Study on the Correlation between the Course of Hypertension and Autonomic Nervous System Damage in Elderly Patients in Primary Care Institutions in Ningxia Hui Autonomous Region Based on Single Lead Wearable Electrocardiogram Device

YU Xinyan¹, YANG Jianyun², JIANG Qingru¹, CHEN Tao³, SU Peng⁴, WANG Siyang⁵, LUO Zhanwu⁶, ZHANG Haicheng⁷*

¹Xinhua Street Community Health Service Center, The First People's Hospital, Yinchuan 750001, China
²Jingcheng Community Health Service Center, Binhe New District, Yinchuan 750001, China
³Functional Department, Liangzhou Integrated Traditional Chinese and Western Medicine Hospital, Wuwei 733000, China
⁴School of Public Health, North China University of Science and Technology, Tangshan 063210, China
⁵Qujing Health Service Center, Qingtongxia 751600, China
⁶People's Hospital of Jiangsibao District, Wuzhong 751900, China
⁷Department of Cardiology, Peking University People's Hospital, Beijing 100044, China

*Corresponding author: ZHANG Haicheng, Chief Physician; E-mail: haichengzhang@bjmu.edu.cn

Abstract

Background: Hypertension is a major chronic disease managed by primary healthcare institutions in China. Cardiac autonomic dysfunction is a key cause of blood pressure regulation imbalance and adverse cardiovascular events. Therefore, in primary healthcare institutions, the use of efficient and portable single lead wearable electrocardiogram (ECG) devices can help explore the relationship between elderly hypertensive patients and the autonomic nervous system, providing a simple, efficient, low-cost, and sustainable suitable method and objective basis for the management and prevention of cardiovascular disease in primary hypertensive patients.

Objective: To investigate the correlation between the duration of hypertension and autonomic nervous system damage in elderly patients in primary care institutions in the Ningxia Hui Autonomous Region using a single lead wearable ECG device.

Methods: A total of 2,137 elderly hypertension patients aged 65 years or older from 20 primary medical institutions in the Ningxia Hui Autonomous Region were enrolled in this study. ECG data of 72 hours, along with basic information, psychological health, and lifestyle data, were collected and uploaded to the cloud platform using a single-lead wearable ECG device from January 2022 to December 2022. Based on the heart rate variability (HRV) time-domain parameter standard deviation of all sinus rhythm RR intervals (SDNN), subjects were divided into two groups: a normal group (SDNN>100 ms, n=470) and an abnormal group (SDNN ≤ 100 ms, n=1,667). Propensity score matching was used to match subjects on a 1:1 basis, adjusting for confounding factors with a caliper value of 0.02, and the matching effect was verified using inverse probability weighting. Single-factor and multivariate logistic regression analyses were conducted to investigate the relationship between the onset of hypertension and autonomic nerve damage, and the matching effect was further verified. Subgroup analyses were performed using propensity score matching based on the onset of hypertension and autonomic nerve damage, with sensitivity analyses conducted before and after matching. Additionally, a nonlinear association between the duration of hypertension and autonomic nerve damage was examined using restricted cubic splines (RCS) analysis to test interaction effects.

Results: Among the patients, 479 had grade 1 hypertension and 1,658 had grade 2 hypertension. Disease duration was categorized as follows: 1,203 patients had less than 5 years of disease duration, 753 had 5-10 years, 110 had 10-15 years, 41 had 15-20 years, 26 had 20-30 years, and 4 had more than 30 years. The multivariate logistic regression analysis showed that the relationship between autonomic nerve damage and hypertension duration was positive both before and after matching (P<0.001). Subgroup analysis showed that the relationship between hypertension duration and autonomic nerve damage was stronger in patients under 80 years old, with lower education levels, without comorbid coronary heart disease, female patients, and those with obstructive sleep apnea (OSA) both before and after matching, with statistically significant differences (P<0.05) and interaction effects (P<0.05). However, the RCS analysis showed no nonlinear relationship between the course of hypertension and autonomic nerve damage before or after matching.

Conclusion: There was a positive correlation between the course of hypertension and autonomic nerve damage in elderly hypertension patients from the primary healthcare centers of the Ningxia Hui Autonomous Region. Primary healthcare facilities should strengthen health education for patients to improve their treatment adherence, which can delay autonomic nerve damage in hypertensive patients.

Keywords: Hypertension; Elderly; Disease course; Autonomic nerve; Primary health service centers; Single-lead wearable electrocardiogram monitoring

1. Methods

1.1 Study Population and Design

A total of 2,137 hypertensive patients aged 65 years or older from 20 primary healthcare institutions in the Ningxia Hui Autonomous Region were selected as study subjects between December 2021 and December 2022. These patients had their ECG data collected using single-lead wearable ECG devices and uploaded to a cloud platform. The study collected 72-hour ECG data along with basic patient information, mental health status, and lifestyle data from the platform.

Inclusion criteria: (1) Age ≥ 65 years; (2) History of grade 1 or 2 hypertension; (3) All patients were taking antihypertensive medication regularly under the guidance of primary care physicians; (4) No cognitive impairment.

Exclusion criteria: (1) Incomplete data on ECG, basic information, lifestyle habits, or mental health; (2) Patients who had never taken antihypertensive medication.

This study followed the Declaration of Helsinki and was approved by the Ethics Committee of Yinchuan First People's Hospital (KT-2021-116). All participants provided informed consent.

1.2 Hypertension Grading and Disease Course

Hypertension grading and diagnosis time were recorded by primary care physicians in the patient management cloud platform. Hypertension grading followed the Chinese Guidelines for the Management of Hypertension in the Elderly 2023 [5] (Grade 1: systolic blood pressure 140-159 mmHg and/or diastolic blood pressure 90-99 mmHg; Grade 2: systolic blood pressure 160-179 mmHg and/or diastolic blood pressure 100-109 mmHg). Disease course was calculated from the time of diagnosis to the time of data collection and categorized as <5 years, 5-10 years, 10-15 years, 15-20 years, 20-30 years, and ≥30 years.

1.3 Arrhythmia Diagnosis and SDNN Analysis

Primary care physicians fitted patients with a single-lead wearable dynamic ECG recorder (Model 401, Sichuan Medical Device Registration No. 20212070096) produced by Chengdu Xinhui Juyuan Technology Co., Ltd. to collect 72-hour ECG data, which were uploaded to the cloud platform. Professional ECG physicians accessed the data from the platform, excluded interference that could not be recognized by the system, correctly marked RR intervals that were misidentified by the system (normal sinus rhythm and arrhythmic RR intervals), diagnosed arrhythmia types, and analyzed the automatically calculated SDNN values based on the algorithm [22].

1.4 Assessment and Grouping of Autonomic Nerve Damage

Based on the HRV time-domain parameter SDNN, autonomic nerve damage was assessed and patients were divided into a normal group (SDNN ≥ 100 ms) and an abnormal group (SDNN < 100 ms) [22].

1.5 Assessment of Other Variables

All covariates were collected from the patient management cloud platform and included: (1) Basic patient information: age, sex, BMI, ethnicity, occupation, education level, urban/rural distribution, and medical history [diabetes, coronary heart disease, and stroke diagnosed according to the Chinese Clinical Guidelines for the Prevention and Treatment of Type 2 Diabetes in the Elderly (2022 Edition) [23], 2023 AHA/ACC/ACCP/ASPC/NLA/PCNA Guideline for the Management of Patients with Chronic Coronary Disease [24], and Guidelines for the Diagnosis and Treatment of Ischemic Stroke in Primary Care (2021) [25]]; (2) Obstructive sleep apnea (OSA): Assessed using the apnea-hypopnea index (AHI) calculated by the cyclic variation of heart rate (CVHR) technology [26] from the single-lead wearable device, with moderate to severe OSA defined as AHI ≥ 15 events/hour (CVHR technology has high sensitivity and specificity for identifying moderate to severe OSA [26-27]); (3) Mental health: Normal mental health defined as Self-Rating Depression Scale (SDS) score <53 and Self-Rating Anxiety Scale (SAS) score <50, while abnormal defined as SDS score ≥53 and/or SAS score ≥50 [28-29]; (4) Lifestyle habits: Smoking status (never, former [quit >5 years], current), alcohol consumption (never, former [quit >5 years], current), tea drinking (never, occasionally [<3 days/week], regularly [≥4 days/week]), and exercise (never, short duration [<1 hour/day], long duration [≥1 hour/day]).

1.6 Statistical Methods

Data were collected through the independently developed patient management cloud platform and analyzed using RStudio 4.1.1 and Python statistical software. Normally distributed continuous baseline data were expressed as (x̄±s) and compared between two groups using t-tests or among multiple groups using one-way ANOVA. Categorical data were expressed as frequencies and percentages and compared using χ² tests. Propensity score matching was performed at a 1:1 ratio to adjust for confounding factors with a caliper value of 0.02. Inverse probability weighting was used as a reference to verify the matching effect. Single-factor and multivariate logistic regression analyses were conducted to examine the relationship between hypertension duration and autonomic nerve damage. Subgroup analyses were performed before and after matching based on hypertension duration and autonomic nerve damage. Restricted cubic splines (RCS) analysis was used to test for nonlinear associations and interaction effects between hypertension duration and autonomic nerve damage. Statistical significance was defined as P<0.05.

2. Results

2.1 Baseline Characteristics Before and After Matching

Among the collected patients, 479 had grade 1 hypertension and 1,658 had grade 2 hypertension. Disease duration distribution was: 1,203 patients (<5 years), 753 (5-10 years), 110 (10-15 years), 41 (15-20 years), 26 (20-30 years), and 4 (≥30 years). Before matching, significant differences were observed between the two groups in sex, age, urban/rural distribution, education level, occupation, and comorbidities including coronary heart disease, diabetes, supraventricular premature beats, and sinoatrial block (P<0.05) (Table 1 [TABLE:1]).

To reduce data bias and confounding effects, this study performed 1:1 propensity score matching with a caliper value of 0.02. Using inverse probability weighting as a reference, the matching effect was verified. After matching, both groups included 445 patients. The weighted groups showed good balance across baseline characteristics, with standardized mean differences (SMD) <0.1 for all variables except BMI grouping (SMD=0.109), which was considered acceptable given its distribution relative to the outcome variable (Table 2 [TABLE:2]). The data distribution after matching and weighting is shown in Figure 1 [FIGURE:1].

2.2 Relationship Between Hypertension Duration and Autonomic Nerve Damage

2.2.1 Baseline Characteristics by Hypertension Duration Segment Before Matching

Before matching, significant differences were observed across duration segments in age, age grouping, urban/rural distribution, education level, occupation, coronary heart disease history, stroke history, parallel rhythm, ventricular pre-excitation, and SDNN abnormality (P<0.05). See Schedule 1.

2.2.2 Multivariate Logistic Regression Analysis of Hypertension Duration and Autonomic Nerve Damage

Using SDNN grouping as the dependent variable (normal group=0, abnormal group=1) and hypertension duration as the independent variable, three logistic regression models were constructed to analyze the relationship between hypertension duration and autonomic nerve damage in elderly patients before and after matching, with variables showing statistical significance in univariate analysis included as covariates (Table 3 [TABLE:3]). In all models before matching, hypertension duration was positively correlated with autonomic nerve damage in the total hypertensive population, grade 1 hypertension patients, and grade 2 hypertension patients (P<0.001). Although this association was attenuated after matching, it remained statistically significant (P<0.001). These findings suggest that hypertension duration is an independent risk factor for autonomic nerve damage. Notably, both before and after matching, the correlation between duration and autonomic nerve damage appeared stronger in grade 1 hypertension patients compared to grade 2 patients across all three models (Table 4 [TABLE:4]).

2.2.3 Subgroup Analysis Before and After Matching

To avoid heterogeneity and examine potential interactions between hypertension duration and autonomic nerve damage across different populations, subgroup analyses were performed. Before matching, the correlation was stronger among patients under 80 years old, with lower education levels, and without comorbid coronary heart disease, with statistically significant differences (P<0.05) and significant interaction effects (P<0.05). After matching, the correlation was stronger among female patients and those with comorbid OSA, with statistically significant differences (P<0.05) and significant interaction effects (P<0.05). Forest plots of subgroup interactions are shown in Figure 2 [FIGURE:2] and Figure 3 [FIGURE:3].

2.2.4 Nonlinear Association Analysis Between Hypertension Duration and Autonomic Nerve Damage

Based on Model 3, RCS was used to visualize the relationship between hypertension duration and autonomic nerve damage before and after matching. The RCS analysis showed no nonlinear association between hypertension duration and autonomic nerve damage (P=0.0554 before matching, P=0.2902 after matching). However, both before and after matching, the analysis suggested that grade 2 hypertension patients experienced autonomic nerve damage earlier than grade 1 patients: grade 2 patients showed conversion to a risk factor at 4 years of duration, while grade 1 patients showed conversion at 5.3 years and 10.3 years before and after matching, respectively (Figure 4 [FIGURE:4]).

3. Discussion

3.1 Relationship Between Hypertension Duration and Autonomic Nerve Damage

Previous studies have shown that hypertensive patients have varying degrees of autonomic nerve function damage [30], which is not only independent of other causes of elevated blood pressure [31] but also independently associated with ventricular remodeling in hypertensive patients [32]. Tao [33] compared HRV parameters between 41 hypertensive patients and 30 healthy controls and found that the degree of autonomic nerve damage was related to hypertension severity.

Although current research suggests that the pathogenesis of hypertension is mainly related to increased sympathetic nerve activity, which positively correlates with the magnitude of blood pressure elevation [34-35], scholars have also found [36-37] that the dual innervation of the cardiac autonomic nervous system is not equally balanced. In most cases, both humans and animals show predominant vagal regulation. Cardiac autonomic function involves not only independent feedback mechanisms of sympathetic and vagal nerves but also mutual inhibition between them to achieve dynamic balance. In hypertensive patients, impaired cardiac autonomic function leads to reduced vagal tone and dominant sympathetic influence on the heart and peripheral vessels [38]. Pal et al. [39] found that vagal inhibition plays an important role in both prehypertensive and progressive stages. Guo [37] also noted that evaluation of autonomic regulation should assess the combined effects of sympathetic and vagal nerves. This study used SDNN as an indicator of autonomic nerve function, which is a recognized parameter reflecting total sympathetic and vagal tone [22]. Additionally, grade 1 hypertension patients have lower disease awareness and treatment adherence compared to grade 2 patients [40]. As pharmacological treatment is the first of five major prescriptions for hypertension management [41] and plays a necessary role in achieving blood pressure control [42], these findings suggest that primary care physicians should emphasize vagal function regulation and balance cardiac autonomic function, which may be more important than simply reducing sympathetic tone. This also highlights the need to strengthen health education for grade 1 hypertension patients to improve their disease awareness and medication adherence.

Although this study showed no nonlinear relationship between hypertension duration and autonomic nerve damage, both before and after matching, grade 2 hypertension patients experienced autonomic nerve damage earlier than grade 1 patients. This suggests that primary care physicians should screen and evaluate autonomic function earlier in grade 2 hypertension patients.

3.2 Factors Influencing the Relationship Between Hypertension Duration and Autonomic Nerve Damage

In male arterial blood pressure reflex mechanisms, sympathetic activity positively correlates with total peripheral resistance and negatively correlates with cardiac output [43]. In females, sympathetic activity couples with blood pressure primarily through vasodilation mediated by vascular β-adrenergic receptors [44]. However, with aging, β-adrenergic receptor sensitivity decreases in women [45], and estrogen levels that promote vasodilation via nitric oxide or β-adrenergic receptors also decline [46]. Decreased estrogen levels in women are associated with reduced SDNN [47-48].

Previous studies have shown that the prevalence of frailty is higher in elderly hypertensive patients aged ≥80 years compared to younger elderly patients [49]. Blood pressure variability (BPV) is positively correlated with frailty in elderly hypertensive patients [50], and elevated BPV indicates increased sympathetic-mediated vascular reactivity [51]. Most frail elderly hypertensive patients also have chronic diseases related to autonomic nerve damage [52] and more frequently have two or more comorbidities [53], which affect frailty severity [54]. Therefore, autonomic nerve damage in very elderly hypertensive patients is more closely related to frailty status than disease duration. Primary care physicians should strengthen frailty screening in patients over 80 years old, as frailty is considered an important indicator for determining which patients benefit from blood pressure reduction [55]. Additionally, blood pressure shows a U-shaped relationship with mortality in very elderly patients [56], suggesting that both BPV and HRV should be evaluated to provide more precise references for clinical treatment decisions.

Coronary heart disease patients often experience acute autonomic imbalance due to myocardial ischemia [57]. However, the preferred antihypertensive medications for elderly hypertensive patients with coronary heart disease—β-blockers and angiotensin-converting enzyme inhibitors—have significant sympathetic inhibitory effects [5]. Statins, commonly used in coronary heart disease patients, also inhibit sympathetic excitation [58]. Moreover, awareness of coronary heart disease hazards is higher than that of hypertension among Chinese elderly [3,59], which improves treatment adherence when the two conditions coexist. OSA disrupts normal breathing during sleep, causing nocturnal cardiac autonomic dysfunction and damage [60], which is exacerbated when combined with hypertension. In this study, OSA patients were screened using single-lead wearable ECG devices based on CVHR technology. Even among patients previously diagnosed with OSA, low awareness and treatment rates [61] meant that most did not receive standardized treatment.

Patients with lower education levels have fewer channels to obtain hypertension knowledge, which is an independent factor affecting medication adherence [62], while medication adherence is crucial for blood pressure control [42,63-64]. Additionally, lower education levels are associated with lower socioeconomic returns [65], and low income not only reduces medication adherence [66] but also creates chronic stress factors [65]. Under stress, the autonomic nervous system is rapidly activated, leading to imbalance [67]. Furthermore, individuals with lower education and socioeconomic status have reduced access to healthcare services [68].

These analyses suggest that although hypertension duration is positively correlated with autonomic nerve damage, improving medication adherence through health education can delay autonomic nerve damage. Health education significantly improves treatment and medication adherence in primary hypertensive patients [69]. Therefore, primary care physicians should strengthen health education for elderly hypertensive patients, while health administrative departments need to formulate policies to increase healthcare accessibility for low-income populations. These measures will benefit patients and families and are significant for hypertension management and cardiovascular disease prevention in primary care settings [3,70].

Conclusion

In elderly hypertensive patients from primary healthcare institutions in the Ningxia Hui Autonomous Region, hypertension duration is positively correlated with autonomic nerve damage, suggesting that hypertension duration is an independent risk factor for autonomic nerve damage. However, strengthening health education to improve treatment adherence and blood pressure control can delay autonomic nerve damage in hypertensive patients.

The strengths of this study include: (1) Use of propensity score matching to adjust for confounding factors, with inverse probability weighting as a reference to verify matching effectiveness, improving result credibility and generalizability; (2) Subgroup analyses to explore relationships across different populations and determine precise management strategies; (3) Innovative segmentation of hypertension duration combined with grading to provide objective evidence for primary care management; (4) Real-world study design accurately reflecting clinical practice; (5) Data collection through single-lead wearable ECG devices and cloud platform, enabling efficient, simple, low-cost, and sustainable arrhythmia screening and autonomic function assessment.

Limitations include: (1) Covariates such as income, medication type, and adherence were not included; (2) OSA diagnosis relied on algorithm-based assessment from the single-lead device, without polysomnography confirmation for most patients; (3) Only SDNN was used to assess autonomic nerve damage, without additional HRV parameters for more comprehensive sympathetic and vagal function evaluation. These aspects require improvement in future studies.

References

[1] World Health Organization. Guideline for the pharmacological treatment of hypertension in adults[J]. Geneva: World Health Organization, 2021: 1-61.

[2] PAN H, HIBINO M, KOBEISSI E, et al. Blood pressure, hypertension and the risk of sudden cardiac death: a systematic review and meta-analysis of cohort studies[J]. Eur J Epidemiol, 2020, 35(5): 443-454. DOI: 10.1007/s10654-019-00593-4.

[3] The Writing Committee of the Report on Cardiovascular Health and Diseases in China. Summary of the 2022 Report on Cardiovascular Health and Diseases in China[J]. Chinese Circulation Journal, 2023, 38(6): 583-612.

[4] Zhang M, Wu J, Zhang X, et al. Study on the prevalence and control of hypertension among adult residents in China in 2018[J]. Chinese Journal of Epidemiology, 2021, 42(10): 1780-1789. DOI: 10.3760/cma.j.cn112338-20210508-00379.

[5] Hypertension Branch of Chinese Geriatrics Society, Beijing Hypertension Prevention and Treatment Association, National Clinical Research Center for Geriatric Diseases (Chinese PLA General Hospital), et al. Chinese Guidelines for the Management of Hypertension in the Elderly 2023[J]. Chinese Journal of Hypertension, 2023, 31(6): 508-538. DOI: 10.16439/j.issn.1673-7245.2023.06.003.

[6] TOCCI G, CICERO A F, SALVETTI M, et al. Attitudes and preferences for the clinical management of hypertension and hypertension-related cardiac disease in general practice: results of the Italian Hypertension and Heart Survey[J]. J Hum Hypertens, 2015, 29(7): 409-416. DOI: 10.1038/jhh.2014.115.

[7] SCHROEDER E B, LIAO D P, CHAMBLESS L E, et al. Hypertension, blood pressure, and heart rate variability: the Atherosclerosis Risk in Communities (ARIC) study[J]. Hypertension, 2003, 42(6): 1106-1111. DOI: 10.1161/01.HYP.0000100444.71069.73.

[8] CARDOSO C R, MORAES R A, LEITE N C, et al. Relationships between reduced heart rate variability and pre-clinical cardiovascular disease in patients with type 2 diabetes[J]. Diabetes Res Clin Pract, 2014, 106(1): 110-117. DOI: 10.1016/j.diabres.2014.07.005.

[9] GRISK O, RETTIG R. Interactions between the sympathetic nervous system and the kidneys in arterial hypertension[J]. Cardiovasc Res, 2004, 61(2): 238-246. DOI: 10.1016/j.cardiores.2003.11.024.

[10] KANG J, CHANG Y, KIM Y, et al. Ten-second heart rate variability, its changes over time, and the development of hypertension[J]. Hypertension, 2022, 79(6): 1308-1318. DOI: 10.1161/HYPERTENSIONAHA.121.18589.

[11] JARCZOK M N, KOENIG J, WITTLING A, et al. First evaluation of an index of low vagally-mediated heart rate variability as a marker of health risks in human adults: proof of concept[J]. J Clin Med, 2019, 8(11): 1940. DOI: 10.3390/jcm8111940.

[12] NOLAN R P, JONG P, BARRY-BIANCHI S M, et al. Effects of drug, biobehavioral and exercise therapies on heart rate variability in coronary artery disease: a systematic review[J]. Eur J Cardiovasc Prev Rehabil, 2008, 15(4): 386-396. DOI: 10.1097/hjr.0b013e3283030a97.

[13] SESSA F, ANNA V, MESSINA G, et al. Heart rate variability as predictive factor for sudden cardiac death[J]. Aging (Albany NY), 2018, 10(2): 166-177. DOI: 10.18632/aging.101386.

[14] ZEPPENFELD K, TFELT-HANSEN J, DE RIVA M, et al. 2022 ESC Guidelines for the management of patients with ventricular arrhythmias and the prevention of sudden cardiac death[J]. Eur Heart J, 2022, 43(40): 3997-4126. DOI: 10.1093/eurheartj/ehac262.

[15] LAI Y R, HUANG C C, CHENG B C, et al. Feasibility of combining heart rate variability and electrochemical skin conductance as screening and severity evaluation of cardiovascular autonomic neuropathy in type 2 diabetes[J]. J Diabetes Investig, 2021, 12(9): 1671-1679. DOI: 10.1111/jdi.13518.

[16] SCHERBAKOV N, BARKHUDARYAN A, EBNER N, et al. Early rehabilitation after stroke: relationship between the heart rate variability and functional outcome[J]. ESC Heart Fail, 2020, 7(5): 2983-2991. DOI: 10.1002/ehf2.12917.

[17] YUE W W, YIN J, CHEN B, et al. Analysis of heart rate variability in masked hypertension[J]. Cell Biochem Biophys, 2014, 70(1): 201-204. DOI: 10.1007/s12013-014-9882-y.

[18] Chu JK, Zhao Q, Men L, et al. Application of the internet plus medical consortium diagnosis and treatment model based on wearable ECG devices in cardiovascular diseases[J]. Chinese Journal of Cardiovascular Research, 2023, 21(2): 160-165. DOI: 10.3969/j.issn.1672-5301.2023.02.012.

[19] Wang QS, Chen T, Han BS, et al. Application of intelligent wearable 12-lead ECG in remote diagnosis and treatment of cardiovascular diseases[J]. Chinese Circulation Journal, 2022, 37(7): 738-744.

[20] GAWAŁKO M, DUNCKER D, MANNINGER M, et al. The European TeleCheck-AF project on remote app-based management of atrial fibrillation during the COVID-19 pandemic: centre and patient experiences[J]. Europace, 2021, 23(7): 1003-1015. DOI: 10.1093/europace/euab050.

[21] Yu XY, Zhao XD, Zhao XY, et al. Real-world study of arrhythmia screening protocols based on community mobile healthcare[J]. Chinese General Practice, 2023, 26(2): 192-200, 209.

[22] Heart rate variability. Standards of measurement, physiological interpretation, and clinical use. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology[J]. Eur Heart J, 1996, 17(3): 354-381.

[23] Chinese Clinical Guideline Writing Group for Prevention and Treatment of Diabetes in the Elderly, Geriatric Endocrinology and Metabolism Branch of Chinese Geriatrics Society, Geriatric Endocrinology and Metabolism Branch of Chinese Association of Gerontology and Geriatrics, et al. Chinese Clinical Guidelines for the Prevention and Treatment of Type 2 Diabetes in the Elderly (2022 Edition)[J]. Chinese Journal of Internal Medicine, 2022, 61(1): 12-50. DOI: 10.3760/cma.j.cn112138-20211027-00751.

[24] VIRANI S S, NEWBY L K, ARNOLD S V, et al. 2023 AHA/ACC/ACCP/ASPC/NLA/PCNA guideline for the management of patients with chronic coronary disease: a report of the American heart association/American college of cardiology joint committee on clinical practice guidelines[J]. Circulation, 2023, 148(9): e9-e119. DOI: 10.1161/CIR.0000000000001168.

[25] Chinese Medical Association, Chinese Medical Journals Publishing House, General Practice Branch of Chinese Medical Association, et al. Guidelines for the Diagnosis and Treatment of Ischemic Stroke in Primary Care (2021)[J]. Chinese Journal of General Practitioners, 2021, 20(9): 927-946. DOI: 10.3760/cma.j.cn114798-20210508-00379.

[26] HAYANO J, WATANABE E, SAITO Y, et al. Screening for obstructive sleep apnea by cyclic variation of heart rate[J]. Circ Arrhythm Electrophysiol, 2011, 4(1): 64-72. DOI: 10.1161/CIRCEP.110.958009.

[27] QASEEM A, HOLTY J C, OWENS D K, et al. Management of obstructive sleep apnea in adults: a clinical practice guideline from the American College of Physicians[J]. Ann Intern Med, 2013, 159(7): 471-483. DOI: 10.7326/0003-4819-159-7-201310010-00004.

[28] Chinese Medical Association, Chinese Medical Journals Publishing House, General Practice Branch of Chinese Medical Association, et al. Guidelines for the Diagnosis and Treatment of Generalized Anxiety Disorder in Primary Care (2021)[J]. Chinese Journal of General Practitioners, 2021, 20(12): 1232-1241. DOI: 10.3760/cma.j.cn114798-20211025-00790.

[29] Chinese Medical Association, Chinese Medical Journals Publishing House, General Practice Branch of Chinese Medical Association, et al. Guidelines for the Diagnosis and Treatment of Depression in Primary Care (2021)[J]. Chinese Journal of General Practitioners, 2021, 20(12): 1249-1260. DOI: 10.3760/cma.j.cn114798-20211025-00790.

[30] VIRTANEN R, JULA A, KUUSELA T, et al. Reduced heart rate variability in hypertension: associations with lifestyle factors and plasma renin activity[J]. J Hum Hypertens, 2003, 17(3): 171-179. DOI: 10.1038/sj.jhh.1001529.

[31] KATHOLI R E, ROCHA-SINGH K J. The role of renal sympathetic nerves in hypertension: has percutaneous renal denervation refocused attention on their clinical significance[J]. Prog Cardiovasc Dis, 2009, 52(3): 243-248. DOI: 10.1016/j.pcad.2009.09.003.

[32] TADIC M, CUSPIDI C, PENCIC B, et al. Relationship between right ventricular remodeling and heart rate variability in arterial hypertension[J]. J Hypertens, 2015, 33(5): 1090-1097. DOI: 10.1097/HJH.0000000000000511.

[33] Tao G. Clinical study of heart rate variability in hypertension[J]. Journal of Practical Electrocardiology, 2005, 14(1): 35-36. DOI: 10.13308/j.issn.1008-0740.2005.01.026.

[34] DELALIO L J, SVED A F, STOCKER S D. Sympathetic nervous system contributions to hypertension: updates and therapeutic relevance[J]. Can J Cardiol, 2020, 36(5): 712-720. DOI: 10.1016/j.cjca.2020.03.003.

[35] Jin JJ, Yao YQ, Wu YT, et al. Analysis of heart rate variability in primary hypertension patients with different blood pressure grades[J]. Chinese Journal of Evidence-Based Cardiovascular Medicine, 2013, 5(3): 275-277. DOI: 10.3969/j.1674-4055.2013.03.021.

[36] YANG T, LEVY M N. The phase-dependency of the cardiac chronotropic responses to vagal stimulation as a factor in sympathetic-vagal interactions[J]. Circ Res, 1984, 54(6): 703-710. DOI: 10.1161/01.res.54.6.703.

[37] Guo JH. Detection of heart rate deceleration capacity[J]. Journal of Clinical Electrocardiology, 2009, 18(1): 59-68. DOI: 10.3969/j.issn.1005-0272.2009.01.019.

[38] GRASSI G, RAM V S. Evidence for a critical role of the sympathetic nervous system in hypertension[J]. J Am Soc Hypertens, 2016, 10(5): 457-466. DOI: 10.1016/j.jash.2016.02.015.

[39] PAL G K, ADITHAN C, AMUDHARAJ D, et al. Assessment of sympathovagal imbalance by spectral analysis of heart rate variability in prehypertensive and hypertensive patients in Indian population[J]. Clin Exp Hypertens, 2011, 33(7): 478-483. DOI: 10.3109/10641963.2010.549275.

[40] Tan YJ, Liang JR, Zhang HR. Investigation of medication use and analysis of related influencing factors in community hypertensive patients[J]. Inner Mongolia Medical Journal, 2019, 51(12): 1443-1444. DOI: 10.16096/J.cnki.nmgyxzz.2019.51.12.014.

[41] Revision Committee of the Chinese Guidelines for the Primary Management of Hypertension. Chinese Guidelines for the Primary Management of Hypertension (2014 Revision)[J]. Chinese Journal of Health Management, 2015, 1(1): 10-30.

[42] Hu DY. Hypertension control requires implementation of five prescriptions[J]. Chinese Journal of Hypertension, 2015, 23(9): 801. DOI: 10.16439/j.issn.1673-7245.2015.09.001.

[43] CHARKOUDIAN N, JOYNER M J, JOHNSON C P, et al. Balance between cardiac output and sympathetic nerve activity in resting humans: role in arterial pressure regulation[J]. J Physiol, 2005, 568(Pt 1): 315-321. DOI: 10.1113/jphysiol.2005.090076.

[44] HART E C, CHARKOUDIAN N, WALLIN B G, et al. Sex and ageing differences in resting arterial pressure regulation: the role of the β-adrenergic receptors[J]. J Physiol, 2011, 589(Pt 21): 5285-5297. DOI: 10.1113/jphysiol.2011.212753.

[45] KEARNEY P M, WHELTON M, REYNOLDS K, et al. Global burden of hypertension: analysis of worldwide data[J]. Lancet, 2005, 365(9455): 217-223. DOI: 10.1016/S0140-6736(05)17741-1.

[46] HART E C, CHARKOUDIAN N, WALLIN B G, et al. Sex differences in sympathetic neural-hemodynamic balance: implications for human blood pressure regulation[J]. Hypertension, 2009, 53(3): 571-576. DOI: 10.1161/HYPERTENSIONAHA.108.126391.

[47] Li T, Wu XF, Wu J, et al. Autonomic nervous function status and significance reflected by heart rate variability during female physiological cycles[J]. Journal of Practical Electrocardiology, 2021, 30(5): 337-341, 345. DOI: 10.13308/j.issn.2095-9354.2021.05.008.

[48] Zhang HC, Bai WP, Guo JH, et al. Relationship between menopausal symptoms and heart rate variability in perimenopausal women[J]. Chinese Journal of Cardiac Arrhythmias, 1999, 3(1): 25-27. DOI: 10.3760/cma.j.issn.1007-6638.1999.01.007.

[49] Fan L, Li JH, Hu YX, et al. Detection rate of frailty in elderly hypertensive patients with different complications[J]. Chinese Journal of Hypertension, 2015, 23(12): 1151-1155. DOI: 10.16439/j.cnki.1673-7245.2015.12.018.

[50] Han HH, Hu S, Liu J, et al. Study on the correlation between blood pressure variability and frailty in the elderly[J]. Chinese Journal of Geriatric Heart Brain and Vessel Diseases, 2019, 21(3): 258-260. DOI: 10.3969/j.issn.1009-0126.2019.03.009.

[51] NIIRANEN T J. Increased blood pressure variability: a marker of augmented sympathetic vascular reactivity[J]. Am J Hypertens, 2019, 32(6): 533-534. DOI: 10.1093/ajh/hpz030.

[52] GYAMLANI G, BASU A, GERACI S, et al. Depression, screening and quality of life in chronic kidney disease[J]. Am J Med Sci, 2011, 342(3): 186-191. DOI: 10.1097/MAJ.0b013e3182113d9e.

[53] SHIMADA H, MAKIZAKO H, DOI T, et al. Combined prevalence of frailty and mild cognitive impairment in a population of elderly Japanese people[J]. J Am Med Dir Assoc, 2013, 14(7): 518-524. DOI: 10.1016/j.jamda.2013.03.010.

[54] CASTELL M V, SÁNCHEZ M, JULIÁN R, et al. Frailty prevalence and slow walking speed in persons age 65 and older: implications for primary care[J]. BMC Fam Pract, 2013, 14: 86. DOI: 10.1186/1471-2296-14-86.

[55] INOUE T, MATSUOKA M, SHINJO T, et al. Blood pressure, frailty status, and all-cause mortality in elderly hypertensives; The Nambu Cohort Study[J]. Hypertens Res, 2022, 45(1): 146-154. DOI: 10.1038/s41440-021-00769-0.

[56] GAO H, WANG K, AHMADIZAR F, et al. Changes in late-life systolic blood pressure and all-cause mortality among oldest-old people in China: the Chinese longitudinal healthy longevity survey[J]. BMC Geriatr, 2021, 21(1): 562. DOI: 10.1186/s12877-021-02456-2.

[57] SCHWARTZ P J, ZAZA A, PALA M, et al. Baroreflex sensitivity and its evolution during the first year after myocardial infarction[J]. J Am Coll Cardiol, 1988, 12(3): 629-636. DOI: 10.1016/S0735-1097(88)80048-2.

[58] SIŃSKI M, LEWANDOWSKI J, CIARKA A, et al. Atorvastatin reduces sympathetic activity and increases baroreceptor reflex sensitivity in patients with hypercholesterolaemia and systemic arterial hypertension[J]. Kardiol Pol, 2009, 67(6): 613-620.

[59] Shen J, Miao Q, Wang X, et al. Investigation on the development trend of coronary heart disease patients and current status of self-management of cardiovascular risk factors in Nanyang City from 2016 to 2019[J]. Chinese General Practice, 2023, 26(S1): 21-24.

[60] JAVAHERI S, BARBE F, CAMPOS-RODRIGUEZ F, et al. Sleep apnea types, mechanisms, and clinical cardiovascular consequences[J]. J Am Coll Cardiol, 2017, 69(7): 841-858. DOI: 10.1016/j.jacc.2016.11.069.

[61] DRAGER L F, GENTA P R, PEDROSA R P, et al. Characteristics and predictors of obstructive sleep apnea in patients with systemic hypertension[J]. Am J Cardiol, 2010, 105(8): 1135-1139. DOI: 10.1016/j.amjcard.2009.12.017.

[62] Han EJ, Zhao JY, Zhang Y, et al. Current status and influencing factors of active aging among elderly people in nursing homes in Zhengzhou City[J]. Chinese Journal of Gerontology, 2019, 39(1): 206-209. DOI: 10.3969/j.issn.1005-9202.2019.01.070.

[63] Wang BY, Jian WY. Study on social determinants of hypertension awareness and blood pressure control in the elderly[J]. Chinese General Practice, 2015, 18(2): 152-156.

[64] Wu X, Qin Y, Cui L, et al. Current status of antihypertensive drug combination use and blood pressure control in hypertensive patients in Jiangsu Province[J]. Chinese Journal of Hypertension, 2022, 30(6): 571-576. DOI: 10.16439/j.issn.1673-7245.2022.06.013.

[65] HARRAP S B. Hypertension: genes versus environment[J]. Lancet, 1994, 344(8916): 169-171. DOI: 10.1016/s0140-6736(94)92762-6.

[66] Li XL. Investigation and analysis of influencing factors on medication adherence in elderly hypertensive patients in primary care[D]. Haikou: Hainan Medical University, 2022. DOI: 10.27952/d.cnki.ghnyx.2022.000072.

[67] MORAIS-SILVA G, COSTA-FERREIRA W, GOMES-DE-SOUZA L, et al. Cardiovascular outcomes related to social defeat stress: New insights from resilient and susceptible rats[J]. Neurobiol Stress, 2019, 11: 100181. DOI: 10.1016/j.ynstr.2019.100181.

[68] HE J, MUNTNER P, CHEN J, et al. Factors associated with hypertension control in the general population of the United States[J]. Arch Intern Med, 2002, 162(9): 1051. DOI: 10.1001/archinte.162.9.1051.

[69] Wang ZW, Wang X, Zhang LF, et al. Community hypertension control: evaluation of blood pressure management effectiveness[J]. Chinese Journal of Epidemiology, 2010, 31(1): 1-4.

[70] Chinese Hypertension Prevention and Treatment Guidelines Revision Committee, Hypertension Alliance China, Cardiovascular Disease Branch of Chinese Medical Association, et al. Chinese Guidelines for the Prevention and Treatment of Hypertension (2018 Revision)[J]. Chinese Journal of Cardiovascular Medicine, 2019, 24(1): 24-56. DOI: 10.3969/j.issn.1672-5301.2019.03.001.

[71] YU X Y, GUAN L E, SU P, et al. Study on OSA screening and influencing factors in community-based elderly hypertensive patients based on single-lead wearable ECG devices[J]. Sleep Breath, 2024, 28(6): 2445-2456.

[72] Yu XY, Zhao XY, Yang JY, et al. Analysis of the application of wearable single-lead remote ECG monitoring devices combined with scatter plots outside the hospital[J]. Chinese Circulation Journal, 2021, 36(11): 1096-1100.

[73] Yu XY, Su P, Yuan XJ, et al. Study on influencing factors of comprehensive cardiovascular disease risk in elderly patients with chronic diseases in primary care[J]. Chinese General Practice, 2024, 27(10): 1186-1193, 1200.

Schedule 1

Baseline Characteristics of Patients by Hypertension Duration Segment Before Matching

Characteristic <5 years (n=1,203) 5-10 years (n=753) 10-15 years (n=110) 15-20 years (n=41) 20-30 years (n=26) ≥30 years (n=4) χ²(F) value P value Sex [n (%)] Male 727 (60.4) 459 (61.0) 69 (62.7) 23 (56.1) 20 (76.9) 3 (75.0) Female 476 (39.6) 294 (39.0) 41 (37.3) 18 (43.9) 6 (23.1) 1 (25.0) Age (x̄±s, years) 71.39±4.84 70.74±4.96 71.76±4.26 74.73±5.66 73.42±5.18 75.5±5.07 <0.001 Age grouping [n (%)] <80 years 1,115 (92.7) 702 (93.2) 105 (95.5) 33 (80.5) 21 (80.8) 3 (75.0) ≥80 years 88 (7.3) 51 (6.8) 5 (4.5) 8 (19.5) 5 (19.2) 1 (25.0) Ethnicity [n (%)] Han 1,092 (90.8) 683 (90.7) 99 (90.0) 37 (90.2) 24 (92.3) 4 (100.0) Hui 93 (7.7) 66 (8.8) 11 (10.0) 3 (7.3) 2 (7.7) Other 18 (1.5) 4 (0.5) Urban/rural distribution [n (%)] Urban 414 (34.4) 239 (31.7) 66 (60.0) 24 (58.5) 12 (46.2) 2 (50.0) Rural 789 (65.6) 514 (68.3) 44 (40.0) 17 (41.5) 14 (53.8) 2 (50.0) BMI (x̄±s, kg/m²) 24.89±3.14 24.84±3.12 24.55±3.03 25.26±2.80 25.06±2.37 24.8±2.9 0.403 BMI grouping [n (%)] Underweight 16 (1.3) 13 (1.7) 1 (0.9) Normal 463 (38.5) 292 (38.8) 48 (43.6) 12 (29.3) 8 (30.8) Overweight 540 (44.9) 339 (45.0) 48 (43.6) 22 (53.7) 15 (57.7) Obese 184 (15.3) 109 (14.5) 14 (12.7) 6 (14.6) 3 (11.5) 2 (50.0) Education level [n (%)] Illiterate 400 (33.3) 248 (32.9) 64 (58.2) 20 (48.8) 15 (57.7) 1 (25.0) Primary school 316 (26.3) 209 (27.8) 20 (18.2) 13 (31.7) 8 (30.8) 1 (25.0) Junior high school 335 (27.8) 190 (25.2) 14 (12.7) 7 (17.1) 1 (25.0) High school and above 152 (12.6) 106 (14.1) 12 (10.9) 1 (2.4) 3 (11.5) 1 (25.0) Occupation [n (%)] Retired 358 (29.8) 281 (37.3) 24 (21.8) 11 (26.8) 8 (30.8) 1 (25.0) Other 481 (40.0) 258 (34.3) 27 (24.5) 8 (19.5) 7 (26.9) 1 (25.0) Agriculture/forestry/animal husbandry/fishery 364 (30.3) 214 (28.4) 59 (53.6) 22 (53.7) 11 (42.3) 2 (50.0) Mental health [n (%)] Normal 877 (72.9) 540 (71.7) 78 (70.9) 22 (53.7) 16 (61.5) 4 (100.0) Abnormal 326 (27.1) 213 (28.3) 32 (29.1) 19 (46.3) 10 (38.5) Smoking [n (%)] Never 1,050 (87.3) 659 (87.5) 98 (89.1) 38 (92.7) 24 (92.3) 3 (75.0) Former 7 (0.6) 3 (0.4) 1 (25.0) Current 146 (12.1) 91 (12.1) 12 (10.9) 3 (7.3) 2 (7.7) Alcohol consumption [n (%)] Never 1,094 (90.9) 680 (90.3) 100 (90.9) 38 (92.7) 22 (84.6) 3 (75.0) Former 9 (0.7) 12 (1.6) 1 (25.0) Current 100 (8.3) 61 (8.1) 10 (9.1) 3 (7.3) 4 (15.4) Tea drinking [n (%)] Never 1,031 (85.7) 641 (85.1) 100 (90.9) 35 (85.4) 22 (84.6) 4 (100.0) Occasionally 3 (0.2) 2 (0.3) Regularly 169 (14.0) 110 (14.6) 10 (9.1) 6 (14.6) 4 (15.4) Exercise [n (%)] Never 496 (41.2) 310 (41.2) 37 (33.6) 22 (53.7) 10 (38.5) 2 (50.0) Short duration 351 (29.2) 215 (28.6) 34 (30.9) 8 (19.5) 6 (23.1) Long duration 356 (29.6) 228 (30.3) 39 (35.5) 11 (26.8) 10 (38.5) 2 (50.0) Diabetes history [n (%)] 301 (25.0) 211 (28.0) 30 (27.3) 6 (14.6) 10 (38.5) <0.001 Coronary heart disease history [n (%)] 156 (13.0) 92 (12.2) 55 (50.0) 30 (73.2) 17 (65.4) 4 (100.0) <0.001 Stroke history [n (%)] 15 (1.2) 7 (0.9) 6 (5.5) 3 (7.3) 4 (15.4) <0.001 OSA [n (%)] 947 (78.7) 596 (79.2) 88 (80.0) 35 (85.4) 17 (65.4) 3 (75.0) Supraventricular premature beats [n (%)] 1,134 (94.3) 706 (93.8) 108 (98.2) 41 (100.0) 25 (96.2) 4 (100.0) Ventricular premature beats [n (%)] 991 (82.4) 619 (82.2) 98 (89.1) 36 (87.8) 23 (88.5) 4 (100.0) Ventricular tachycardia [n (%)] 19 (1.6) 20 (2.7) 4 (3.6) Parallel rhythm [n (%)] 17 (1.4) 11 (1.5) Ventricular pre-excitation [n (%)] 8 (0.7) 6 (0.8) Atrioventricular block [n (%)] 31 (2.6) 24 (3.2) Atrial flutter [n (%)] 1 (0.1) 2 (0.3) Atrial tachycardia [n (%)] 352 (29.3) 196 (26.0) 27 (24.5) 10 (24.4) 6 (23.1) 1 (25.0) Atrial fibrillation [n (%)] 15 (1.2) 13 (1.7) Bundle branch block [n (%)] 70 (5.8) 43 (5.7) Sinoatrial block [n (%)] 22 (1.8) 12 (1.6) Long pause [n (%)] 24 (2.0) 22 (2.9) SDNN abnormality [n (%)] 226 (18.8) 160 (21.2) 60 (54.5) 15 (36.6) 8 (30.8) 1 (25.0) <0.001

Note: SDNN = standard deviation of NN intervals; BMI <18.5 kg/m² = underweight, 18.5-24.9 kg/m² = normal, 25-28 kg/m² = overweight, ≥28 kg/m² = obese.