Smartwatches and Fitness Trackers: How Meaningful Are Health Data?

Within just a few years, smartwatches and fitness trackers have evolved from simple step counters into versatile health companions. Modern devices continuously record heart rate, sleep, activity, heart rate variability, oxygen saturation, skin temperature and, in some cases, even a single-lead electrocardiogram. At the same time, expectations are growing that these data can help people understand their bodies better and identify health risks at an early stage.

Wearables now provide valuable insights into physiological processes that were previously measured almost exclusively in medical settings. Nevertheless, one principle remains important: not every measurement is equally informative. Many values are based on indirect measurements or algorithmic estimates and should be interpreted accordingly.

The key question is therefore not whether smartwatches are useful in principle. What matters far more is which health data are genuinely reliable and how they can be interpreted meaningfully.

Table of contents

1. Why wearables can measure health at all
2. Which sensors modern smartwatches use
3. Which health data are genuinely informative?
4. Where smartwatches reach their limits
5. Can wearables detect diseases early?
6. What role smartwatches play in longevity
7. When health data do more harm than good
8. How to use health data meaningfully
9. Conclusion
10. Sources

Why wearables can measure health at all

The human body continuously produces measurable signals. Every heartbeat changes blood flow, every movement generates acceleration forces and every breath influences heart rate and oxygen supply. Many of these parameters also change depending on sleep, physical activity, stress or illness.

Modern wearables record these signals using various sensors and analyse them with statistical methods and machine learning. Unlike a single examination in a doctor’s office, they enable continuous observation under everyday conditions. This is their greatest advantage.

A single measurement often has only limited informative value. However, when the same parameters are tracked over weeks or months, individual changes become visible that may provide clues about training, recovery or health-related strain.

Which sensors modern smartwatches use

The reliability of health data depends largely on which sensors are built into the device and how the measured signals are processed.

Optical heart-rate measurement

Most smartwatches measure heart rate using photoplethysmography, or PPG. LEDs shine green or infrared light into the skin. Because the volume of blood in the small vessels changes with every heartbeat, the amount of light returning to the sensor also changes. The watch calculates heart rate from these fluctuations.

Under resting conditions, this method now achieves a high level of accuracy. During intense movement, with a loosely fitted strap or when the hands are cold, however, substantial measurement errors may occur. Smartwatches are therefore most suitable for long-term observation of heart rate rather than medical precision measurements.

Accelerometers

Every smartwatch contains an accelerometer that records movement in several spatial directions. This makes it possible to track steps, activity duration, types of exercise and movement during sleep.

This technology also forms the basis of actigraphy, which has been used in sleep research for many years. It reliably detects whether a person is moving or resting. Interpretation becomes more difficult when quiet wakefulness is incorrectly classified as sleep.

Additional sensors

Depending on the model, additional sensors may include:

• skin temperature
• oxygen saturation
• skin conductance
• GPS
• barometric pressure
• single-lead ECG

Not every one of these sensors provides directly medically relevant information. They often complement other measurements and thereby improve the accuracy of the algorithms.

Which health data are genuinely informative?

Modern wearables record a wide range of parameters. However, their scientific validity differs considerably. While some measurements are now well validated, others should be interpreted much more cautiously.

Heart rate

Resting heart rate is one of the most reliable measurements provided by modern smartwatches. Validation studies show that high-quality devices often differ only slightly from medical reference methods under resting conditions.

However, the development of the value over time is even more important than the absolute number. A resting heart rate that remains elevated over several nights may be associated with sleep deprivation, alcohol consumption, high training load or the onset of infection. Individual measurements, by contrast, do not allow reliable conclusions.

Heart rate variability

Alongside heart rate, heart rate variability, or HRV, has become one of the most important metrics provided by modern wearables. It describes the time differences between two consecutive heartbeats and provides indirect information about the autonomic nervous system.

In simplified terms, HRV reflects the interaction between the sympathetic and parasympathetic nervous systems. Higher variability often indicates that the organism is adapting well to changing demands. If HRV declines substantially over several days, however, this may indicate physical or psychological stress, sleep deprivation, intensive training or an emerging illness.

One important limitation applies to practical interpretation: the personal long-term trend is what matters. Age, sex, genetic factors and training status influence HRV considerably, which means that comparisons with average values are only of limited use.

Sleep data

Sleep tracking is now one of the most widely used functions of modern smartwatches. Wearables measure sleep duration and regular sleeping times particularly reliably.

Determining individual sleep stages is considerably more difficult. Deep sleep and REM sleep can be distinguished reliably only using polysomnography, which measures the brain’s electrical activity, among other signals. Smartwatches do not have access to this information. Instead, they estimate sleep stages using heart rate, movement patterns and other sensor signals.

For everyday use, this means that sleep duration and sleep regularity generally provide much more robust information than the precise distribution of individual sleep stages.

Oxygen saturation

Many smartwatches now also measure blood oxygen saturation, or SpO₂. As with heart rate, photoplethysmography is used. Different wavelengths of light make it possible to estimate the proportion of oxygenated and deoxygenated haemoglobin.

Under optimal conditions, high-quality devices can provide reasonably useful values. In everyday life, however, movement, poor fit, cold skin or reduced circulation can significantly impair accuracy. Smartwatches also measure at the wrist, while medical pulse oximeters are generally placed on the fingertip, where signal quality is often higher.

For healthy people, SpO₂ measurement therefore mainly has a supplementary value. Individual low readings are usually not concerning. Repeated abnormalities or symptoms such as shortness of breath should always be assessed medically.

Activity and energy expenditure

Step count is one of the best-studied measurements provided by modern wearables. During normal walking, high-quality devices generally agree well with reference methods. Larger deviations occur mainly during very slow movement, cycling, strength training or other sports involving little arm movement.

Calorie expenditure calculations are less reliable. Devices combine information such as age, body weight, heart rate and movement data with mathematical models. Because metabolism and energy expenditure vary greatly between individuals, the calculated values can differ substantially from actual calorie consumption.

For everyday use, the calorie display is therefore better regarded as guidance than as an exact measurement. Anyone planning training or weight management should not consider the value in isolation.

The greatest strength of wearables lies not in individual measurements but in long-term trends.Changes over days or weeks are generally more informative than short-term fluctuations.

Where smartwatches reach their limits

The more health data modern smartwatches record, the easier it becomes to assume that they can represent a person’s health status completely. In reality, however, many metrics are based on indirect measurements and statistical models.

A good example is sleep scores or so-called readiness scores. They combine different parameters such as sleep duration, heart rate, HRV and activity into a single figure. Such scores may help place developments in context over time, but their calculations differ considerably between manufacturers.

Because the underlying algorithms are usually not published and are only partly validated scientifically, a particular score has no universally valid medical meaning. A sleep score of 90 is therefore not automatically better than a score of 80 on another device.

Determining individual sleep stages also remains challenging. In a sleep laboratory, classification is based on polysomnography, which records brain waves, eye movements and muscle activity simultaneously. Smartwatches do not have access to this information. Classification into REM sleep, light sleep or deep sleep is therefore based on probability models.

Individual interpretation of measurements is equally important. A low HRV or higher resting heart rate does not necessarily indicate disease. Only clear and persistent changes from the personal baseline may provide a meaningful reason to investigate the causes more closely.

Can wearables detect diseases early?

One of the most exciting developments of recent years is the question of whether wearables can detect disease before the first symptoms appear.

The detection of atrial fibrillation is currently the best-studied application. Large prospective studies such as the Apple Heart Study and the Fitbit Heart Study examined hundreds of thousands of users and showed that smartwatches can identify people at increased risk. The devices do not diagnose the condition itself but record unusual heart rhythms that must subsequently be confirmed through medical testing.

Wearables also became a focus of research during the COVID-19 pandemic. Several studies observed that resting heart rate, HRV or skin temperature could change several days before the first symptoms appeared. These findings demonstrate the potential of continuous measurement but are not sufficient for reliably diagnosing infections.

Researchers are also investigating whether wearables may in future provide indications of metabolic disorders, neurological diseases or chronic inflammatory processes. However, many of these approaches remain at an early stage of research and do not yet have an established role in clinical practice.

What role smartwatches play in longevity

From the perspective of longevity research, the greatest value of wearables lies not in diagnosing individual diseases but in continuously recording health-related behaviours.

Sleep, movement and recovery are among the most important modifiable factors for healthy ageing. Changes in these areas are often reflected in parameters such as sleep duration, activity level, resting heart rate and HRV.

These data do not provide a direct insight into biological ageing processes such as mitochondrial dysfunction or epigenetic changes. However, they can help make behaviours visible that are associated over the long term with several hallmarks of ageing.

Wearables become particularly valuable when they support sustainable changes. Someone who sleeps more regularly, moves more or adjusts training load more effectively to personal recovery is likely to benefit more than someone who merely tries to optimise individual metrics.

When health data do more harm than good

The ability to observe one’s own body around the clock can be motivating. At the same time, it creates a risk of overinterpreting normal daily fluctuations. The more data are available, the greater the temptation to view every change as evidence of a health-related cause.

A well-known example is orthosomnia. The term describes people who increasingly judge their sleep according to smartwatch data and focus heavily on sleep scores or supposedly unfavourable sleep stages. Even though they often feel fit during the day, concern about the measurements creates additional stress, which may in turn impair sleep.

Similar effects can be observed with other health data. Lower HRV after intensive training, a slightly elevated resting heart rate after a short night or a poorer sleep rating after travel are often normal physiological responses. Such changes become more informative only when they persist for several days or occur together with symptoms.

Wearables therefore provide their greatest benefit when they offer guidance rather than perfection. The goal should not be to optimise every measurement but to understand long-term developments better and derive sensible lifestyle changes from them.

How to use health data meaningfully

The scientific value of modern smartwatches lies less in individual readings than in continuous observation of the body. The data become particularly helpful when linked with concrete behaviours.

Instead of focusing exclusively on individual values, it is useful to ask questions such as:

• How does my resting heart rate change after several nights of insufficient sleep?
• Does my HRV recover after a lighter training week?
• Does regular exercise improve my sleep duration or sleep quality?
• Which habits are associated with better recovery over the long term?

Such relationships can often be identified more effectively in everyday life than through individual laboratory examinations because wearables document the personal course over weeks or months.

It is equally important to accept the limitations of the technology. Smartwatches provide health information, not medical diagnoses. Abnormal measurements should therefore always be assessed in the context of symptoms, personal medical history and, where appropriate, a medical examination.

The greatest value of a smartwatch lies not in a single number but in long-term change.People who focus on trends rather than daily readings generally obtain the most reliable information about their health.

Conclusion

Smartwatches and fitness trackers now provide far more than step counts. Heart rate, heart rate variability, sleep duration and activity levels offer continuous insights into physiological processes that, only a few years ago, could be measured almost exclusively in medical settings.

At the same time, individual measurements differ considerably in their reliability. Resting heart rate, activity data, sleep duration and long-term changes in HRV are among the parameters best supported scientifically. Sleep stages, sleep scores and individual oxygen readings should be interpreted more cautiously because they are based on indirect measurements and manufacturer-specific algorithms.

For longevity research, wearables open up new possibilities because they continuously document health-related behaviours. However, their greatest value does not lie in diagnosing disease or determining biological age precisely. Rather, they help make relationships between sleep, movement, recovery and personal well-being visible.

Used correctly, smartwatches are therefore not medical diagnostic devices but tools for informed self-observation. Anyone who interprets the data in context, follows long-term trends and understands their limitations can gain valuable insights into personal lifestyle patterns and develop health-promoting habits more deliberately.

Sources

• Dunn J, Runge R, Snyder M. Wearables and the Medical Revolution. npj Digital Medicine. 2018;1:5.
• Fuller D, Colwell E, Low J, et al. Reliability and Validity of Commercially Available Wearable Devices for Measuring Steps, Energy Expenditure, and Heart Rate: Systematic Review. JMIR mHealth and uHealth. 2020;8(9):e18694.
• Haghayegh S, Khoshnevis S, Smolensky MH, Diller KR. Accuracy of Wristband Fitbit Models in Assessing Sleep: Systematic Review and Meta Analysis. Journal of Medical Internet Research. 2019;21(11):e16273.
• Chinoy ED, Cuellar JA, Huwa KE, et al. Performance of Seven Consumer Sleep Tracking Devices Compared With Polysomnography. Sleep. 2021;44(5):zsaa291.
• Perez MV, Mahaffey KW, Hedlin H, et al. Large Scale Assessment of a Smartwatch to Identify Atrial Fibrillation. New England Journal of Medicine. 2019;381(20):1909-1917.
• Lubitz SA, Faranesh AZ, Atlas SJ, et al. Detection of Atrial Fibrillation in a Large Population Using Wearable Devices. Circulation. 2022;146(19):1415-1424.
• Bent B, Goldstein BA, Kibbe WA, Dunn JP. Investigating Sources of Inaccuracy in Wearable Optical Heart Rate Sensors. npj Digital Medicine. 2020;3:18.
• Baron KG, Abbott S, Jao N, Manalo N, Mullen R. Orthosomnia: Are Some Patients Taking the Quantified Self Too Far? Journal of Clinical Sleep Medicine. 2017;13(2):351-354.