SPECIALTY TOPICS · GUIDE

Wearables and Health Trackers: What They Measure and How Accurate They Are

What consumer wearables measure directly, what they estimate and what they calculate, how accurate studies found them, and why accuracy is not proof of better health.

Content last updated October 22, 2026. This page has not been medically reviewed.

Back to Biohacking

Quick answer

Consumer wearables are reasonably good at some measures, such as heart rate and steps at moderate effort, and poor at others, such as calories burned. They are not diagnostic tools. A device can be accurate and still not make anyone healthier, and those are two separate questions. This page labels evidence as established, promising but limited, mixed, insufficient evidence or marketing claim; the labels are explained in How We Label Evidence in Biohacking.

Direct, estimated and calculated

Three kinds of numbers appear in a wearable app, and they deserve different levels of trust.

Measured directly: a signal the sensor actually picks up, such as the pulse seen by light on the skin (photoplethysmography, or PPG) or motion from an accelerometer.
Estimated: a number worked out from that signal with a model, such as calories burned, sleep stages or aerobic fitness.
Calculated by the company: a proprietary score such as "readiness" or "recovery". Its formula is usually not public, and it has no universal clinical meaning.

What studies found

Measure	What studies found	Label
Heart rate (wrist)	A review of 65 articles found the Apple Watch had an error under 10% in 2 studies [1]; a meta-analysis of 56 studies found no heart-rate subgroup above a 10% error threshold for that watch [2]	promising but limited (devices studied only)
Steps	For the Fitbit Charge family, error was under 25% across 20 studies [1]; Fitbit devices met a set accuracy standard about half the time [3]	mixed
Calories burned	Error above 30% for all brands in one review [1]; accuracy depended on activity type [4]	no good evidence of accuracy
Sleep versus wake	Sensitivity of at least 95% for three devices in a 35-person lab night [5]	promising but limited
Sleep stages	Sensitivity 50% to 86% [5]; agreement with a sleep lab varied widely across 11 trackers [6]	mixed
Heart-rate variability	Small average difference from electrocardiogram (ECG) but very uneven across studies [7]	promising but limited

Only about 11% of consumer wearables released have been validated for at least one outcome, and the validation studies done amount to roughly 3.5% of what a full evaluation would need, according to a living umbrella review. [8] Results for one model do not carry over to another, or from an older model to a newer one.

Accuracy is not benefit

Measuring correctly is different from improving health. In a meta-analysis of randomized trials in healthy adults, wearable trackers were associated with modest short-term increases in physical activity, with no clear benefit for weight reduction in subgroup analyses and unknown long-term effect. [9] A separate meta-analysis of 19 trials reported moderate effects on body weight and waist circumference, larger for people with obesity or chronic disease and for use of 12 weeks or more. [10] The two reviews differ, so the weight-loss evidence is mixed. Both are about using a tracker as a motivational tool, not about the accuracy of any one number.

Blood sugar, illness and other big claims

The FDA says not to buy or use smartwatches or smart rings that claim to measure blood glucose, and that it has not authorized, cleared or approved any smartwatch or smart ring intended to measure or estimate blood glucose values on its own. [11] Wearable heart-rate variability is only an exploratory measure for inflammation, according to a 2026 systematic review. [12]

In the United States, the FDA's general wellness guidance (January 2026) describes a compliance policy for low-risk products that promote a healthy lifestyle. [13] Clearance of one feature for one use is not approval of everything a device claims.

Who should be careful

A 2026 neurology review notes that heightened anxiety is an unintended consequence of wearables seen in people with neurologic conditions. [14] Clinicians have described patients chasing perfect sleep data. [15] If tracking feeds worry, perfectionism or disordered eating, take a break and talk with a clinician; see Orthosomnia and Sleep-Tracking Anxiety. People with heart rhythm problems or diabetes should rely on medical-grade devices chosen with their care team.

Where to go next

Device types: Smartwatches, Fitness Bands and Activity Trackers, Smart Rings, Chest-Strap Heart-Rate Monitors and Smart Scales.
Method: Device vs Clinical Reference and How to Evaluate a Biohack.
Sleep context: Wearables vs Sleep Studies.

Frequently asked questions

Are smartwatches accurate for heart rate?

In the studies reviewed, one watch had low heart-rate error at rest and moderate effort, with larger error at higher running speeds. [1] [16] This does not apply to every model.

Can my watch count calories burned accurately?

Not well. Error was large across devices and varied by activity. [1] [4]

Can a smart ring or watch measure blood sugar?

The FDA warns against devices that claim to, and has not authorized any that do so on its own. [11]

Should I trust my sleep stages?

Treat them as rough estimates. Sleep versus wake was detected well in one lab study, but stage sensitivity varied from 50% to 86%. [5]

Does wearing a tracker make people healthier?

Trials show modest short-term increases in activity in healthy adults, but long-term benefit is not established. [9]

References

Education, not medical advice. This page is for general information. It does not diagnose or treat any condition and is not a substitute for care from a qualified health professional. Natural does not automatically mean safe. Ask a clinician or pharmacist before changing your care, especially if you take medication, are pregnant or breastfeeding, or have a health condition.