Authors: Hugo Vieira Pereira , Rodrigo Martins , Mafalda Antunes e Paulo Pires (exercise physiology) Wearables have become part of everyday life. They include smartwatches, fitness bands, rings, and other devices worn on the body that count steps, track exercise, estimate calorie expenditure, and monitor sleep. Some of them also provide information about fitness levels, recovery, or even alert users to changes in heart rhythm. After a walk, they display distance covered, pace, calories burned, and sometimes an estimate of fitness level. The following morning, they may rate your sleep and suggest whether you are ready to train again. This is a large amount of information generated by different sensors and algorithms. All this data can be useful, but not all of it has the same level of accuracy or meaning. Understanding what is directly measured, what is estimated, and how to turn that information into realistic action is essential for using these devices properly. Measuring and estimating are not the same thing Data collected by wearables can help users understand their habits, monitor exercise, and identify patterns over time. To make the most of these devices, it is important to understand what is measured, what is estimated, and the limitations of the information provided. Sensors collect signals from both movement and the body. An accelerometer detects movement acceleration, GPS helps estimate speed and distance travelled and an optical wrist sensor analyses changes in light associated with blood flow. Based on these signals, the device software calculates various indicators. As a result, not every value shown on the screen is a direct measurement. Many are estimates produced by algorithms that combine multiple signals with information about the user. This helps explain why not all indicators have the same level of accuracy. Step count, activity time, distance, pace, and heart rate can be useful for tracking trends. However, accuracy varies according to the device model and software version, how the device is worn, skin contact, GPS quality, type of movement, and exercise intensity. Even step counting is not perfect. Walking very slowly may lead to some steps not being recorded, while certain repetitive arm or wrist movements may occasionally be interpreted as steps. In addition, devices worn on different parts of the body may produce different counts. Results from a large study published in 2024 concluded that only around 11% of identified devices had been validated for at least one biometric indicator. Furthermore, validation for one metric does not guarantee the same accuracy for all others. When the data are truly useful In practice, wearable data are most informative when their evolution is monitored over time using the same device under similar conditions. A single measurement, or a direct comparison between values obtained from devices from different manufacturers, should be interpreted with caution. Indicators that require extra care Calories and energy expenditure Margins of error can be substantial and vary between activities and devices. These estimates should not be treated as exact values when deciding how much to eat or as a definitive measure of workout quality. Maximum oxygen uptake (VO₂ Max) This indicator reflects the body ability to take in, transport, and use oxygen during exercise and is an important measure of cardiorespiratory fitness. On a smartwatch, VO₂ max is usually an estimate based on the relationship between heart rate, pace, and other available data. It can be useful for tracking personal trends but is not equivalent to a direct measurement obtained during a cardiopulmonary exercise test with respiratory gas analysis. Training load and recovery These scores combine data such as heart rate, heart rate variability, sleep, and exercise history. They can help organize information but do not account for the entire picture. Illness, stress, heat, medication, pain, and fatigue also influence how the body responds to training. Therefore, a single recovery score should not determine on its own whether you should train or rest. Device data should always be considered alongside how you feel, your perceived effort, symptoms, and overall circumstances. Sleep Wearables can be useful for monitoring sleep schedules, regularity, and approximate sleep duration. Estimates of sleep stages are less reliable and do not replace clinical assessment or sleep study. Persistent daytime sleepiness, heavy snoring, observed breathing pauses during sleep, or prolonged insomnia should prompt medical advice, even when a wearable reports normal values. Oxygen saturation and electrocardiogram (ECG) Some devices estimate peripheral oxygen saturation, while others can record a simplified electrocardiogram. An alert from a device can be useful in prompting further assessment. However, it is not the same as a clinical evaluation and does not replace the medical tests required in a particular situation. A device alert alone does not establish a diagnosis and, equally, an apparently normal result does not rule out disease. From data to behavior change Activity trackers can help people become more physically active. Research has shown that users of these devices often walk more and spend more time being active. However, this does not mean that simply purchasing or wearing a wearable will automatically produce these benefits. For a behavior to occur and be maintained, people need the capability, opportunity, and motivation to perform it. Devices can support some of these factors through techniques such as self-monitoring, goal setting, feedback, and reminders. Nevertheless, a reminder on your wrist cannot solve a lack of time, an unsafe environment, symptoms that limit exercise, or the absence of an appropriate plan. Turn information into real action To get the most from your wearable: Wearables do not replace healthcare If you have a medical condition, take medication that affects heart rate, or have concerns about exercising safely, tracking your wearable data and sharing it with the healthcare team looking after you may be useful. Select the relevant period, specify the device model, describe the circumstances surrounding any changes, and mention any associated symptoms. A well-described pattern is far more useful than dozens of isolated values without context. It is also important to remember that chest pain, fainting, severe shortness of breath, palpitations accompanied by feeling unwell, or other acute symptoms require prompt medical attention regardless of what your smartwatch or other wearable indicates. More data do not automatically mean better health. A wearable is most useful when it helps you recognize patterns, choose realistic actions, and communicate more effectively with healthcare professionals. Device-generated information complements personal awareness and clinical assessment; it does not replace them. Sources: Barrera N, Solorzano M, Jimenez Y, et al. Accuracy of smartwatches in the detection of atrial fibrillation: a systematic review and diagnostic meta-analysis . JACC Adv 2025;4:102133. Doherty C, Baldwin M, Keogh A, et al. Keeping pace with wearables: a living umbrella review of systematic reviews evaluating the accuracy of consumer wearable technologies in health measurement . Sports Med 2024;54:2907-26. Ferguson T, Olds T, Curtis R, et al. Effectiveness of wearable activity trackers to increase physical activity and improve health: a systematic review of systematic reviews and meta-analyses. Lancet Digit Health 2022;4:e615-26. Meet the sports medicine center clinical team from Hospital da Luz network. The content published in the Health Guide is intended for the general public and is provided for informational and educational purposes only. It does not replace individual medical assessment, diagnosis, or personalized clinical advice from a qualified healthcare professional. The content is produced or reviewed by physicians and other healthcare professionals at Hospital da Luz and is based on the knowledge available at the time of publication or the most recent update, as indicated at the beginning of each item.
