitting over a genuinely low arterial saturation - is about 1.7 times more common in Black than in White hospitalised patients. All of that evidence comes from hospitals, where readings are checked against arterial blood; there is none from home or outpatient settings, and consumer wrist-worn oximeters are sold as wellness products without the clearance a hospital oximeter carries, with no mass-market wrist device meeting the medical standard for pulse oximeters. No trial shows that a well person checking their oxygen saturation at home changes any outcome, and the failure mode worth naming is the one that causes harm: a falsely reassuring number in someone who is genuinely short of breath.

Full evidence page → · Check price on Amazon ↗ · More Air, Water & Sleep Systems · Sleep