Wearables & Testing Decision Guide
What wearables, biological age tests and testing subscriptions can and cannot tell you, based on published validation studies. We list no prices or model numbers: they change too often to keep accurate, so check each maker's site for current models.
What wearables are good for: counting steps, detecting sleep and tracking overnight heart rate over time. They are weak on calories burned and sleep stages,[1,2] and many key health measures, such as blood pressure, ApoB and HbA1c, still need a cuff or a lab test.
| Measure | What the studies found |
|---|---|
| Steps | Accurate in laboratory tests for Fitbit, Apple Watch and Samsung.[1] |
| Heart rate | Varies by brand: Apple Watch and Garmin were most accurate, and Fitbit tended to underestimate.[1] |
| Calories burned | No brand was accurate.[1] |
| Overnight resting heart rate | Against an ECG over 536 nights, Oura rings agreed most closely (concordance 0.97 to 0.98) and WHOOP moderately (0.91).[3] |
| Overnight HRV | Oura rings agreed most closely with the ECG (0.97 to 0.99), WHOOP moderately (0.94), and the Garmin and Polar models less well (0.87 and 0.82).[3] |
| Sleep versus wake | Sleep was detected well (sensitivity at least 0.93), wake poorly to moderately (specificity 0.18 to 0.54).[2] |
| Sleep stages | Inconsistent across devices.[2] |
| VO₂ max estimate | Estimates from resting data overestimated VO₂ max by 2.17 ml/kg/min on average, with wide limits (about −13 to +17); estimates from exercise data had smaller errors.[4] |
| Irregular-pulse alerts | In the Apple Heart Study, 0.52% of 419,297 participants received a notification of an irregular pulse.[5] |
Epigenetic clocks read DNA methylation from a blood or saliva sample. Different clocks answer different questions: GrimAge was built to predict time to death,[7] and DunedinPACE estimates how fast you are aging, distilled from two decades of decline in 19 measures of organ function.[8]
Results can move for technical reasons alone. In replicate samples, six prominent clocks differed by up to 9 years; versions rebuilt on principal components brought most replicates within 1.5 years.[9] Ask which version a lab uses.
Clocks can respond to interventions, but differently. In the CALERIE trial, two years of calorie restriction slowed DunedinPACE without significantly changing PhenoAge or GrimAge.[10] No clock is yet validated as a treatment target.
Glycan tests measure sugar chains on antibodies. In 5,117 Europeans, three glycans together explained up to 58% of the variation in chronological age.[11]
Questionnaire estimates, including the lifestyle estimate in our Bio Age Calculator, are not validated against health outcomes. The same tool includes a PhenoAge calculator that uses nine routine blood tests.
If you test: use the same lab and the same clock each time, and treat any single result as noisy.[9]
1. Which tests are included, and do guidelines support them for someone like you? Our Blood Panel Builder sorts tests by how strongly guidelines support them.
2. Does a clinician review your results and arrange follow-up? A dashboard is not the same as care.
3. Does the report use guideline categories, or "optimal" ranges that no guideline defines? Our Lab Results Interpreter uses cited guideline categories.
4. What happens after an abnormal result? Find out who orders follow-up tests and who pays for them.
Start free: build a panel with the Blood Panel Builder, read results with the Lab Results Interpreter, and use the Test Guide to decide what to test next. Direct-to-consumer lab ordering is available in many US states; rules vary.
References
- 1Fuller D, et al. "Reliability and Validity of Commercially Available Wearable Devices for Measuring Steps, Energy Expenditure, and Heart Rate: Systematic Review." JMIR Mhealth Uhealth. 2020;8(9):e18694. PubMed · DOI
- 2Chinoy ED, et al. "Performance of seven consumer sleep-tracking devices compared with polysomnography." Sleep. 2021;44(5). PubMed · DOI
- 3Dial MB, et al. "Validation of nocturnal resting heart rate and heart rate variability in consumer wearables." Physiol Rep. 2025;13(16):e70527. PubMed · DOI
- 4Molina-Garcia P, et al. "Validity of Estimating the Maximal Oxygen Consumption by Consumer Wearables: A Systematic Review with Meta-analysis and Expert Statement of the INTERLIVE Network." Sports Med. 2022;52(7):1577-1597. PubMed · DOI
- 5Perez MV, et al. "Large-Scale Assessment of a Smartwatch to Identify Atrial Fibrillation." N Engl J Med. 2019;381(20):1909-1917. PubMed · DOI
- 6Shah VN, et al. "Continuous Glucose Monitoring Profiles in Healthy Nondiabetic Participants: A Multicenter Prospective Study." J Clin Endocrinol Metab. 2019;104(10):4356-4364. PubMed · DOI
- 7Lu AT, et al. "DNA methylation GrimAge strongly predicts lifespan and healthspan." Aging (Albany NY). 2019;11(2):303-327. PubMed · DOI
- 8Belsky DW, et al. "DunedinPACE, a DNA methylation biomarker of the pace of aging." Elife. 2022;11. PubMed · DOI
- 9Higgins-Chen AT, et al. "A computational solution for bolstering reliability of epigenetic clocks: Implications for clinical trials and longitudinal tracking." Nat Aging. 2022;2(7):644-661. PubMed · DOI
- 10Waziry R, et al. "Effect of long-term caloric restriction on DNA methylation measures of biological aging in healthy adults from the CALERIE trial." Nat Aging. 2023;3(3):248-257. PubMed · DOI
- 11Krištić J, et al. "Glycans are a novel biomarker of chronological and biological ages." J Gerontol A Biol Sci Med Sci. 2014;69(7):779-89. PubMed · DOI