Wearable health trackers reveal how accurate your smartwatch really is

By Priyanjana Pramanik, MSc.Reviewed by Benedette Cuffari, M.Sc.Sep 8 2025

A new review shows that while wearables can reliably track steps and heart rate, their accuracy for sleep, stress, and blood pressure remains uneven, raising questions about when these devices will be ready for clinical use.

Study: A guide to consumer-grade wearables in cardiovascular clinical care and population health for non-experts. Image Credit: Nan_got / Shutterstock.com

A recent review published in npj Cardiovascular Health compares the different health parameters measured by common wearable sensors used by consumers and provides evidence of their accuracy, validity, regulatory clearance status, and practical issues.

How consumer devices are shaping public health

Wearable devices like rings, smartwatches, and wristbands are widely popular, allowing consumers to monitor their health and activity continuously in real time. In clinical and research settings, these powerful technologies collect large-scale health data, monitor population trends, develop risk prediction tools, and evaluate interventions.

Most of the health metrics these wearables analyze are based on sensor signals such as photoplethysmography (PPG) or accelerometry, combined with proprietary algorithms that are not publicly disclosed. Frequent updates and new product versions add further complexity, challenging device comparison and clinical value assessment. Additionally, issues such as standardized data acquisition and large-scale analysis pipelines must be addressed because wearables are designed for personal rather than population-level use.

The current review provides practical guidance for healthcare and research professionals on the different types of sensors currently available, their measurable health parameters, validation evidence, regulatory insight, and future implications for population and cardiovascular health.

PPG, ECG, and beyond

PPG is an optical technique that uses light to detect blood volume changes in the skin. As a result, PPG is widely used in smartwatches to estimate heart rate (HR), HR variability (HRV), respiratory rate, oxygen saturation, and blood pressure.

Accelerometers, particularly tri-axial versions, measure movement and body orientation across three planes, making them essential for assessing physical activity and posture in fitness trackers.

Electrocardiography (ECG), traditionally a clinical tool, has also been integrated into consumer wearables to enable short single-lead recordings comparable to lead I of a standard 12-lead ECG. These measurements can help identify rhythm abnormalities such as atrial fibrillation.

Novel sensors are increasingly being developed to expand modern wearable devices' capabilities, including barometers, magnetometers, global positioning systems (GPS), and thermometers for environmental and location tracking. Electrodermal activity (EDA) sensors have also been utilized to detect changes in skin conductance to infer stress levels. Comparatively, bioelectrical impedance (BioZ) provides estimates of body composition and, when combined with ECG, has the potential to predict an individual’s risk of heart failure.

Gyroscopes, which measure angular velocity, may also be paired with accelerometers in a technique known as gyrocardiography to identify subtle cardiac vibrations. These sensors collectively enable wearable devices to monitor cardiovascular, respiratory, metabolic, and behavioral parameters to enhance their role in personal health management and large-scale population research.

Are wearable metrics trustworthy?

Wearable devices measure a wide range of physiological and behavioral parameters relevant to health monitoring and disease prevention. Most devices accurately monitor resting HR, with only minor errors compared to ECG measurements. However, the precision of these measurements declines during physical activity due to motion, sweat, and contact pressure.

HRV, an indication of autonomic nervous system function, can be evaluated from both ECG and PPG. Wearable-derived pulse rate variability demonstrates strong concordance with ECG measurements, particularly at rest.

Wearable devices also play an important role in arrhythmia detection, especially atrial fibrillation (AF). Several models from Apple, Fitbit, Samsung, and Withings have received U.S. FDA clearance for AF detection, although these functions are intended for pre-diagnostic rather than clinical decision-making. In addition to HR, wearable devices can also estimate cardiovascular and respiratory parameters like cardiac intervals, respiratory rate, and oxygen saturation (SpO₂).

Wearable devices frequently provide data on physical activity metrics like step count, distance traveled, and energy expenditure. Step counts are generally more reliable than energy expenditure estimates, which vary significantly between devices.

Sleep monitoring uses HR and accelerometer signals to monitor sleep duration and stages, with moderate accuracy compared to polysomnography. However, wearables often overestimate sleep because they misclassify quiet wakefulness as sleep. Stress detection is based on HRV, respiratory rate, and EDA; however, motion artifacts may reduce the reliability of these measurements.

Women’s health features, such as cycle and ovulation tracking, leverage HRV, temperature, and respiratory changes, with growing evidence for reasonable accuracy. Blood pressure estimation using wearables remains challenging. Cuffless approaches can require calibration against traditional BP cuffs, and accuracy varies between devices and populations.

Overall, wearable devices provide accessible and scalable health insights. However, their accuracy varies by parameter, activity, population group, and device, underscoring the need for continued validation before widespread clinical adoption.

Consumer-grade wearables provide an opportunity to understand public health trends, develop risk stratification tools, and monitor interventions.

Balancing the potential and limitations of wearables

Wearable health devices are powerful tools capable of continuously collecting large-scale long-term health data. As these devices become more affordable, public health researchers hope they can evolve to facilitate the early detection of conditions such as atrial fibrillation and chronic diseases like heart failure. Wearable devices may also support preventive care, improve patient adherence to rehabilitation programs, and generate valuable population-level insights into physical activity, obesity, and mental health.

Nevertheless, restricted access to raw data, inconsistent data formats, and reliance on proprietary algorithms limit standardization and analysis. Privacy, security, and regulatory concerns pose additional barriers, as does the risk of digital exclusion for certain groups.

Thus, while wearables show great potential to transform cardiovascular care and public health research, robust validation, clear clinical pathways, and equitable implementation are essential before widespread adoption.