This is a quantitative, descriptive, cross-sectional study aimed at verifying the reliability of wearable sensors (Samsung Galaxy Watch 6) for measuring BP and SpO₂ in participants aged 50 years and older.

Individuals aged 50 to 89 years were enrolled, comprising Caucasian individuals of Portuguese nationality from the Beira Interior region. Participants were recruited from the general community through local public announcements (eg, community notice boards and local outreach) and by direct invitation. No recruitment was conducted through medical care facilities, and participants were not employees of specific companies or institutions. Exclusion criteria included the inability to calibrate the Samsung Galaxy Watch 6 and the presence of wrist tattoos. Manufacturer specifications deemed systolic values greater than 169 and diastolic values less than 50 unacceptable for smartwatch calibration. Given the potential for various illnesses and medication use in this age group, these factors were not considered grounds for exclusion. BP measurements were conducted between January 2, 2024, and February 1, 2024.

To validate the reliability of the Samsung Galaxy Watch 6 (SM-R930) in terms of BP, we used a clinically validated reference BP device (TONOPORT V D-10829) under controlled conditions. According to North American [20,21], European [22,23], Japanese [24], and Chinese [25] clinical guidelines, reference-grade BP measurement devices are considered the standard for accurate BP assessment [26]. The TONOPORT V has previously demonstrated good accuracy and reliability for systolic and diastolic BP measurements when compared with mercury sphygmomanometry, fulfilling all phases of the European Society of Hypertension International Validation Protocol [27]. However, in this study, the device was used exclusively to perform stationary, office-based BP measurements under controlled conditions.

For calibration purposes, we adhered to the manufacturer’s guidelines. Participants were instructed to abstain from alcohol, caffeine, smoking, and exercise for 30 minutes prior to calibration, ensuring dry skin [9]. Calibration occurred in a tranquil setting, with participants seated comfortably, backs supported, arms resting on a table, legs uncrossed, and feet flat on the floor. Participants rested in this position for 5 minutes before commencement. During calibration, participants were instructed to remain still and breathe normally. Using the Samsung Health Monitor application, on-screen instructions guided the calibration process. Each participant underwent an individual calibration protocol consisting of 3 consecutive BP measurements obtained using a clinically validated reference BP device (TONOPORT V D-10829). Calibration was exclusively required for the BP variable, as per the manufacturer’s standards and instructions.

The cuff was installed in accordance with the device’s standards. Thus, the reference BP device (TONOPORT V D-10829) cuff was placed on the patient’s nondominant arm, at a height of about 2.5 cm above the elbow. The cuff would only be placed on the other arm if there were any limitations. The cuff was to fit snugly on the patient’s arm, but not too tightly [28].

Installation of the Samsung Galaxy Watch 6 followed the manufacturer’s instructions. In this way, the Samsung Galaxy Watch 6 was worn on the wrist contralateral to the one used for the reference BP device. The Galaxy Watch strap was appropriately fitted to the participant’s wrist without excessive tightness. Due to the placement of the reference BP device, it was decided to place the watch on the opposite arm to ensure that measurements did not conflict.

The pulse oximeter was placed on the patient’s right index finger, which was on the same wrist where the smartwatch was worn. To ensure an accurate reading, it was important to place the oximeter snugly against the patient’s skin. This meant that the sensor should be firmly attached to the skin, and there should be no gaps between the sensor and the skin. Once the oximeter was properly adjusted, we waited at least 30 seconds before taking the oxygen saturation reading. This allowed the sensor to acclimate to the patient’s temperature and blood flow. To ensure that painted fingernails would not cause erroneous readings, we placed the probe oriented laterally so that the sensor transmitted the light on the side of the finger [29].

The participant was subjected to a 5-minute rest (remaining seated). After resting for 5 minutes, we calibrated the Samsung Galaxy Watch 6 in relation to BP. After calibration, 3 BP measurements were taken with a 5-minute interval between each measurement. The reference BP device (TONOPORT V D-10829, Germany) and the Samsung Galaxy Watch 6 (SM-R930) were used to measure BP simultaneously.

For analytical purposes, BP data were evaluated using 2 complementary approaches. First, each measurement session was analyzed separately to assess within-session variability and agreement between the smartwatch and the reference device at each time point, without averaging across sessions. In addition to the analysis of each individual measurement session, particular emphasis was placed on the overall analysis calculated as the mean of the 3 consecutive measurements. This approach was adopted to reduce the influence of random variability and isolated measurement errors, which are common in single BP readings [22,23]. The use of averaged values is consistent with standard clinical practice, where digital BP devices typically rely on multiple consecutive measurements to improve accuracy and support clinical decision-making [22,23].

SpO₂ was measured after the BP measurements. One measurement was taken simultaneously with the oximeter (Proficare PO 3104) and the Samsung Galaxy Watch 6. All BP and SpO₂ measurements were performed with participants in a seated position, resting comfortably in a chair with back support. Participants were instructed to keep both feet flat on the floor, legs uncrossed, and arms supported at heart level. A resting period was observed prior to measurements to ensure hemodynamic stabilization. The same body position was maintained throughout the measurement protocol.

Microsoft Office Excel 2013 and the statistical analysis software IBM SPSS version 28.0 were used for data analysis. The calculation of means, SDs, differences, and 95% CIs was based on standard statistical methods. To verify the normality of the data, the Kolmogorov-Smirnov test (n>30) was used. The paired-sample t test (2-tailed) was used to compare the variables. The degree of agreement between measurements was evaluated using the intraclass correlation coefficient (ICC), calculated with a 2-way mixed-effects model with absolute agreement and single measures. ICC values were interpreted as indicating low reliability when below 0.5, moderate reliability between 0.5 and 0.75, good reliability between 0.75 and 0.9, and excellent reliability when greater than 0.90. The significance level for rejecting the null hypothesis was set at an α level of .05. Using Bland-Altman graphs, the difference was calculated as the reference BP device minus the smartwatch measurements. All assessments obtained from the reference BP device and the smartwatch were used to construct the Bland-Altman graphs. The same procedure was also applied to SpO₂ measurements. An a priori sample size evaluation indicated that a sample of approximately 50 participants would provide 80% statistical power to detect small-to-moderate standardized differences between measurement methods (Cohen d=0.4), assuming a 2-sided significant α level of .05.

All procedures followed the guidelines of the Declaration of Helsinki and were approved by the research ethics committee of the University of Beira Interior (approval CE-UBI-Pj-2023‐064-ID1994). All participants signed a declaration of participation. Prior to participation, all individuals received detailed oral and written information regarding the study objectives, procedures, potential risks, and benefits, and they provided written informed consent. Participation was voluntary, and participants were assured of confidentiality and the right to withdraw from the study at any time without any consequences.

This study aimed to assess the reliability of the Samsung Galaxy Watch 6 in measuring BP and SpO₂ among individuals aged 50 to 89 years. Our findings suggest that the Samsung Galaxy Watch 6 demonstrates reliability in these measurements among individuals older than 50 years. The reliability of these data is crucial for facilitating the development of digital biomarkers and supporting clinical research endeavors involving monitoring [15]. Furthermore, this reliability is instrumental in the prevention and detection of adverse cardiac events [15,30]. Improved measurement reliability enhances the potential for wearables to serve as affordable and valuable tools for health monitoring, thereby promoting equitable access to health care [15,16,31,32].

It is important to emphasize that reliability in this context refers to agreement under standardized, calibrated, resting conditions, and should not be interpreted as full interchangeability with clinical-grade reference devices. Agreement between methods was, therefore, interpreted primarily based on Bland-Altman analysis, limits of agreement, and clinically meaningful absolute differences, rather than the absence of statistically significant mean differences.

Our findings indicate that the Samsung Galaxy Watch 6 reliably measures BP, showing a strong correlation compared to the reference method, with results falling within the range of good to excellent. This underscores the reliability of such technology for BP measurement among individuals aged 50 to 89 years. A previous study on the Samsung Galaxy Watch 4, a predecessor to the model used in our research, also demonstrated reliability in individuals aged 18 years, displaying a level of accuracy compared to the reference method [33]. Similarly, research on another wearable device, the Aktiia Bracelet, conducted on a population aged 2 to 65 years, revealed its safety and effectiveness in measuring BP compared to ambulatory blood pressure monitoring (ABPM) [34]. Recent reviews emphasize that smartwatch-based BP technologies should therefore be interpreted as complementary tools rather than substitutes for conventional clinical measurements [19].

In line with these recommendations, the findings indicate acceptable agreement at the group level, while Bland-Altman analysis revealed interindividual variability that should be considered when interpreting single measurements, particularly at higher BP values.

However, in another study validating BP measurement using a Samsung Galaxy Watch Active 2 smartwatch, a previous model used in our research, compared to ABPM, a systematic bias toward a calibration point was observed, resulting in an overestimation of low BP and an underestimation of high BP [19]. One contributing factor to these discrepancies is that the Samsung Galaxy Watch Active 2 smartwatch does not perform automatic measurements; instead, patients had to manually activate the smartwatch measurement for 24 hours following the ABPM measurement [19]. Additionally, differences in firmware versions among the same device model can lead to varying conclusions [9]. In this study, the assessments were meticulously controlled and conducted simultaneously using the Samsung Galaxy Watch 6 (SM-R930) and the TONOPORT V device.

The results of the measurement of SpO₂ by the Samsung Galaxy Watch 6 smartwatch and the portable finger pulse oximeter appear to show a good level of agreement and positive correlation. Our results reinforce the reliability of smartwatches and the high accuracy of these devices compared to a standard pulse oximeter as evaluated in other studies [35,36]. In addition to the reliability of these wearables for use in clinical decision-making [35], smartwatches enable continuous and long-term monitoring [37]. These wearables can detect abnormal fluctuations and more quickly assess changes in the patient’s state of health over time [37]. In addition, this type of technology is particularly advantageous for some pathologies, such as chronic lung disease, sleep apnea, or post-COVID syndrome [37]. On the other hand, wearables take the burden off measurements taken in a medical environment and can minimize errors associated with anxiety induction and increased adrenergic activity in clinical settings, known as “hypertension or white coat syndrome” [4,9]. Moderate reliability indicates that, although the smartwatch may provide approximate SpO₂ estimates under controlled, normoxic resting conditions, its measurements should not be considered interchangeable with those obtained from clinical-grade pulse oximeters. Under these conditions, smartwatch-based SpO₂ measurements may be suitable for general trend monitoring in stable individuals but remain inadequate for clinical decision-making. It is important to note that part of the observed variability may be attributed to differences in the measurement site, as wrist-based assessments are more susceptible to variations in skin temperature and peripheral perfusion compared with finger-based pulse oximetry. Recent advances in wearable photoplethysmography sensor technology and signal processing have improved measurement performance; however, physiological factors, such as peripheral perfusion, skin temperature, and anatomical measurement site remain important sources of variability [15]. Contemporary reviews highlight that wrist-based photoplethysmography sensors are inherently more susceptible to these influences than finger-based pulse oximetry, reinforcing the cautious interpretation of smartwatch SpO₂ measurements [15].

When we validate portable devices, we have to be careful and follow a series of guidelines. In our study, we followed guidelines that can influence reproducibility and scientific rigor [9]. Bearing these guidelines in mind, our research attempted to verify the reliability of Samsung devices in different contexts, including sex and various age groups. Individual differences, sex variability, and demographics must be taken into account when collecting data using an optical sensor (photoplethysmography) [9]. Sex must be assessed to verify possible measurement errors between male participants and female participants [9], as some authors have reported a higher error rate in males than in females [37]. We did not find any significant differences between sexes with this smartwatch, possibly due to its more reliable function.

Much of the research on this topic does not take into account factors such as race and ethnicity, which can impair the observation and interpretation of health results [9]. Another guideline evaluated was skin tone, which should be taken into account when evaluating studies using wearables [9]. Due to the homogeneity of the population studied in terms of race, we did not encounter any difficulties in this regard. Another factor that can influence measurements is whether the participant has tattoos [9], which was not the case in our study. There also seems to be no consensus on the accuracy and absorption of green light in people with lighter skin tones [9]. Wearables that use green LED light have a shorter wavelength, so the amount of light that passes through the tissue can be limited [9]. Our research was carried out on a light-skinned Portuguese population without any tattoos in the area of the optical sensor. According to the results of our research, there were no significant changes in any of the variables evaluated. These results are in line with other studies that have found no differences in the use of optical green light sensors in people with lighter skin tones [31,38].

The adoption of wearable devices has the potential to improve health care system efficiency by enabling the early detection of health problems, monitoring treatment outcomes, and reducing the burden on health care infrastructures [39-41]. Continuous monitoring through digital technologies may support more personalized and effective health management, potentially lowering costs related to hospitalizations and emergency care while improving health outcomes [39-41].

Wearable devices and mobile health platforms may also help reduce inequalities in access to health care by providing affordable monitoring solutions, particularly in resource-limited settings, thereby promoting more equitable health care delivery [41]. Overall, these technologies have the potential to enhance both individual and population health while optimizing health care resources.

This study has several limitations that should be acknowledged, including the relatively small sample size, the homogeneous demographic characteristics of the study population, and the fact that all measurements were conducted on the same day and under resting conditions only, which may not reflect BP variability during daily activities. The study included only 50 participants from a Caucasian Portuguese population, which limits the generalizability of the findings to other ethnicities or populations. Therefore, the results should be interpreted with caution and should not be extrapolated beyond similar demographic contexts without further validation.

An additional limitation is that smartwatch calibration and validation were performed using the same reference BP device within a single session, which may partially inflate agreement and does not allow for the assessment of calibration stability across days. Another limitation is that BP was measured on contralateral arms, and interarm differences may have contributed to measurement variability; therefore, findings should be interpreted as an arm-to-arm comparison under standardized resting conditions.

Therefore, future studies are recommended to (1) assess the reliability of the wearable for BP measurement at different times of the day; (2) evaluate BP measurement reliability after more intense daily activities; (3) examine reliability across different days; (4) verify smartwatch-based SpO₂ measurements using repeated assessments; (5) assess SpO₂ reliability during movement and daily activities; (6) evaluate SpO₂ measurements across a broader range of oxygen saturation values, including hypoxemic conditions; and (7) investigate the influence of BMI and other individual factors, such as skin tone, on measurement accuracy. Future validation protocols should also incorporate calibration and validation sessions performed on different days and assess device performance during unsupervised, real-world use.

Wearable devices may represent a cost-effective and practical approach for collecting longitudinal BP data and constructing individual BP profiles. However, to fully realize this potential, wearable technologies should be capable of continuous and autonomous BP monitoring. The smartwatch used in this study does not currently support fully automatic BP measurements, which represents an additional limitation.

This study assessed the agreement between the Samsung Galaxy Watch 6 and reference devices for BP and SpO₂ measurements in adults aged 50 years and older. The findings indicate that the smartwatch demonstrates good agreement with the reference method for SBP and DBP when averaged across repeated measurements, despite statistically significant differences observed in some individual measurement sessions. These results suggest that the device may be suitable for resting BP monitoring based on repeated measurements, consistent with standard clinical practice.

For SpO₂, the smartwatch showed moderate agreement with the reference pulse oximeter, with no significant systematic bias. This level of agreement supports its use for nonclinical monitoring and trend observation under resting conditions.

In summary, the Samsung Galaxy Watch 6 demonstrated acceptable agreement with reference devices for the assessment of BP and SpO₂ in older adults without decompensated clinical conditions, evaluated under controlled resting conditions. These findings indicate that the device provides reliable measurements within this specific population and context, thereby supporting its use as a complementary assessment tool when measurements are obtained under standardized and physiologically stable conditions.