Smartwatch-Derived Exercise Metrics as Predictors of Early Hypertension: A Prospective Observational Study

Smartwatch-Derived Exercise Metrics as Predictors of Early Hypertension: A Prospective Observational Study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Smartwatch-Derived Exercise Metrics as Predictors of Early Hypertension: A Prospective Observational Study Gökhan Ceyhun, Esma Selva Ateş Ceyhun This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8238959/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 12 You are reading this latest preprint version Abstract Background: Early detection of hypertension is crucial for cardiovascular prevention. Smartwatch-derived physiological and activity metrics may offer a practical tool for identifying individuals at risk before clinical onset. Methods: In this 12-month prospective study, 230 normotensive adults (30–60 years) were monitored using smartwatches recording heart rate variability, resting heart rate, and moderate-to-vigorous physical activity . Results: Twenty-eight participants (12.2%) developed hypertension. Those with new-onset hypertension had lower heart rate variability (42.6 ± 10.8 vs. 56.1 ± 12.9 ms, p < 0.001) and lower moderate-to-vigorous physical activity (38.4 ± 14.2 vs. 57.9 ± 16.7 min/day, p < 0.001). Reduced heart rate variability and moderate-to-vigorous physical activity independently predicted hypertension. Conclusions: Smartwatch-derived autonomic and exercise metrics can identify normotensive adults at risk of developing hypertension, supporting their role as digital biomarkers in preventive cardiology. Hypertension Heart rate variability Physical activity Smartwatch Wearable technology Figures Figure 1 Figure 2 Figure 3 Introduction Hypertension is a leading modifiable risk factor for cardiovascular disease and premature mortality worldwide. Despite extensive public health efforts, nearly half of hypertensive individuals remain undiagnosed or untreated, highlighting the importance of early detection strategies. Conventional office-based blood pressure assessments often fail to capture dynamic physiological changes that precede the onset of sustained hypertension, particularly those related to autonomic dysfunction and physical inactivity. Recent technological advances have enabled smartwatches to collect continuous cardiovascular and behavioral data under real-life conditions. Beyond heart rate tracking, these devices can measure heart rate variability (HRV), resting heart rate (RHR), and physical activity metrics such as moderate-to-vigorous physical activity (MVPA), VO₂max, and sedentary time, offering valuable insight into cardiovascular regulation. Reduced HRV has been associated with sympathetic overactivity and early vascular changes that predispose to hypertension, while low levels of MVPA contribute to endothelial dysfunction and elevated resting blood pressure[ 1 – 3 ]. Accumulating evidence suggests that HRV and daily activity levels may jointly reflect the autonomic–metabolic interplay underlying blood pressure regulation. Prospective studies have shown that lower HRV predicts incident hypertension in normotensive adults[ 4 ], and that regular aerobic activity reduces sympathetic tone and arterial stiffness, thus mitigating long-term hypertension risk[ 5 , 6 ]. However, few studies have integrated multiple smartwatch-derived parameters into a single predictive framework, and most existing data rely on short-term or cross-sectional analyses. Recent validation studies confirm the technical accuracy of smartwatch-based cardiovascular measurements[ 7 – 9 ], yet the clinical relevance of combining exercise and autonomic metrics to detect early hypertension remains largely unexplored. The integration of these digital biomarkers could provide a practical, noninvasive approach for identifying high-risk individuals before overt hypertension develops. Therefore, the present prospective observational study aimed to assess whether smartwatch-derived exercise and autonomic parameters can predict incident hypertension during 12 months of follow-up in normotensive adults. We hypothesized that decreased HRV, reduced MVPA, and their interaction would independently predict early hypertension and that combining these smartwatch metrics with clinical variables would enhance predictive performance compared with conventional models. Methods Study design and population This prospective observational study was conducted between January 2024 and January 2025 to evaluate whether smartwatch-derived exercise and autonomic parameters could predict incident hypertension among normotensive adults. A total of 230 participants aged 30–60 years were enrolled from outpatient cardiology and preventive medicine clinics. All participants provided written informed consent, and the study protocol was approved by the institutional ethics committee (Approval No: B.30.2.ATA.0.01.00/737) in accordance with the Declaration of Helsinki. Inclusion criteria: age 30–60 years, baseline systolic BP < 140 mmHg and diastolic BP < 90 mmHg, continuous smartwatch use ≥ 20 days per month during follow-up, willingness to attend scheduled visits, and provision of written informed consent. Exclusion criteria: known hypertension, coronary artery disease, heart failure or arrhythmia (including atrial fibrillation), current use of antihypertensive, antiarrhythmic, β-blocker, calcium-channel blocker, or ACE inhibitor therapy within 3 months before enrollment, diabetes mellitus, chronic kidney disease (eGFR 20 cigarettes/day) or alcohol consumption (> 20 g/day), shift work or irregular sleep–wake patterns, sleep apnea syndrome, missing or incomplete smartwatch data (> 15% missing), technical incompatibility preventing data synchronization, pregnancy, lactation, hormonal therapy, or inability/unwillingness to attend the 12-month follow-up visit. Smartwatch-derived parameters Participants wore a validated smartwatch (Apple Watch Series 9 or equivalent) continuously for 12 months. The following parameters were automatically recorded and extracted through the device’s health API: resting heart rate (RHR, bpm), heart rate variability (HRV, RMSSD, ms), moderate-to-vigorous physical activity (MVPA, min/day), sedentary time (min/day), estimated VO₂max (ml/kg/min), sleep efficiency (% of time asleep/time in bed), and stress index (0–100). Monthly mean values and 30-day changes (ΔHRV, ΔRHR, ΔMVPA) were computed. Interaction indices (HRV×MVPA, RHR×Sedentary time) were derived to explore autonomic–activity relationships. Clinical evaluation Blood pressure (BP) was measured at baseline and at 12 months using a calibrated oscillometric device after 5 minutes of rest. The mean of the last two of three readings was recorded. Incident hypertension was defined according to the 2023 ESC/ESH guidelines as systolic BP ≥ 140 mmHg and/or diastolic BP ≥ 90 mmHg, or initiation of antihypertensive therapy during follow-up. Statistical analysis All analyses were performed using Python 3.12 ( pandas , statsmodels , xgboost , lightgbm , shap ). Continuous variables were expressed as mean ± SD and categorical variables as percentages. Normality was assessed using the Shapiro–Wilk test. Group comparisons were made with the Student’s t-test or Mann–Whitney U test for continuous variables, and the χ² or Fisher’s exact test for categorical data. Multivariable logistic regression Independent predictors of incident hypertension were identified using multivariable logistic regression including age, sex, BMI, HRV, RHR, MVPA, sedentary time, VO₂max, sleep efficiency, and interaction terms (HRV×MVPA, RHR×Sedentary). Adjusted odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. Model calibration was assessed using the Hosmer–Lemeshow test, and discrimination using the area under the ROC curve (AUC). Machine learning analysis To complement traditional inference, predictive modeling was performed using Extreme Gradient Boosting (XGBoost) and Light Gradient Boosting Machine (LightGBM) with 10-fold stratified cross-validation. Feature importance was determined via SHAP (Shapley Additive Explanations) analysis. Performance metrics included ROC–AUC, F1-score, accuracy, and average precision, with cross-validated mean AUC and 95% CI reported. Exploratory and interaction analyses Restricted cubic spline regression was applied to assess non-linear associations. A partial dependence analysis illustrated the joint effect of HRV and MVPA, demonstrating the highest predicted risk among individuals with both low HRV (< 45 ms) and low MVPA (< 40 min/day). All tests were two-tailed, and a p < 0.05 was considered statistically significant. Results Baseline characteristics A total of 230 participants (mean age 45.8 ± 7.6 years, 52.6% women) completed the 12-month follow-up. During the study period, 28 individuals (12.2%) developed incident hypertension according to the ESC/ESH 2023 criteria. Participants who developed hypertension were older (48.7 ± 6.9 vs. 45.3 ± 7.4 years, p = 0.014), had a higher baseline BMI (28.2 ± 3.7 vs. 26.4 ± 3.1 kg/m², p = 0.009), and demonstrated lower mean HRV values (42.6 ± 10.8 vs. 56.1 ± 12.9 ms, p < 0.001) compared with normotensive subjects. They also showed significantly lower daily MVPA (38.4 ± 14.2 vs. 57.9 ± 16.7 min/day, p < 0.001) and reduced VO₂max (34.7 ± 5.1 vs. 39.8 ± 5.8 mL/kg/min, p = 0.002). Other parameters, including resting heart rate and sleep efficiency, showed non-significant trends toward higher risk in hypertensive converters. Table 1 summarizes baseline characteristics of participants according to hypertension status. Table 1 Baseline characteristics of participants according to incident hypertension status Variable Total (n = 230) No Hypertension (n = 202) Incident Hypertension (n = 28) p value Age (years) 45.8 ± 7.6 45.3 ± 7.4 48.7 ± 6.9 0.014 Female sex, n (%) 121 (52.6) 108 (53.5) 13 (46.4) 0.49 Body mass index (kg/m²) 26.6 ± 3.3 26.4 ± 3.1 28.2 ± 3.7 0.009 Resting heart rate (bpm) 72.4 ± 7.9 72.1 ± 7.7 74.0 ± 8.8 0.22 Heart rate variability (HRV, RMSSD, ms) 54.3 ± 13.5 56.1 ± 12.9 42.6 ± 10.8 < 0.001 Moderate-to-vigorous physical activity (MVPA, min/day) 55.6 ± 17.9 57.9 ± 16.7 38.4 ± 14.2 < 0.001 Sedentary time (min/day) 540 ± 88 532 ± 84 598 ± 90 0.006 Estimated VO₂max (ml/kg/min) 39.2 ± 5.9 39.8 ± 5.8 34.7 ± 5.1 0.002 Sleep efficiency (%) 87.4 ± 4.8 87.6 ± 4.6 86.1 ± 5.5 0.19 Stress index (0–100) 45.1 ± 9.8 44.7 ± 9.6 48.0 ± 10.1 0.11 Smokers, n (%) 47 (20.4) 39 (19.3) 8 (28.6) 0.26 Follow-up duration (months) 12.0 ± 0.2 12.0 ± 0.2 12.0 ± 0.2 0.89 Multivariable logistic regression In multivariable analysis, lower HRV (per 10-ms decrease, adjusted OR = 1.31, 95% CI 1.10–1.58, p = 0.003), lower MVPA (per 10-min decrease, OR = 1.18, 95% CI 1.04–1.36, p = 0.009), and higher BMI (per 1 kg/m² increase, OR = 1.12, 95% CI 1.01–1.26, p = 0.041) were independently associated with incident hypertension. The interaction term HRV × MVPA was statistically significant ( p = 0.012), indicating a synergistic effect between autonomic balance and physical activity. The regression model showed good calibration (Hosmer–Lemeshow p = 0.64) and discrimination (AUC = 0.82, 95% CI 0.75–0.89). Machine learning model performance Using the same feature set, the XGBoost model achieved an average cross-validated ROC – AUC of 0.83 (95% CI 0.76–0.90 ) , accuracy of 0.81, and F1-score of 0.74. The LightGBM model yielded comparable discrimination (AUC = 0.82, accuracy = 0.79). Both models demonstrated consistent calibration, with no evidence of overfitting across folds. The combined smartwatch + clinical model significantly outperformed the clinical-only baseline model (AUC = 0.83 vs. 0.68, p < 0.001 by DeLong test). Figure 1 displays ROC curves comparing the two models. Figure 2 presents the SHAP-based feature importance ranking, showing HRV RHR, and MVPA as the strongest contributors. Exploratory and interaction analyses Partial-dependence analysis revealed a clear non-linear relationship between HRV and MVPA (Fig. 3). Participants with both low HRV (< 45 ms) and low MVPA ( 40%), whereas those maintaining high HRV (> 60 ms) and MVPA (> 60 min/day) had a risk below 10%. No significant sex-based or device-type interactions were observed. Summary of key findings Incident hypertension occurred in 12.2% of previously normotensive adults over 12 months. Reduced HRV, lower MVPA, and higher BMI were independent predictors. Smartwatch-derived parameters substantially improved discrimination compared to clinical data alone. Machine-learning models confirmed the robustness of the physiological markers and their interaction effects. Discussion In this prospective observational study, we found that smartwatch-derived exercise and autonomic parameters, particularly HRV and MVPA, were significant predictors of incident hypertension in normotensive adults during one year of follow-up. Individuals who developed hypertension demonstrated lower HRV, reduced physical activity, and increased sedentary time at baseline compared with those who remained normotensive. The integration of these wearable-derived indices with conventional clinical variables markedly improved predictive accuracy, highlighting the potential role of smartwatch-based metrics in early hypertension detection. Our findings align with previous research suggesting that autonomic dysfunction and physical inactivity are among the earliest physiologic markers of blood pressure dysregulation. Reduced HRV, reflecting impaired parasympathetic modulation and sympathetic predominance, has been consistently associated with the development of hypertension. The Framingham Heart Study showed that diminished HRV independently predicted new-onset hypertension, even after adjusting for age and baseline blood pressure [ 4 ]. Similarly, Diaz and colleagues demonstrated that reduced HRV and elevated resting heart rate increased the risk of hypertension in a large African-American cohort [ 3 ]. These studies, along with our results, support the hypothesis that altered autonomic control precedes the clinical manifestation of sustained hypertension. The inverse association between physical activity and hypertension risk observed in our cohort further reinforces the importance of lifestyle behaviors in cardiovascular regulation. Regular aerobic activity improves endothelial function, reduces arterial stiffness, and enhances baroreflex sensitivity, mechanisms that collectively lower blood pressure [ 10 ]. Participants with higher MVPA in our study exhibited better autonomic profiles and lower incident hypertension rates, suggesting a synergistic relationship between physical activity and autonomic function. The significant HRV×MVPA interaction term found in our regression model underscores this point, indicating that low HRV and low physical activity jointly identify a high-risk phenotype for hypertension development. These findings align with current guidelines emphasizing physical activity as a first-line preventive measure for blood pressure control [ 11 , 12 ]. An important contribution of this study is the demonstration that real-world, continuously recorded smartwatch data can provide reliable physiological information relevant to cardiovascular risk. Prior research has validated the accuracy of smartwatch-derived HRV and heart rate metrics compared with gold-standard ECG and Holter recordings [ 13 , 14 ]. While cuffless blood pressure estimation remains limited in precision, the use of surrogate physiologic markers such as HRV and MVPA circumvents calibration challenges and offers more stable longitudinal trends [ 15 ]. The present findings therefore highlight a pragmatic approach to hypertension risk assessment that relies on metrics already available in most commercially used wearable devices. From a methodological standpoint, the consistency between logistic regression and machine learning results strengthens the robustness of our conclusions. Both XGBoost and LightGBM models identified HRV, resting heart rate, and MVPA as the most influential predictors of incident hypertension, with cross-validated AUC values exceeding 0.82. These models’ calibration performance suggests that complex, non-linear relationships within smartwatch-derived data can be effectively leveraged to identify individuals at increased risk. Importantly, these algorithms do not simply replicate conventional statistics but uncover subtle physiological interactions, supporting their future role in precision cardiovascular screening. Clinically, these findings propose a paradigm shift from episodic blood pressure measurement to continuous physiological monitoring for early identification of at-risk individuals. Integrating wearable-derived autonomic and activity metrics into preventive cardiology could enable proactive lifestyle counseling and closer follow-up before the establishment of fixed hypertension. Such approaches could be particularly beneficial for younger adults, in whom hypertension is often unrecognized until later stages. Furthermore, the scalability and affordability of consumer smartwatches make this strategy feasible for population-level implementation. Nonetheless, several limitations should be acknowledged. Despite careful device calibration, smartwatch-derived HRV and activity data are susceptible to environmental influences and algorithmic variability. The study cohort consisted of health-conscious volunteers, which may limit generalizability. Hypertension diagnosis relied on office-based measurements rather than continuous ambulatory monitoring, and biochemical correlates such as inflammatory or neurohormonal markers were not assessed. Additionally, while machine learning improved discrimination, the observational design precludes causal inference. Future studies should include larger, multi-ethnic populations, standardized device validation, and integration with vascular and metabolic biomarkers to enhance mechanistic understanding. In conclusion, smartwatch-derived HRV and MVPA were independent predictors of early hypertension in normotensive adults, and their combination with clinical parameters substantially improved predictive accuracy. These findings underscore the potential of wearable devices as noninvasive, scalable tools for early detection and prevention of hypertension in real-life settings. As wearable technology becomes increasingly integrated into daily health monitoring, its application in cardiovascular risk stratification may transform hypertension prevention from reactive management to proactive digital surveillance. Funding Declaration Declarations Funding Declaration Funding: No financial support was received for this study. Clinical Trial Number Clinical trial number: Not applicable. Author Contribution G.C.: Conceptualization, study design, data collection, data analysis, interpretation of results, manuscript drafting and critical revision.E.S. A.C.: Data collection, clinical evaluation, data interpretation, literature review, manuscript review and editing.All authors read and approved the final manuscript. Data Availability The data that support the findings of this study are available from Atatürk University Health Practice and Research Center (Atatürk Üniversitesi Sağlık Uygulama ve Araştırma Merkezi). However, restrictions apply to the availability of these data, which were accessed under institutional licence for the present study and are therefore not publicly available. The data may be obtained upon reasonable request and with permission from the Atatürk University Health Practice and Research Center. References Schroeder EB, Liao D, Chambless LE, Prineas RJ, Evans GW, Heiss G. Hypertension, blood pressure, and heart rate variability: the Atherosclerosis Risk in Communities (ARIC) study. Hypertension. 2003;42(6):1106–11. Kang J, Chang Y, Kim Y, Shin H, Ryu S. Ten-Second Heart Rate Variability, Its Changes Over Time, and the Development of Hypertension. Hypertension. 2022;79(6):1308–18. Diaz KM, Booth JN 3rd, Seals SR, Abdalla M, Dubbert PM, Sims M, et al. Physical Activity and Incident Hypertension in African Americans: The Jackson Heart Study. Hypertension. 2017;69(3):421–7. Singh JP, Larson MG, Tsuji H, Evans JC, O'Donnell CJ, Levy D. Reduced heart rate variability and new-onset hypertension: Insights from the Framingham Heart Study. Hypertension. 1998;32(2):293–7. Seals DR, Desouza CA, Donato AJ, Tanaka H. Habitual exercise and arterial aging. J Appl Physiol (1985). 2008;105(4):1323–32. Grassi G, Seravalle G, Quarti-Trevano F. The 'neuroadrenergic hypothesis' in hypertension: Current evidence. Exp Physiol. 2010;95(5):581–6. Shcherbina A, Mattsson CM, Waggott D, Salisbury H, Christle JW, Hastie T, et al. Accuracy in wrist-worn, sensor-based measurements of heart rate and energy expenditure in a diverse cohort. J Pers Med. 2017;7(2):3. Bent B, Goldstein BA, Kibbe WA, Dunn JP. Investigating sources of inaccuracy in wearable optical heart rate sensors. NPJ Digit Med. 2020;3:18. Boudreaux BD, Hebert EP, Hollander DB, Williams BM, Cormier CL, Naquin MR, et al. Validity of wearable activity monitors during cycling and resistance exercise. Med Sci Sports Exerc. 2018;50(3):624–33. Green DJ, Hopman MT, Padilla J, Laughlin MH, Thijssen DH. Vascular adaptation to exercise in humans: Role of hemodynamic stimuli. Physiol Rev. 2017;97(2):495–528. Kang J, Chang Y, Kim Y, Shin H, Ryu S. Ten-Second Heart Rate Variability, Its Changes Over Time, and the Development of Hypertension. Hypertension. 2022;79(6):1308–18. Williams B, Mancia G, Spiering W, Agabiti Rosei E, Azizi M, Burnier M, et al. 2018 ESC/ESH Guidelines for the management of arterial hypertension. Eur Heart J. 2018;39(33):3021–104. Bent B, Goldstein BA, Kibbe WA, Dunn JP. Investigating sources of inaccuracy in wearable optical heart rate sensors. NPJ Digit Med. 2020;3:18. Hernando D, Roca S, Sancho J, Alesanco Á, Bailón R. Validation of the Apple Watch for heart rate variability measurements during relax and mental stress in healthy subjects. Sens (Basel). 2018;18(8):2619. Mukkamala R, Stergiou GS, Avolio AP. Cuffless Blood Pressure Measurement. Annu Rev Biomed Eng. 2022;6(24):203–30. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 06 Jan, 2026 Reviews received at journal 05 Jan, 2026 Reviews received at journal 29 Dec, 2025 Reviewers agreed at journal 26 Dec, 2025 Reviews received at journal 24 Dec, 2025 Reviewers agreed at journal 22 Dec, 2025 Reviewers agreed at journal 21 Dec, 2025 Reviewers invited by journal 19 Dec, 2025 Editor invited by journal 11 Dec, 2025 Editor assigned by journal 08 Dec, 2025 Submission checks completed at journal 08 Dec, 2025 First submitted to journal 29 Nov, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8238959","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":564341672,"identity":"b56b1037-02c4-4ade-b390-8213434dc8a9","order_by":0,"name":"Gökhan Ceyhun","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7UlEQVRIiWNgGAWjYFACNjYILQHEH0B8dlK0MM4A8ZlJ0cLMA2IQ0iLv3pb24OcOOzmD2z1mj21+bZPnY2Zg/PAxB7cWwzPHjhv2nkk2Nrhzxtw4t++2YRszA7PkzG14tMxIb5PgbTuQuOFGjpl0bs9tRqAWNmZeAlok/7YdqAdrsey5bU9Qi7xE2jFpoC0JBiAtDD9uJxLUYsBzLE1ati3ZcOaNtDLJ3obbyW3MjM14/SLf3mYm+bbNTp7vRvI2iR9/btvOb28++OEjPlsOQBkKIAZjG4jJ2IBbPciWBhTGH7yKR8EoGAWjYIQCAFmTUDkKs+eSAAAAAElFTkSuQmCC","orcid":"","institution":"Ataturk University Faculty of Medicine, Department of Cardiology","correspondingAuthor":true,"prefix":"","firstName":"Gökhan","middleName":"","lastName":"Ceyhun","suffix":""},{"id":564341683,"identity":"74cb4383-3345-4505-9b5b-8d3e0db1f8d0","order_by":1,"name":"Esma Selva Ateş Ceyhun","email":"","orcid":"","institution":"Department of Cardiology, Erzurum City Hospital","correspondingAuthor":false,"prefix":"","firstName":"Esma","middleName":"Selva Ateş","lastName":"Ceyhun","suffix":""}],"badges":[],"createdAt":"2025-11-29 19:38:15","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8238959/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8238959/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":99319006,"identity":"45faff43-f4ae-4271-9009-d8dce5b77785","added_by":"auto","created_at":"2025-12-31 16:35:59","extension":"png","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":244480,"visible":true,"origin":"","legend":"","description":"","filename":"figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-8238959/v1/4c2df7ea8ac32ae4dcb0abeb.png"},{"id":99317911,"identity":"fdf8c1be-65b9-4d02-875e-ba62acfc8604","added_by":"auto","created_at":"2025-12-31 16:30:56","extension":"png","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":152080,"visible":true,"origin":"","legend":"","description":"","filename":"figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8238959/v1/d8248409d1500dd16ef7c7e6.png"},{"id":99190551,"identity":"6d056cb9-4f70-444e-bb6c-7347c4ccee04","added_by":"auto","created_at":"2025-12-30 00:51:48","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":25277,"visible":true,"origin":"","legend":"","description":"","filename":"Maindocument.docx","url":"https://assets-eu.researchsquare.com/files/rs-8238959/v1/f480b4c44de6d0ffbc9b3589.docx"},{"id":99317317,"identity":"68b6d4a6-19c1-42f6-8ca9-36d6cb783d43","added_by":"auto","created_at":"2025-12-31 16:30:00","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":26963,"visible":true,"origin":"","legend":"","description":"","filename":"Table1.docx","url":"https://assets-eu.researchsquare.com/files/rs-8238959/v1/143b711e2fa81446450ccbcb.docx"},{"id":99190559,"identity":"87088f43-dc09-4005-b89e-a6ba82ff1627","added_by":"auto","created_at":"2025-12-30 00:51:48","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":216907,"visible":true,"origin":"","legend":"","description":"","filename":"figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-8238959/v1/954f1c68ff92d935c74dc87f.png"},{"id":99318873,"identity":"ac4c2f1d-19d7-4a0a-8373-b95dd03bcc54","added_by":"auto","created_at":"2025-12-31 16:35:32","extension":"json","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4324,"visible":true,"origin":"","legend":"","description":"","filename":"ec4bcf1fcf99428c94c50fa9fa01cbee.json","url":"https://assets-eu.researchsquare.com/files/rs-8238959/v1/52d937ce39167e81a3171709.json"},{"id":99190555,"identity":"59fa1dab-8e58-4494-9b6d-6b9b9f1a37a1","added_by":"auto","created_at":"2025-12-30 00:51:48","extension":"xml","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":58020,"visible":true,"origin":"","legend":"","description":"","filename":"ec4bcf1fcf99428c94c50fa9fa01cbee1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8238959/v1/64d5765f2c365eff517e8a1e.xml"},{"id":99190557,"identity":"d47c3854-cfec-410b-bd6d-534d58c41bd6","added_by":"auto","created_at":"2025-12-30 00:51:48","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":244480,"visible":true,"origin":"","legend":"","description":"","filename":"figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-8238959/v1/ed68785545971252760657f9.png"},{"id":99190560,"identity":"87820dba-5dbc-4c11-a836-9e0b1d57ea9d","added_by":"auto","created_at":"2025-12-30 00:51:49","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":152080,"visible":true,"origin":"","legend":"","description":"","filename":"figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8238959/v1/64856edbeaef3fc6541107ce.png"},{"id":99316728,"identity":"7fa0d03f-b7ef-43d3-9a92-5fb9100c4d78","added_by":"auto","created_at":"2025-12-31 16:29:05","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":216907,"visible":true,"origin":"","legend":"","description":"","filename":"figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-8238959/v1/fcfe8d4ba79e32c3e2a9d0e5.png"},{"id":99190558,"identity":"26078c2a-ed31-4080-9031-dc29593fe67c","added_by":"auto","created_at":"2025-12-30 00:51:48","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":84481,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefigure1.png","url":"https://assets-eu.researchsquare.com/files/rs-8238959/v1/4aa62c7acb8d97b6a8886d9a.png"},{"id":99190562,"identity":"bc8aee99-c0d1-4b19-bd4d-ddb5d9407845","added_by":"auto","created_at":"2025-12-30 00:51:49","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":69497,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefigure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8238959/v1/77e4c4b4d6622c1a4777fab5.png"},{"id":99190561,"identity":"2d03ab9d-72bd-491e-992c-e2739f47ab83","added_by":"auto","created_at":"2025-12-30 00:51:49","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":57194,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefigure3.png","url":"https://assets-eu.researchsquare.com/files/rs-8238959/v1/49a840e8716b008561005ff6.png"},{"id":99190565,"identity":"c6f7bc17-0bdc-4c21-8702-340e6f98003c","added_by":"auto","created_at":"2025-12-30 00:51:49","extension":"xml","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":54688,"visible":true,"origin":"","legend":"","description":"","filename":"ec4bcf1fcf99428c94c50fa9fa01cbee1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8238959/v1/88a70654ad0a25b72d2f46e1.xml"},{"id":99190563,"identity":"cdf174ab-3305-4086-972f-607574bb3843","added_by":"auto","created_at":"2025-12-30 00:51:49","extension":"html","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":64160,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8238959/v1/75149eec4ce854c35f062135.html"},{"id":99190548,"identity":"7940f492-02cf-469f-a917-0942a941c726","added_by":"auto","created_at":"2025-12-30 00:51:48","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":244480,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-8238959/v1/5ff40a90e399f67c91f3cc7b.png"},{"id":99190553,"identity":"e1394989-f705-427b-ae7d-02806e4414f7","added_by":"auto","created_at":"2025-12-30 00:51:48","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":152080,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8238959/v1/57a5b97b5872e2a017beb87f.png"},{"id":99190552,"identity":"1e3e390c-9d51-49dd-91f0-a4a7aad91353","added_by":"auto","created_at":"2025-12-30 00:51:48","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":216907,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-8238959/v1/4e3b4cd53b9832e9c2deb2d5.png"},{"id":99323808,"identity":"2859ee8e-2e7b-4179-aed5-22edc244998c","added_by":"auto","created_at":"2025-12-31 16:46:19","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1127727,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8238959/v1/75a81653-21b2-4a57-b311-a1b4321c4475.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Smartwatch-Derived Exercise Metrics as Predictors of Early Hypertension: A Prospective Observational Study","fulltext":[{"header":"Introduction","content":"\u003cp\u003eHypertension is a leading modifiable risk factor for cardiovascular disease and premature mortality worldwide. Despite extensive public health efforts, nearly half of hypertensive individuals remain undiagnosed or untreated, highlighting the importance of early detection strategies. Conventional office-based blood pressure assessments often fail to capture dynamic physiological changes that precede the onset of sustained hypertension, particularly those related to autonomic dysfunction and physical inactivity.\u003c/p\u003e \u003cp\u003eRecent technological advances have enabled smartwatches to collect continuous cardiovascular and behavioral data under real-life conditions. Beyond heart rate tracking, these devices can measure heart rate variability (HRV), resting heart rate (RHR), and physical activity metrics such as moderate-to-vigorous physical activity (MVPA), VO₂max, and sedentary time, offering valuable insight into cardiovascular regulation. Reduced HRV has been associated with sympathetic overactivity and early vascular changes that predispose to hypertension, while low levels of MVPA contribute to endothelial dysfunction and elevated resting blood pressure[\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eAccumulating evidence suggests that HRV and daily activity levels may jointly reflect the autonomic\u0026ndash;metabolic interplay underlying blood pressure regulation. Prospective studies have shown that lower HRV predicts incident hypertension in normotensive adults[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e], and that regular aerobic activity reduces sympathetic tone and arterial stiffness, thus mitigating long-term hypertension risk[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eHowever, few studies have integrated multiple smartwatch-derived parameters into a single predictive framework, and most existing data rely on short-term or cross-sectional analyses.\u003c/p\u003e \u003cp\u003eRecent validation studies confirm the technical accuracy of smartwatch-based cardiovascular measurements[\u003cspan additionalcitationids=\"CR8\" citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e], yet the clinical relevance of combining exercise and autonomic metrics to detect early hypertension remains largely unexplored. The integration of these digital biomarkers could provide a practical, noninvasive approach for identifying high-risk individuals before overt hypertension develops.\u003c/p\u003e \u003cp\u003eTherefore, the present prospective observational study aimed to assess whether smartwatch-derived exercise and autonomic parameters can predict incident hypertension during 12 months of follow-up in normotensive adults. We hypothesized that decreased HRV, reduced MVPA, and their interaction would independently predict early hypertension and that combining these smartwatch metrics with clinical variables would enhance predictive performance compared with conventional models.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eStudy design and population\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis prospective observational study was conducted between January 2024 and January 2025 to evaluate whether smartwatch-derived exercise and autonomic parameters could predict incident hypertension among normotensive adults. A total of 230 participants aged 30–60 years were enrolled from outpatient cardiology and preventive medicine clinics. All participants provided written informed consent, and the study protocol was approved by the institutional ethics committee (Approval No: B.30.2.ATA.0.01.00/737) in accordance with the Declaration of Helsinki.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eInclusion criteria:\u003c/strong\u003e age 30–60 years, baseline systolic BP \u0026lt; 140 mmHg and diastolic BP \u0026lt; 90 mmHg, continuous smartwatch use ≥ 20 days per month during follow-up, willingness to attend scheduled visits, and provision of written informed consent.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eExclusion criteria:\u003c/strong\u003e known hypertension, coronary artery disease, heart failure or arrhythmia (including atrial fibrillation), current use of antihypertensive, antiarrhythmic, β-blocker, calcium-channel blocker, or ACE inhibitor therapy within 3 months before enrollment, diabetes mellitus, chronic kidney disease (eGFR \u0026lt; 60 mL/min/1.73 m²), thyroid dysfunction, autonomic neuropathy, BMI ≥ 35 kg/m², heavy smoking (\u0026gt; 20 cigarettes/day) or alcohol consumption (\u0026gt; 20 g/day), shift work or irregular sleep–wake patterns, sleep apnea syndrome, missing or incomplete smartwatch data (\u0026gt; 15% missing), technical incompatibility preventing data synchronization, pregnancy, lactation, hormonal therapy, or inability/unwillingness to attend the 12-month follow-up visit.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSmartwatch-derived parameters\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eParticipants wore a validated smartwatch (Apple Watch Series 9 or equivalent) continuously for 12 months. The following parameters were automatically recorded and extracted through the device’s health API: resting heart rate (RHR, bpm), heart rate variability (HRV, RMSSD, ms), moderate-to-vigorous physical activity (MVPA, min/day), sedentary time (min/day), estimated VO₂max (ml/kg/min), sleep efficiency (% of time asleep/time in bed), and stress index (0–100). Monthly mean values and 30-day changes (ΔHRV, ΔRHR, ΔMVPA) were computed. Interaction indices (HRV×MVPA, RHR×Sedentary time) were derived to explore autonomic–activity relationships.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eClinical evaluation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBlood pressure (BP) was measured at baseline and at 12 months using a calibrated oscillometric device after 5 minutes of rest. The mean of the last two of three readings was recorded. Incident hypertension was defined according to the 2023 ESC/ESH guidelines as systolic BP ≥ 140 mmHg and/or diastolic BP ≥ 90 mmHg, or initiation of antihypertensive therapy during follow-up.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll analyses were performed using Python 3.12 (\u003cem\u003epandas\u003c/em\u003e, \u003cem\u003estatsmodels\u003c/em\u003e, \u003cem\u003exgboost\u003c/em\u003e, \u003cem\u003elightgbm\u003c/em\u003e, \u003cem\u003eshap\u003c/em\u003e). Continuous variables were expressed as mean ± SD and categorical variables as percentages. Normality was assessed using the Shapiro–Wilk test. Group comparisons were made with the Student’s t-test or Mann–Whitney U test for continuous variables, and the χ² or Fisher’s exact test for categorical data.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMultivariable logistic regression\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIndependent predictors of incident hypertension were identified using multivariable logistic regression including age, sex, BMI, HRV, RHR, MVPA, sedentary time, VO₂max, sleep efficiency, and interaction terms (HRV×MVPA, RHR×Sedentary). Adjusted odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. Model calibration was assessed using the Hosmer–Lemeshow test, and discrimination using the area under the ROC curve (AUC).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMachine learning analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo complement traditional inference, predictive modeling was performed using Extreme Gradient Boosting (XGBoost) and Light Gradient Boosting Machine (LightGBM) with 10-fold stratified cross-validation. Feature importance was determined via SHAP (Shapley Additive Explanations) analysis. Performance metrics included ROC–AUC, F1-score, accuracy, and average precision, with cross-validated mean AUC and 95% CI reported.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eExploratory and interaction analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRestricted cubic spline regression was applied to assess non-linear associations. A partial dependence analysis illustrated the joint effect of HRV and MVPA, demonstrating the highest predicted risk among individuals with both low HRV (\u0026lt; 45 ms) and low MVPA (\u0026lt; 40 min/day). All tests were two-tailed, and a \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05 was considered statistically significant.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eBaseline characteristics\u003c/h2\u003e \u003cp\u003eA total of 230 participants (mean age 45.8\u0026thinsp;\u0026plusmn;\u0026thinsp;7.6 years, 52.6% women) completed the 12-month follow-up. During the study period, 28 individuals (12.2%) developed incident hypertension according to the ESC/ESH 2023 criteria. Participants who developed hypertension were older (48.7\u0026thinsp;\u0026plusmn;\u0026thinsp;6.9 vs. 45.3\u0026thinsp;\u0026plusmn;\u0026thinsp;7.4 years, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.014), had a higher baseline BMI (28.2\u0026thinsp;\u0026plusmn;\u0026thinsp;3.7 vs. 26.4\u0026thinsp;\u0026plusmn;\u0026thinsp;3.1 kg/m\u0026sup2;, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.009), and demonstrated lower mean HRV values (42.6\u0026thinsp;\u0026plusmn;\u0026thinsp;10.8 vs. 56.1\u0026thinsp;\u0026plusmn;\u0026thinsp;12.9 ms, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) compared with normotensive subjects.\u003c/p\u003e \u003cp\u003eThey also showed significantly lower daily MVPA (38.4\u0026thinsp;\u0026plusmn;\u0026thinsp;14.2 vs. 57.9\u0026thinsp;\u0026plusmn;\u0026thinsp;16.7 min/day, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and reduced VO₂max (34.7\u0026thinsp;\u0026plusmn;\u0026thinsp;5.1 vs. 39.8\u0026thinsp;\u0026plusmn;\u0026thinsp;5.8 mL/kg/min, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.002).\u003c/p\u003e \u003cp\u003eOther parameters, including resting heart rate and sleep efficiency, showed non-significant trends toward higher risk in hypertensive converters. Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e summarizes baseline characteristics of participants according to hypertension status.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cem\u003eBaseline characteristics of participants according to incident hypertension status\u003c/em\u003e\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTotal (n\u0026thinsp;=\u0026thinsp;230)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNo Hypertension (n\u0026thinsp;=\u0026thinsp;202)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eIncident Hypertension (n\u0026thinsp;=\u0026thinsp;28)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ep value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge (years)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e45.8\u0026thinsp;\u0026plusmn;\u0026thinsp;7.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e45.3\u0026thinsp;\u0026plusmn;\u0026thinsp;7.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e48.7\u0026thinsp;\u0026plusmn;\u0026thinsp;6.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.014\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale sex, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e121 (52.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e108 (53.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e13 (46.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.49\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBody mass index (kg/m\u0026sup2;)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e26.6\u0026thinsp;\u0026plusmn;\u0026thinsp;3.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e26.4\u0026thinsp;\u0026plusmn;\u0026thinsp;3.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e28.2\u0026thinsp;\u0026plusmn;\u0026thinsp;3.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.009\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eResting heart rate (bpm)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e72.4\u0026thinsp;\u0026plusmn;\u0026thinsp;7.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e72.1\u0026thinsp;\u0026plusmn;\u0026thinsp;7.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e74.0\u0026thinsp;\u0026plusmn;\u0026thinsp;8.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.22\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHeart rate variability (HRV, RMSSD, ms)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e54.3\u0026thinsp;\u0026plusmn;\u0026thinsp;13.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e56.1\u0026thinsp;\u0026plusmn;\u0026thinsp;12.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e42.6\u0026thinsp;\u0026plusmn;\u0026thinsp;10.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModerate-to-vigorous physical activity (MVPA, min/day)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e55.6\u0026thinsp;\u0026plusmn;\u0026thinsp;17.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e57.9\u0026thinsp;\u0026plusmn;\u0026thinsp;16.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e38.4\u0026thinsp;\u0026plusmn;\u0026thinsp;14.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSedentary time (min/day)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e540\u0026thinsp;\u0026plusmn;\u0026thinsp;88\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e532\u0026thinsp;\u0026plusmn;\u0026thinsp;84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e598\u0026thinsp;\u0026plusmn;\u0026thinsp;90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.006\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEstimated VO₂max (ml/kg/min)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e39.2\u0026thinsp;\u0026plusmn;\u0026thinsp;5.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e39.8\u0026thinsp;\u0026plusmn;\u0026thinsp;5.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e34.7\u0026thinsp;\u0026plusmn;\u0026thinsp;5.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSleep efficiency (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e87.4\u0026thinsp;\u0026plusmn;\u0026thinsp;4.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e87.6\u0026thinsp;\u0026plusmn;\u0026thinsp;4.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e86.1\u0026thinsp;\u0026plusmn;\u0026thinsp;5.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.19\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eStress index (0\u0026ndash;100)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e45.1\u0026thinsp;\u0026plusmn;\u0026thinsp;9.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e44.7\u0026thinsp;\u0026plusmn;\u0026thinsp;9.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e48.0\u0026thinsp;\u0026plusmn;\u0026thinsp;10.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.11\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSmokers, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e47 (20.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e39 (19.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8 (28.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.26\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFollow-up duration (months)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e12.0\u0026thinsp;\u0026plusmn;\u0026thinsp;0.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12.0\u0026thinsp;\u0026plusmn;\u0026thinsp;0.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12.0\u0026thinsp;\u0026plusmn;\u0026thinsp;0.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.89\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eMultivariable logistic regression\u003c/h2\u003e \u003cp\u003eIn multivariable analysis, lower HRV (per 10-ms decrease, adjusted OR\u0026thinsp;=\u0026thinsp;1.31, 95% CI 1.10\u0026ndash;1.58, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.003), lower MVPA (per 10-min decrease, OR\u0026thinsp;=\u0026thinsp;1.18, 95% CI 1.04\u0026ndash;1.36, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.009), and higher BMI (per 1 kg/m\u0026sup2; increase, OR\u0026thinsp;=\u0026thinsp;1.12, 95% CI 1.01\u0026ndash;1.26, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.041) were independently associated with incident hypertension. The interaction term HRV \u0026times; MVPA was statistically significant (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.012), indicating a synergistic effect between autonomic balance and physical activity. The regression model showed good calibration (Hosmer\u0026ndash;Lemeshow \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.64) and discrimination (AUC\u0026thinsp;=\u0026thinsp;0.82, 95% CI 0.75\u0026ndash;0.89).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eMachine learning model performance\u003c/h2\u003e \u003cp\u003eUsing the same feature set, the XGBoost model achieved an average cross-validated ROC\u003cb\u003e\u0026ndash;\u003c/b\u003eAUC of 0.83 (95% CI 0.76\u0026ndash;0.90\u003cb\u003e)\u003c/b\u003e, accuracy of 0.81, and F1-score of 0.74. The LightGBM model yielded comparable discrimination (AUC\u0026thinsp;=\u0026thinsp;0.82, accuracy\u0026thinsp;=\u0026thinsp;0.79). Both models demonstrated consistent calibration, with no evidence of overfitting across folds. The combined smartwatch\u0026thinsp;+\u0026thinsp;clinical model significantly outperformed the clinical-only baseline model (AUC\u0026thinsp;=\u0026thinsp;0.83 vs. 0.68, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001 by DeLong test). Figure\u0026nbsp;1 displays ROC curves comparing the two models. Figure\u0026nbsp;2 presents the SHAP-based feature importance ranking, showing HRV RHR, and MVPA as the strongest contributors.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eExploratory and interaction analyses\u003c/h2\u003e \u003cp\u003ePartial-dependence analysis revealed a clear non-linear relationship between HRV and MVPA (Fig.\u0026nbsp;3). Participants with both low HRV (\u0026lt;\u0026thinsp;45 ms) and low MVPA (\u0026lt;\u0026thinsp;40 min/day) exhibited the highest predicted risk (\u0026gt;\u0026thinsp;40%), whereas those maintaining high HRV (\u0026gt;\u0026thinsp;60 ms) and MVPA (\u0026gt;\u0026thinsp;60 min/day) had a risk below 10%. No significant sex-based or device-type interactions were observed.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eSummary of key findings\u003c/h2\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eIncident hypertension occurred in \u003cb\u003e12.2%\u003c/b\u003e of previously normotensive adults over 12 months.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eReduced HRV, lower MVPA, and higher BMI were independent predictors.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eSmartwatch-derived parameters substantially improved discrimination compared to clinical data alone.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eMachine-learning models confirmed the robustness of the physiological markers and their interaction effects.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this prospective observational study, we found that smartwatch-derived exercise and autonomic parameters, particularly HRV and MVPA, were significant predictors of incident hypertension in normotensive adults during one year of follow-up. Individuals who developed hypertension demonstrated lower HRV, reduced physical activity, and increased sedentary time at baseline compared with those who remained normotensive. The integration of these wearable-derived indices with conventional clinical variables markedly improved predictive accuracy, highlighting the potential role of smartwatch-based metrics in early hypertension detection.\u003c/p\u003e \u003cp\u003eOur findings align with previous research suggesting that autonomic dysfunction and physical inactivity are among the earliest physiologic markers of blood pressure dysregulation. Reduced HRV, reflecting impaired parasympathetic modulation and sympathetic predominance, has been consistently associated with the development of hypertension. The Framingham Heart Study showed that diminished HRV independently predicted new-onset hypertension, even after adjusting for age and baseline blood pressure [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Similarly, Diaz and colleagues demonstrated that reduced HRV and elevated resting heart rate increased the risk of hypertension in a large African-American cohort [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. These studies, along with our results, support the hypothesis that altered autonomic control precedes the clinical manifestation of sustained hypertension.\u003c/p\u003e \u003cp\u003eThe inverse association between physical activity and hypertension risk observed in our cohort further reinforces the importance of lifestyle behaviors in cardiovascular regulation. Regular aerobic activity improves endothelial function, reduces arterial stiffness, and enhances baroreflex sensitivity, mechanisms that collectively lower blood pressure [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Participants with higher MVPA in our study exhibited better autonomic profiles and lower incident hypertension rates, suggesting a synergistic relationship between physical activity and autonomic function. The significant HRV\u0026times;MVPA interaction term found in our regression model underscores this point, indicating that low HRV and low physical activity jointly identify a high-risk phenotype for hypertension development. These findings align with current guidelines emphasizing physical activity as a first-line preventive measure for blood pressure control [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eAn important contribution of this study is the demonstration that real-world, continuously recorded smartwatch data can provide reliable physiological information relevant to cardiovascular risk. Prior research has validated the accuracy of smartwatch-derived HRV and heart rate metrics compared with gold-standard ECG and Holter recordings [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. While cuffless blood pressure estimation remains limited in precision, the use of surrogate physiologic markers such as HRV and MVPA circumvents calibration challenges and offers more stable longitudinal trends [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. The present findings therefore highlight a pragmatic approach to hypertension risk assessment that relies on metrics already available in most commercially used wearable devices.\u003c/p\u003e \u003cp\u003eFrom a methodological standpoint, the consistency between logistic regression and machine learning results strengthens the robustness of our conclusions. Both XGBoost and LightGBM models identified HRV, resting heart rate, and MVPA as the most influential predictors of incident hypertension, with cross-validated AUC values exceeding 0.82. These models\u0026rsquo; calibration performance suggests that complex, non-linear relationships within smartwatch-derived data can be effectively leveraged to identify individuals at increased risk. Importantly, these algorithms do not simply replicate conventional statistics but uncover subtle physiological interactions, supporting their future role in precision cardiovascular screening.\u003c/p\u003e \u003cp\u003eClinically, these findings propose a paradigm shift from episodic blood pressure measurement to continuous physiological monitoring for early identification of at-risk individuals. Integrating wearable-derived autonomic and activity metrics into preventive cardiology could enable proactive lifestyle counseling and closer follow-up before the establishment of fixed hypertension. Such approaches could be particularly beneficial for younger adults, in whom hypertension is often unrecognized until later stages. Furthermore, the scalability and affordability of consumer smartwatches make this strategy feasible for population-level implementation.\u003c/p\u003e \u003cp\u003eNonetheless, several limitations should be acknowledged. Despite careful device calibration, smartwatch-derived HRV and activity data are susceptible to environmental influences and algorithmic variability. The study cohort consisted of health-conscious volunteers, which may limit generalizability. Hypertension diagnosis relied on office-based measurements rather than continuous ambulatory monitoring, and biochemical correlates such as inflammatory or neurohormonal markers were not assessed. Additionally, while machine learning improved discrimination, the observational design precludes causal inference. Future studies should include larger, multi-ethnic populations, standardized device validation, and integration with vascular and metabolic biomarkers to enhance mechanistic understanding.\u003c/p\u003e \u003cp\u003eIn conclusion, smartwatch-derived HRV and MVPA were independent predictors of early hypertension in normotensive adults, and their combination with clinical parameters substantially improved predictive accuracy. These findings underscore the potential of wearable devices as noninvasive, scalable tools for early detection and prevention of hypertension in real-life settings. As wearable technology becomes increasingly integrated into daily health monitoring, its application in cardiovascular risk stratification may transform hypertension prevention from reactive management to proactive digital surveillance.\u003c/p\u003e \u003cp\u003e \u003cb\u003eFunding Declaration\u003c/b\u003e \u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding Declaration\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFunding: No financial support was received for this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eClinical Trial Number\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eClinical trial number: Not applicable.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eG.C.: Conceptualization, study design, data collection, data analysis, interpretation of results, manuscript drafting and critical revision.E.S. A.C.: Data collection, clinical evaluation, data interpretation, literature review, manuscript review and editing.All authors read and approved the final manuscript.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe data that support the findings of this study are available from Atat\u0026uuml;rk University Health Practice and Research Center (Atat\u0026uuml;rk \u0026Uuml;niversitesi Sağlık Uygulama ve Araştırma Merkezi). However, restrictions apply to the availability of these data, which were accessed under institutional licence for the present study and are therefore not publicly available. The data may be obtained upon reasonable request and with permission from the Atat\u0026uuml;rk University Health Practice and Research Center.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eSchroeder EB, Liao D, Chambless LE, Prineas RJ, Evans GW, Heiss G. Hypertension, blood pressure, and heart rate variability: the Atherosclerosis Risk in Communities (ARIC) study. Hypertension. 2003;42(6):1106\u0026ndash;11.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKang J, Chang Y, Kim Y, Shin H, Ryu S. Ten-Second Heart Rate Variability, Its Changes Over Time, and the Development of Hypertension. Hypertension. 2022;79(6):1308\u0026ndash;18.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDiaz KM, Booth JN 3rd, Seals SR, Abdalla M, Dubbert PM, Sims M, et al. Physical Activity and Incident Hypertension in African Americans: The Jackson Heart Study. Hypertension. 2017;69(3):421\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSingh JP, Larson MG, Tsuji H, Evans JC, O'Donnell CJ, Levy D. Reduced heart rate variability and new-onset hypertension: Insights from the Framingham Heart Study. Hypertension. 1998;32(2):293\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeals DR, Desouza CA, Donato AJ, Tanaka H. Habitual exercise and arterial aging. J Appl Physiol (1985). 2008;105(4):1323\u0026ndash;32.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrassi G, Seravalle G, Quarti-Trevano F. The 'neuroadrenergic hypothesis' in hypertension: Current evidence. Exp Physiol. 2010;95(5):581\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShcherbina A, Mattsson CM, Waggott D, Salisbury H, Christle JW, Hastie T, et al. Accuracy in wrist-worn, sensor-based measurements of heart rate and energy expenditure in a diverse cohort. J Pers Med. 2017;7(2):3.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBent B, Goldstein BA, Kibbe WA, Dunn JP. Investigating sources of inaccuracy in wearable optical heart rate sensors. NPJ Digit Med. 2020;3:18.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoudreaux BD, Hebert EP, Hollander DB, Williams BM, Cormier CL, Naquin MR, et al. Validity of wearable activity monitors during cycling and resistance exercise. Med Sci Sports Exerc. 2018;50(3):624\u0026ndash;33.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGreen DJ, Hopman MT, Padilla J, Laughlin MH, Thijssen DH. Vascular adaptation to exercise in humans: Role of hemodynamic stimuli. Physiol Rev. 2017;97(2):495\u0026ndash;528.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKang J, Chang Y, Kim Y, Shin H, Ryu S. Ten-Second Heart Rate Variability, Its Changes Over Time, and the Development of Hypertension. Hypertension. 2022;79(6):1308\u0026ndash;18.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWilliams B, Mancia G, Spiering W, Agabiti Rosei E, Azizi M, Burnier M, et al. 2018 ESC/ESH Guidelines for the management of arterial hypertension. Eur Heart J. 2018;39(33):3021\u0026ndash;104.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBent B, Goldstein BA, Kibbe WA, Dunn JP. Investigating sources of inaccuracy in wearable optical heart rate sensors. NPJ Digit Med. 2020;3:18.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHernando D, Roca S, Sancho J, Alesanco \u0026Aacute;, Bail\u0026oacute;n R. Validation of the Apple Watch for heart rate variability measurements during relax and mental stress in healthy subjects. Sens (Basel). 2018;18(8):2619.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMukkamala R, Stergiou GS, Avolio AP. Cuffless Blood Pressure Measurement. Annu Rev Biomed Eng. 2022;6(24):203\u0026ndash;30.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-cardiovascular-disorders","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bcar","sideBox":"Learn more about [BMC Cardiovascular Disorders](http://bmccardiovascdisord.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bcar/default.aspx","title":"BMC Cardiovascular Disorders","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Hypertension, Heart rate variability, Physical activity, Smartwatch, Wearable technology","lastPublishedDoi":"10.21203/rs.3.rs-8238959/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8238959/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground:\u003c/h2\u003e \u003cp\u003eEarly detection of hypertension is crucial for cardiovascular prevention. Smartwatch-derived physiological and activity metrics may offer a practical tool for identifying individuals at risk before clinical onset.\u003c/p\u003e\u003ch2\u003eMethods:\u003c/h2\u003e \u003cp\u003eIn this 12-month prospective study, 230 normotensive adults (30\u0026ndash;60 years) were monitored using smartwatches recording heart rate variability, resting heart rate, and moderate-to-vigorous physical activity .\u003c/p\u003e\u003ch2\u003eResults:\u003c/h2\u003e \u003cp\u003eTwenty-eight participants (12.2%) developed hypertension. Those with new-onset hypertension had lower heart rate variability (42.6\u0026thinsp;\u0026plusmn;\u0026thinsp;10.8 vs. 56.1\u0026thinsp;\u0026plusmn;\u0026thinsp;12.9 ms, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and lower moderate-to-vigorous physical activity (38.4\u0026thinsp;\u0026plusmn;\u0026thinsp;14.2 vs. 57.9\u0026thinsp;\u0026plusmn;\u0026thinsp;16.7 min/day, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Reduced heart rate variability and moderate-to-vigorous physical activity independently predicted hypertension.\u003c/p\u003e\u003ch2\u003eConclusions:\u003c/h2\u003e \u003cp\u003eSmartwatch-derived autonomic and exercise metrics can identify normotensive adults at risk of developing hypertension, supporting their role as digital biomarkers in preventive cardiology.\u003c/p\u003e","manuscriptTitle":"Smartwatch-Derived Exercise Metrics as Predictors of Early Hypertension: A Prospective Observational Study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-30 00:51:43","doi":"10.21203/rs.3.rs-8238959/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-01-06T11:50:22+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-05T12:40:07+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-29T16:54:30+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"65990014609167866936716795502278635023","date":"2025-12-26T17:32:14+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-24T11:30:15+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"64842420779565443656163312584316408927","date":"2025-12-22T09:15:09+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"100434565942954413054487508964967470879","date":"2025-12-21T20:09:42+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-12-19T20:06:54+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-12-11T17:42:52+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-12-08T14:43:51+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-12-08T14:43:29+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Cardiovascular Disorders","date":"2025-11-29T19:27:40+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-cardiovascular-disorders","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bcar","sideBox":"Learn more about [BMC Cardiovascular Disorders](http://bmccardiovascdisord.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bcar/default.aspx","title":"BMC Cardiovascular Disorders","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"2a1c08a8-8643-4f0c-89f1-c246d1c14a5b","owner":[],"postedDate":"December 30th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-29T18:09:21+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-30 00:51:43","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8238959","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8238959","identity":"rs-8238959","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}