Falls and related injuries requiring rehabilitation are a frequent occurrence in older people. There are several commercially available assistive technologies available that aim to prevent and detect falls. The latest advancement is automatic falls detection in the form of wrist-worn technology referred to as a smartwatch. Sold as a consumer item, this paper explores the potential clinical application of smartwatches to occupational therapy practice and aimed to understand occupational therapists' perceptions of using smartwatches to detect falls in older adults (aged over 60). An exploratory quantitative study using an online self-administered survey was used to gather data including: (1) multiple choice questions on demographics, (2) Likert scale questions using the Technology Acceptance Model to understand therapists' perceptions, and (3) open-ended questions to explore facilitators and barriers to using smartwatches to detect falls. A total of 36 participants fully completed the survey, showing that occupational therapists are open to prescribing smartwatches for fall detection purposes. Personal, environmental, and occupational facilitators and barriers to utilising smartwatches for falls detection were identified by thematic analysis. Smartwatches are perceived as a promising option as a fall detection device for some of the population. Identified barriers reported relate to the end user's ability to manage a smartwatch, therapist skills and knowledge regarding smartwatches and limited insight into funding. Smartwatches are not a solution for everyone, however, do provide an additional choice to keep older people at risk of falls safe and the reassurance that if they fall, help can be alerted. The number of older people is increasing, as is the demand to prevent falls and keep people safe and independent in their homes. There are a range of falls detection devices marketed to assist with the knowledge that should they fall, help can be alerted. Ownership of smartwatches is increasing and include several health assistive functions such as fall detection. As rehabilitation specialists, occupational therapists have a role of assessing the older persons falls risk in the home, prescribing suitable devices, and supporting access to fund to these items. As smartwatches are seen as a fashion accessory, they have the potential to be more acceptable to some individuals, however, are unlikely to be considered as a strategy by occupational therapists when prescribing falls detection systems. Using a survey, it was found that occupational therapists are open to the idea, however identified barriers that relate to the end user’s ability to manage a smartwatch, (such as physical disabilities or sensory limitations), skills and knowledge of therapists around the strengths and limitations of smartwatches and limited insight into funding sources. It is recognised that smartwatches are not a solution for everyone, however, do provide an additional choice to keep older people at risk of falls safe and the reassurance that if they fall, help can be alerted.
Sleep is routinely assessed in the management of mental health conditions. Wearable technologies like smartwatches offer a non-intrusive method to quantitatively measure sleep. However, there are limited empirical benchmarks for sleep duration and sleep quality measured by wearables against user reports. This study aims to evaluate the concordance between user-reported and smartwatch-measured hours of sleep and sleep quality. Participants were recruited from two decentralized digital health well-being studies and completed a 7-day sleep diary while simultaneously wearing their smartwatch to sleep (November 7, 2023 - June 30, 2024). Participants self-reported sleep timestamps and perceived sleep quality using the Sleep Quality Scale. Sleep timestamps and quality were also derived from their smartwatches (Garmin Vívoactive 4 2019, Garmin Venu 2 Plus 2022, and Garmin Venu 3/3S 2023). Statistical analyses included paired t-tests, equipercentile linking, and chi-square tests to assess agreement between smartwatch and self-reported sleep parameters. Exploratory analyses established the difference between reported and recorded sleep duration in healthcare shift workers. From 841 sleep instances reported by 130 participants wearing three different generation smartwatches, the mean difference in sleep duration between smartwatch-recorded and participant-reported was 21.22 (Garmin Vívoactive), 11.67 (Garmin Venu 2 Plus), and 6.58 (Garmin Venu 3/3S) minutes, respectively. There were statistically significant between-group differences in mean sleep durations assessed by participant self-report vs. Vívoactive 4 smartwatches, but not self-report vs. Venu 2 Plus or Venu 3/3S smartwatches. Equipercentile linking revealed concordance between smartwatch sleep scores and self-reported sleep quality using the Sleep Quality Scale (SQS) between 4 and 7, with disagreements observed at the SQS ranges from 0-4 and 7-10. These results suggest that wearables can reliably measure sleep duration, and future research warrants improvements in algorithms that estimate sleep quality with validations across different wearable vendors.
We previously developed and validated a 20-second upper extremity function (UEF) test for frailty assessment using wearable-derived motor and heart rate (HR) metrics. To evaluate, among older adults in hospital, the validity of a smartwatch for extracting UEF motor and HR outcomes relative to wearable sensors, and the test-retest reliability of UEF motor and HR outcomes across two back-to-back trials. Adults ≥65 years performed the UEF task. Motor outcomes included elbow angular speed, moment, range of motion, speed reduction, and speed variability. HR outcomes included mean HR during baseline, task, and post-task, and measures of HR increase and recovery. In Experiment I (validation), motor outcomes were compared between motion sensor and smartwatch (n = 43; age=68.98 ± 13.3), and HR outcomes between ECG sensor and smartwatch (n = 23; age=77.26 ± 8.62). In Experiment II, test-retest reliability was quantified for motor outcomes (n = 133; age=73.04 ± 10.99) and HR outcomes (n = 85; age=74.53 ± 9.83). Agreement and repeatability were quantified using two-way mixed effects intraclass correlation coefficients (ICC). Smartwatch demonstrated moderate-to-excellent agreement with the motion sensor (ICC = 0.67-0.90) and good-to-excellent agreement for most (78%) HR features (ICC = 0.75-0.94). Test-retest reliability within repeated trials was good-to-excellent for most (83%) motor outcomes (ICC = 0.75-0.96), and for most (78%) HR outcomes (ICC = 0.80-0.96). A brief seated UEF assessment provides reproducible motor and HR-derived measures in clinic. Further a smartwatch may capture UEF motor outcomes and most HR features with agreement relative to reference sensors.
Smartwatches are widely used to monitor physiological responses during exercise; however, their accuracy in measuring heart rate (HR) and energy expenditure (EE) across different exercise modalities remains insufficiently characterized. This study evaluated the accuracy of HR and EE measurements obtained from four commercially available smartwatches in comparison with gold-standard reference methods. Sixty-two healthy adult men performed standardized endurance and resistance exercise protocols while simultaneously wearing four smartwatches (Apple, Galaxy, Fitbit, and Garmin). HR was measured using electrocardiography (ECG), and EE was determined using indirect calorimetry. Measurement accuracy was assessed using repeated-measures analysis of variance, Pearson's correlation analysis, intraclass correlation coefficients (ICCs), and Bland-Altman analyses. All smartwatches demonstrated high accuracy in HR measurements during both endurance and resistance exercises. During endurance exercise, HR measurements from all smartwatch brands were comparable to those obtained via ECG, whereas during resistance exercise, only the Apple Watch showed no significant difference from the ECG. HRs showed strong correlations with ECG readings (r = 0.64-0.97), excellent reliability (ICC > 0.94), and narrow limits of agreement (approximately ±10 bpm). In contrast, the EE measurements exhibited limited accuracy across all devices. During endurance exercise, EE was consistently underestimated with wide limits of agreement. EE accuracy further deteriorated during resistance exercise, showing weak correlations with indirect calorimetry (r = 0.10-0.34) and poor reliability (ICC < 0.45). Overall, smartwatches provide accurate HR measurements across endurance and resistance exercise modalities, supporting their use in exercise intensity monitoring and HR-based training. However, smartwatch-derived EE estimates do not accurately reflect the metabolic demands, particularly during resistance exercises. Future research should focus on improving EE estimation algorithms through multimodal biosignal integration and machine-learning approaches, and validating these methods across diverse populations and exercise modalities.
Introduction: Palpitations are one of the most common cardiovascular complaints, affecting approximately 6% to 11% of the general population. Since palpitations often occur sporadically and resolve before medical evaluation, diagnosing the underlying rhythm disturbance requires documentation via an electrocardiogram (ECG) recorded during the symptomatic episode. The standard tool for this purpose has long been the 24-h Holter monitor, which has significant limitations, with diagnostic yields as low as 10% to 15%. Objective: This study aims to evaluate the feasibility and diagnostic yield of single-lead ECG recordings from smartwatches in patients presenting with palpitations. Methods: From 1 May 2023 to 1 May 2025, we conducted a prospective, real-world cohort study among consecutive adults referred to the University Hospital of Ferrara-based arrhythmia outpatient clinics for evaluation of palpitations. Eligibility required patients to be ≥21 years of age, report palpitations for which ambulatory documentation was clinically indicated, and already own a compatible smartwatch capable of single-lead ECG. Participants were trained to record a 30-s single-lead ECG at the onset of symptoms. Tracings were transmitted securely and independently reviewed by two blinded electrophysiologists. Results: Fifty-nine patients were enrolled (mean age 52 years, 64% male). Thirty-one patients (52%) transmitted at least one smartwatch-derived electrocardiographic tracing. Seventy-seven smartwatch tracings were received. Of these, 73 (95%) were interpretable; 57 (78%) showed an arrhythmia, whereas 16 (22%) demonstrated normal sinus rhythm. Four recordings (5%) were non-interpretable. From the 57 arrhythmic tracings, 44 distinct arrhythmic diagnoses were identified. Paroxysmal atrial fibrillation (AF) accounted for 16 episodes. Other diagnosed arrhythmias included atrial flutter (n = 6), paroxysmal supraventricular tachycardia (PSVT) (n = 4), premature atrial complexes (PAC) (n = 6), premature ventricular complexes (PVC) (n = 9), inappropriate sinus tachycardia (n = 12), and second-degree atrioventricular (AV) block type I (n = 4). Conclusions: Smartwatch-based ECG monitoring in symptomatic patients is feasible and provides a high diagnostic yield for a broad spectrum of arrhythmias. Unlike large-scale population screening approaches, which generate vast datasets with limited clinical benefit, a symptom-driven strategy applied to carefully selected, educated, and motivated patients proves both clinically valuable and organizationally sustainable. Indeed, the mean number of tracings transmitted per patient was low (1.3), confirming the clinical and operational sustainability of this patient-triggered, real-world approach.
Hypertension is a significant risk factor for cardiovascular diseases and premature mortality, with its prevalence increasing due to population aging and lifestyle factors. Accurate measurement of blood pressure (BP) and arterial oxygen saturation is crucial for disease prevention and monitoring, and wearable devices have emerged as a promising alternative. However, their clinical reliability requires validation, particularly in older populations. The aim of this research was to evaluate and compare the measurement of BP and arterial oxygen saturation in older people using a smartwatch in comparison with reference devices. We recruited 50 participants aged between 50 and 89 years (mean 70.60, SD 12.03 y), including 34 female participants and 16 male participants. A total of 3 BP measurements were taken simultaneously using the smartwatch and an ambulatory BP monitoring device (reference device). Arterial oxygen saturation was measured simultaneously using the smartwatch and the oximeter. The paired-sample t test (2-tailed) was used to compare variables, and the intraclass correlation coefficient (ICC) was used to verify the correlation. When averaged values were considered, no significant differences were observed between the Samsung Galaxy Watch 6 and the reference device for systolic BP (P=.31) or diastolic BP (P=.88), with good agreement for both parameters (systolic BP ICC=0.88; diastolic BP ICC=0.88). Arterial oxygen saturation showed no significant difference between devices (P=.10), with moderate agreement (ICC=0.68). Heart rate measurements also showed no significant differences between devices (P=.54), demonstrating good agreement. The Samsung Galaxy Watch 6 demonstrated acceptable agreement with reference devices for BP and arterial oxygen saturation measurements 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 when measurements are obtained under standardized and physiologically stable conditions.
Early identification of individuals at risk for hypertension is essential for effective cardiovascular disease. Physiological and activity metrics derived from consumer smartwatches may offer a practical, noninvasive approach to identify individuals at increased risk before the clinical onset of hypertension. In this 12-month prospective observational study, 230 normotensive adults aged 30-60 years were followed using consumer smartwatches. Baseline wearable predictors were calculated as the mean of the first 30 days of valid data after enrollment and included heart rate variability, resting heart rate, and time spent in moderate-to-vigorous physical activity. Incident hypertension was defined according to current European guidelines using standardized office blood pressure measurements obtained at follow-up. During follow-up, 28 participants (12.2%) developed hypertension. Individuals who developed hypertension exhibited lower baseline heart rate variability and spent less time in moderate-to-vigorous physical activity compared with those who remained normotensive. In multivariable logistic regression analysis, lower heart rate variability, lower levels of physical activity, and higher body mass index were independently associated with incident hypertension. An interaction between autonomic variability and physical activity was observed, indicating that individuals with both reduced autonomic regulation and low physical activity had the highest predicted risk. Machine-learning models showed improved statistical discrimination compared with clinical variables alone and were used as complementary exploratory analyses. Smartwatch-derived autonomic and physical activity metrics were independently associated with the development of hypertension over a 12-month period. These findings from an observational study suggest a potential role for wearable-derived physiological parameters as digital biomarkers for early hypertension risk stratification, although further validation in larger and externally replicated cohorts is required.
Randomized controlled trials (RCTs) aim to maximize statistical power while minimizing cost and recruitment burden. In practice, randomization is often stratified or restricted using demographic variables such as age and sex, while physiological heterogeneity that may influence treatment response is rarely incorporated. Consumer smartwatches are now widely used and provide continuous, real-world measurements of cardiovascular physiology and daily activity patterns, including resting heart rate, heart rate variability, sleep timing and regularity, and physical activity, capturing stable individual-level characteristics outside clinical settings. Leveraging these data, we developed Smartwatch-Informed Matching (SIM), a pre-randomization framework that groups physiologically similar participants and applies constrained randomization to assign participants to intervention and control arms. Using a prospective cohort of 4,795 individuals, we compared SIM with conventional age- and sex-based stratification. SIM improved covariate balance and increased similarity in symptom severity (Spearman ρ = 0.176 vs. 0.012) and physiological response profiles (Pearson r = 0.245 vs. 0.112). Power analyses showed that SIM reduced the sample size required to maintain statistical power by 9-18% across a range of effect sizes. These findings demonstrate that incorporating smartwatch-derived physiological similarity into pre-randomization design can enhance the efficiency and precision of randomized clinical trials. The SIM framework is also readily applicable to retrospective matched analyses that aim to reduce confounding.
Although consumer-grade smartwatches with a photoplethysmograph (PPG) can measure heart rate, their reliability and validity in healthy participants under free-living conditions remain unclear. Therefore, this study compared the heart rates measured by a smartwatch with PPG and a clinically accepted Holter electrocardiograph in healthy adult participants under completely free-living conditions. Ten participants wore a Holter recorder on their left chest and a Garmin vivosmart 5 on their non-dominant wrist simultaneously for 72-96 h. Averages were calculated every 2 min from the data obtained, and timestamps were used for alignment. Agreement between the measurements taken by the two devices was verified using intraclass correlation and Bland-Altman analyses. Error trends for Garmin vivosmart 5 were examined using linear regression analysis. The overall intraclass correlation coefficient range was 0.819-0.937 (mean: 0.902), indicating strong agreement. The mean bias was 1.16 bpm, and the mean limit of agreement was ±12.4 (7.87-20.6) bpm. Furthermore, the linear regression line intercept was negative for all participants, indicating that Garmin vivosmart 5 tended to underestimate the heart rate; however, the slope was positive and close to 0. Overall, the smartwatch maintained a certain accuracy level under fluctuating heart rates and demonstrated reasonable reliability during sleep or daily activities.
Vasovagal syncope (VVS) can cause injury and impaired quality of life, and effective prevention requires timely warning before loss of consciousness. To evaluate whether smartwatch photoplethysmography (PPG)-derived heart rate variability (HRV) can predict VVS before symptom onset, and to identify an optimal observation window and lead time. We prospectively enrolled 132 patients with suspected neurally mediated syncope who underwent head-up tilt (HUT) testing while wearing a wrist-worn Samsung Galaxy Watch 6 for continuous multiwavelength PPG acquisition (25 Hz). The HRV features (n = 107) were extracted. An Extra Trees classifier (600 trees) was trained using an 80/20 subject-level split and evaluated on a hold-out test set. Model performance was assessed using AUROC and threshold metrics, including specificity, at a fixed sensitivity of 0.90. Sixty-three participants were HUT-positive, and 69 were HUT-negative. The 5-min presyncope window achieved the highest discrimination (AUROC, 0.91; 95% CI 0.77-1.00). At 90% sensitivity, specificity was 0.64 (95% CI 0.40-0.85). Using a fixed 5-min window, early prediction remained robust at a 5-min lead time (AUROC 0.91; 95% CI 0.76-1.00; accuracy 84.6%; 95% CI 0.65-0.92). The most informative predictors included nonlinear complexity metrics (approximate entropy and composite multiscale entropy) and autonomic balance indices (normalized low-frequency, log-transformed high-frequency, and the cardiac vagal index). Artificial intelligence-enabled analysis of smartwatch PPG-derived HRV can prospectively predict VVS during HUT using a short 5-min observation window while maintaining clinically meaningful performance at a 5-min lead time, supporting the feasibility of wearable, real-time warning systems.
This study introduces the development of a next-generation wearable biosensor for non-invasive glucose monitoring in human sweat, leveraging the unique properties of advanced two-dimensional (2D) nanomaterials. Central to this innovation is a smartwatch-based platform designed for real-time health monitoring. The research focuses on three distinct field-effect transistor (FET) configurations, each utilizing a copper electrode array as the source, drain, and gate, and different 2D composite materials as the transistor channel. The FET channels were fabricated using a hydrothermal synthesis method to produce four high-performance composites: silver nanowires/Ti3C2 MXene, Ti3C2 MXene/MoS2, AgNWs/MoS2. and a ternary combination of MoS2/Ti3C2 MXene/silver nanowires. Each material contributes unique advantages-silver nanowires provide high electrical conductivity for efficient charge transport, MoS2 provides abundant active sites for glucose oxidation, and Ti3C2 MXene enhances both electrical conductivity and surface functionalization, facilitating improved glucose detection. In addition to glucose sensing, the developed smartwatch can simultaneously monitor pH and temperature, factors that are critical for accurate biosensing in sweat. Comprehensive material analysis, fabrication methods, and performance evaluations are detailed in this work. The proposed Bio-FET sensors demonstrate outstanding performance metrics, including an ultra-low detectable concentration of 0.001 µM, a sensitivity of 355 mA·mM⁻1, and a broad span of linear detection from 0 to 10 mM. The sensors also exhibit excellent repeatability, reproducibility, and long-term stability.
Long RP tachycardias share similar clinical presentations and challenging electrocardiographic features which frequently lead to diagnostic uncertainty. When unrecognized or misdiagnosed, the incessant behavior may lead to tachycardiomyopathy. A 28-year-old asymptomatic professional cyclist presented after his Apple Watch repeatedly reported sustained elevated resting heart rate. Physical examination, laboratory results, imaging, and echocardiography were normal. Electrocardiogram showed narrow QRS tachycardia with negative P waves in inferior leads. Electrophysiological study excluded accessory pathways, localizing an ectopic atrial substrate at the posteroseptal tricuspid annulus. Radiofrequency ablation successfully eliminated the focal atrial tachycardia, restoring sinus rhythm. The Apple Watch confirmed stable heart rate normalization postprocedure and at follow-up. We emphasize the role of wearable devices in early arrhythmia detection and the diagnostic challenges posed by long RP tachycardias. Smartwatches can fasten arrhythmias detection and management in asymptomatic athletes, and electrophysiological study remains essential for definitive supraventricular arrhythmias diagnosis.
High ambient temperatures are associated with a variety of negative outcomes, from exacerbated mental illness to aggression to increased dementia symptoms. One possible proximal mechanism influencing these impacts is heat-related changes in cognition. These effects of extreme heat on cognition have been widely investigated; however, the relationship between thermal discomfort at typically experienced temperatures and everyday cognition has received minimal attention. This pilot study evaluates the feasibility of a smartwatch-based ecological momentary assessment (EMA) design for effectively assessing this relationship. We examine whether thermal discomfort and distracting temperatures are sufficient to impair both objective (N-back task performance) and subjective (self-reported alertness) cognitive function. Results demonstrated that thermal discomfort led to worse performance on the N-back task. These effects were not affected by time of day but did show an interaction with acclimatization effects. The presence of distracting temperatures was also associated with lower scores on the N-back task. Taken together, the results of this pilot study demonstrate that deviations from comfortable temperature conditions can impair executive attention and cognitive control in daily life. Further, they highlight the utility of using combined EMA surveys and cognitive tasks to examine the effects of the physical environment on cognitive performance.
Couples generally manage chronic diseases together and the management takes an emotional toll on both patients and their romantic partners. Consequently, recognizing the emotions of each partner in daily life could provide insight into their emotional well-being in chronic disease management. Currently, the process of assessing each partner's emotions is manual, time-intensive, and costly. Despite the existence of works on emotion recognition among couples, none of these works have used data collected from couples' interactions in daily life. In this work, we collected 85 h (1021 5-min samples) of real-world multimodal smartwatch sensor data (speech, heart rate, accelerometer, and gyroscope) and self-reported emotion data (n = 612) from 26 partners (13 couples) managing diabetes mellitus type 2 in daily life. We extracted physiological, movement, acoustic, and linguistic features, and trained machine learning models (support vector machine and random forest) to recognize each partner's self-reported emotions (valence and arousal). Our results from the best models-balanced accuracies of 63.8% and 78.1% for arousal and valence respectively-are better than the results from (1) chance, (2) prior work that also used data from German-speaking, Swiss-based couples, and (3) partners' perceptions of each other's emotions. This work contributes toward building automated emotion recognition systems that would eventually enable partners to monitor their emotions in daily life and enable the delivery of interventions to improve their emotional well-being.
Background/Objectives: The itch-scratch cycle is a key driver of exacerbation in atopic dermatitis (AD) and requires objective monitoring, yet patient-reported itch scores are often unreliable in children. This study aimed to evaluate smartwatch-derived nocturnal scratching metrics as digital biomarkers of disease activity and treatment response in pediatric AD. Methods: In this prospective observational study, 50 children (median age 9 years) with physician-diagnosed AD wore an Apple Watch with the Itch Tracker application for 5-14 nights during initiation of topical therapy. Three scratch metrics-scratch count rate (SCR), scratch duration ratio (SDR), and scratch burden index (SBI, duration × intensity)-were analyzed. Associations with clinical outcomes [Eczema Area and Severity Index (EASI), Patient-Oriented Eczema Measure (POEM)], serum thymus and activation-regulated chemokine (TARC), and itch numerical rating scale (NRS) were examined. Logistic regression models were evaluated to examine whether these metrics could identify children who achieved clinically meaningful improvement, defined as EASI-50 plus ≥ 4-point POEM reduction. Results: All scratch metrics correlated with baseline EASI (r = 0.60-0.64, p < 0.001) and serum TARC (r = 0.58-0.60, p < 0.001). Reductions in scratching paralleled clinical improvement (r = 0.67-0.71, p < 0.0001). Among models, the SBI-based logistic regression demonstrated the best discriminative performance (AUC = 0.78, 95% CI: 0.64-0.92). Conclusions: Wearable-derived nocturnal scratching metrics showed moderate but consistent associations with disease severity and short-term improvement. Although predictive capability remains to be established, these metrics may serve as treatment-responsive digital measures. Given the cross-sectional nature of biomarker analyses and other study limitations, further prospective validation is required before clinical application in real-world pediatric AD monitoring.
暂无摘要（点击查看详情）
College students face increasing mental health challenges, including elevated stress and poor sleep quality. Exposure to natural environments has been associated with psychological and physiological benefits, yet the multidimensional relationships between green space exposure (GSE) and mental health remain insufficiently understood, particularly regarding exposure duration, frequency, and cumulative weekly exposure. This study examined associations between campus green space exposure and mental health outcomes among college students during a six-week wearable-based observational study using Apple Watch data. 43 healthy students from Shanghai Jiao Tong University participated in repeated assessments across baseline and intervention periods. Distinct exposure dimensions were associated with different mental health outcomes. Single-session exposure durations of approximately 17-35 minutes were associated with sleep-related outcomes. Weekly exposure frequency of approximately 3 sessions per week showed the strongest associations with daytime functioning, while cumulative weekly exposure of approximately 70-120 minutes was associated with emotional outcomes and differences in HRV amplitude patterns. These findings suggest that multidimensional green space exposure may relate differently to specific psychological and physiological outcomes among university students. This study demonstrates the feasibility of using wearable devices to support fine-grained ecological exposure assessment and may help inform future research on green space exposure and mental health in university settings.
Sleep duration has been shown to impact cardiovascular outcomes; however, the impact on wearable-derived metrics such as oxygen saturation and resting heart rate remains underinvestigated. This study aims to determine whether average sleep duration of ≤ 6 h per night impacts resting heart rate (RHR) and maximum nocturnal oxygen saturation (MaxSpO₂). Using data from the HEARTBEAT study, sleep duration and maximum oxygen saturation during sleep were collected via the Samsung Galaxy Watch. Participants were stratified into two groups based on average sleep duration (≤ 6 h vs > 6 h per night) and propensity-matched for baseline characteristics. The primary outcomes were RHR and MaxSpO₂. Statistical analysis was performed using R, version 4.5.1. After propensity score matching, there were 168 patients in each group with well-balanced baseline characteristics and comorbidities. Participants with average sleep duration ≤ 6 h demonstrated higher RHR compared to those with average sleep duration > 6 h (61.0 ± 5.8 bpm vs 57.7 ± 6.8 bpm; p < 0.001). Additionally, MaxSpO₂ was lower in the group with shorter sleep duration (95.1 ± 2.4% vs 96.2 ± 1.5%; p < 0.001). Sleeping ≤ 6 h per night is associated with higher resting heart rate and lower maximum nocturnal oxygen saturation.
暂无摘要（点击查看详情）
暂无摘要（点击查看详情）