Front. Public Health Frontiers in Public Health Front. Public Health 2296-2565 Frontiers Media S.A. 10.3389/fpubh.2025.1630953 Original Research Analysis of heart rate variability and its influencing factors in different blood pressure groups in Changchun City Guo YuHang 1 Data curation Visualization Formal analysis Validation Methodology Software Writing – review & editing Investigation Writing – original draft Li Jian 2 Writing – review & editing Conceptualization Supervision Project administration Data curation Guo Feng 1 Project administration Writing – review & editing Data curation Investigation Sun HaoRan 1 Validation Writing – review & editing Investigation Ning YiJin 1 Validation Investigation Writing – review & editing Wang XinYe 3 * Funding acquisition Supervision Conceptualization Resources Project administration Writing – review & editing Validation Methodology School of Clinical Medicine, Changchun University of Chinese Medicine, Changchun, China Office of the Ethics Committee, Affiliated Hospital of Changchun University of Chinese Medicine, Changchun, China School of Management, Changchun University of Chinese Medicine, Changchun, China Correspondence: XinYe Wang, 117186961@qq.com 15 12 2025 2025 13 1630953 19 05 2025 05 11 2025 29 11 2025 Copyright © 2025 Guo, Li, Guo, Sun, Ning and Wang. 2025 Guo, Li, Guo, Sun, Ning and Wang https://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Objective

To evaluate the Heart rate variability (HRV) in different blood pressure populations, analyze its influencing factors, rank their importance, and propose targeted health strategies.

 Methods

A total of 240 individuals with different blood pressure levels were recruited from a group in Changchun City, Jilin Province, using cluster sampling. Heart rate variability was assessed using the DHD-6000 HRV detector, and influencing factors were evaluated through questionnaires such as the “Health Risk Assessment Questionnaire,” Pittsburgh Sleep Quality Index (PSQI), Subjective Cognitive Decline Questionnaire (SCD-Q9), Generalized Anxiety Disorder 7-item scale (GAD-7), and Patient Health Questionnaire-9 (PHQ-9). Additionally, medical electronic sphygmomanometers (YXY-6) were used to collect blood pressure values. The study comprehensively evaluated Heart rate variability and its influencing factors in different blood pressure populations.

 Results

Among the different blood pressure populations in Changchun City, 12 individuals had hypotension (5.0%), 121 had normal blood pressure (50.4%), 87 had prehypertension (36.3%), and 20 had hypertension (8.3%). The detection rate of low TP was 30.4%, indicating weakened autonomic nervous system regulation in this population. Random forest model variable importance ranking showed that hyperlipidemia had the most significant impact on this population. Multivariate linear regression analysis revealed that blood pressure, lipids, marital status, alcohol consumption, age, and exercise had statistically significant effects on Heart rate variability (p < 0.05).

 Conclusion

Heart rate variability is influenced by multiple factors including lifestyle, blood pressure, lipids, age, and marital status. Among these, hyperlipidemia has the greatest impact on Heart rate variability. Compared to individuals with normal blood pressure, those with prehypertension and hypertension exhibit reduced autonomic nervous system activity and regulatory function. It is recommended to develop reasonable exercise plans, improve unhealthy lifestyle behaviors, control blood pressure and lipids, enhance autonomic nervous system function, and reduce the risk of cardiovascular events.

 heart rate variability blood pressure population autonomic nervous system function cardiovascular event risk health education and health promotion disease prevention The author(s) declared that financial support was received for this work and/or its publication. This study was supported by Jilin Province Science and Technology Development Plan Project (Application Research on the Use of Traditional Chinese Medicine (TCM) Characteristic Health Management in Chronic Disease Risk Models in Elderly Care Institutions) (20220203117SF). section-at-acceptance Public Health Education and Promotion Introduction

With the development of the socio-economic environment, improvements in living standards, and changes in lifestyle, the risk of cardiovascular events has increased, and the incidence rate is showing a trend toward younger populations. The activity and regulatory capacity of the autonomic nervous system are significantly related to the occurrence of cardiovascular events. Autonomic imbalance is an important factor in the occurrence of cardiovascular events and is also a direct manifestation of unhealthy behavioral lifestyles and psychological issues. A commonly used quantitative indicator to assess autonomic nervous activity is heart rate variability (HRV) (1–5). Research has found (6) that the measured values of HRV in hypertensive patients are relatively lower compared to healthy populations. The autonomic nervous system is affected, and the physical health condition is gradually deteriorating. This study focuses on analyzing the factors affecting HRV, conducting a comprehensive analysis of HRV levels and influencing factors in different blood pressure groups, and proposing targeted health management strategies.

 Objects and methods Survey participants

This study is a cross-sectional survey conducted in Changchun, Jilin Province, China, from January to May 2025. The participants were drawn from a resident health-management cohort established by a community health service center in an urban district of Changchun. A cluster-sampling method was used, with the following steps: first, five resident grids were randomly selected from all the grids managed by the community health service center. Then, all permanent residents within these five selected grids who met the inclusion criteria were defined as the survey “clusters.” We contacted every target resident in these grids through the community health-record system and invited them to take part in the survey. Considering an anticipated non-response rate of about 20% for a cross-sectional study, we initially set a target sample size of 300 participants. Ultimately, during the survey period, 260 individuals expressed willingness to participate; after data quality control, the final effective sample size for analysis was 240. According to the “Chinese Guidelines for the Prevention and Treatment of Hypertension (2018 revised edition),” all participants were required to rest in a quiet environment for at least 5 min, then have their seated blood pressure measured three times with a calibrated electronic sphygmomanometer, and the average of the three readings was used for group classification. Inclusion Criteria: ① Age ≥ 18 years; ② Consciousness clear, without cognitive impairment; ③ Adherence to the principle of informed consent, voluntary participation in this survey. Exclusion Criteria: ① Those with severe mental or cognitive impairments; ② Deafness or intellectual disabilities that prevent smooth communication; ③ Participants who dropped out of the study; ④ Those with mental illnesses that prevent self-evaluation; ⑤ Those with severe arrhythmias.

This study has been approved by the Ethics Committee of the Affiliated Hospital of Changchun University of Traditional Chinese Medicine (Approval number CCYFYLL-SQ-2024). We confirm that this research was conducted in accordance with the Declaration of Helsinki (1964) and its subsequent amendments. Before participating in this study, all participants signed an informed consent form. The survey began on January 3, 2025, and ended on May 3, 2025. Non-invasive cardiovascular diagnostic techniques were used, which did not cause physical harm to the participants. Participants’ privacy was strictly protected, and their data were anonymized. This study did not involve minors.

 Survey tools

① The “Health Risk Assessment Questionnaire” is used to investigate sociodemographic and behavioral lifestyle factors, including gender, age, marital status, educational level, monthly family income, smoking, alcohol consumption, dietary preferences, water drinking habits, and exercise frequency (7).

② The internationally recognized Pittsburgh Sleep Quality Index (PSQI) is applied to assess sleep quality. The total score ranges from 0 to 21. A PSQI score greater than 7 indicates poor sleep quality, while a score of 7 or less indicates good sleep quality (8). The PSQI has a test–retest reliability of 0.994, a split-half reliability coefficient of 0.824, and a Cronbach’s α coefficient of 0.842. Additionally, the structural validity of the PSQI is 0.76, and the classification validity is 0.81, indicating good reliability and validity (9).

③ The application of the Subjective Cognitive Decline Questionnaire-9 (SCD-Q9) indicates that a score of ≥5 suggests potential memory impairment. The Chinese version of the SCD-Q9 has a Cronbach’s α coefficient of 0.886 and a split-half reliability coefficient above 0.800, indicating good internal consistency (10).

④ The GAD-7 (Generalized Anxiety Disorder-7) scale is a brief self-report measure designed to assess the presence and severity of generalized anxiety disorder (GAD). According to the evidence provided, a score of ≥5 on the GAD-7 indicates the presence of an anxiety state. The Cronbach’s α coefficient for the GAD-7 is 0.867, which suggests good internal consistency reliability. Additionally, the test–retest reliability coefficient is 0.823, indicating that the GAD-7 has good reliability and validity (11).

⑤ The PHQ-9 is a self-administered questionnaire used to assess the severity of depressive symptoms. If the score on the PHQ-9 is ≥5, it indicates the presence of depressive symptoms. The internal consistency coefficient of the PHQ-9, Cronbach’s α, is 0.8325, and the test–retest reliability coefficient is 0.932. The sensitivity of the PHQ-9 is 88%, and the specificity is 99%, indicating good reliability and validity (12).

⑥ The DHD-6000 heart rate variability (HRV) detector is used to measure the heart rate variability of subjects. HRV frequency domain analysis can accurately and systematically evaluate the autonomic nervous function comprehensively (13). The significance of common frequency domain indicators is shown in Table 1. Since the actual values of HRV frequency domain indicators do not follow a normal distribution, logarithmic transformation is applied to each HRV frequency domain indicator to achieve a normal distribution before statistical analysis. Normal reference values for short-term HRV frequency domain analysis: TP (6.7 ~ 8.1 ln(ms2)), LF (4.7 ~ 7.0 ln(ms2)), HF (3.5 ~ 6.8 ln(ms2)) (14).

Commonly used frequency domain indicators in HRV analysis.

Frequency domain indicator Description Unit Meaning
TP Total power ms2 Reflects the overall activity of the autonomic nervous system
LF Low frequency Power ms2 Closely related to blood pressure changes, mainly reflects the tension of the sympathetic and parasympathetic nervous systems
HF High frequency power ms2 Related to respiratory arrhythmia, mainly reflects the tension of the parasympathetic nervous system

⑦ The YXY-6 medical electronic blood pressure monitor is used to measure the blood pressure levels of subjects. According to the hypertension prevention and treatment guidelines in China, blood pressure is classified into three categories: normal blood pressure (systolic pressure <120 mmHg and diastolic pressure <80 mmHg), Prehypertension (systolic pressure 120–139 mmHg and/or diastolic pressure 80–89 mmHg), and hypertension (systolic pressure ≥140 mmHg and/or diastolic pressure ≥90 mmHg). This classification is applicable to adults aged 18 years and older (15).

 Statistical methods

The database was established using Epidata 3.1 software, independent sample T-tests and one-way ANOVA were conducted using SPSS 26.0 software, related variables were visualized using GraphPad Prism 8.0 software, Random forest model analysis was performed using R 4.3.3 software, variable selection was conducted using LASSO regression, and the selected variables were included in multiple linear regression analysis with a significance level of α = 0.05.

 Quality control

Data on different blood pressure groups and heart rate variability were collected to fully understand the background of this study. Surveyors underwent professional training, were clear about the matters that needed attention during the survey process, strictly controlled the use of instruments according to taboos, patiently explained ambiguous questions, and corrected any issues found promptly. After the survey, the surveyors reviewed all questionnaires twice, eliminated invalid ones, checked for any issues, and signed off to ensure the completeness and validity of the survey questionnaires.

 Results Comparison of HRV frequency domain indicators among different blood pressure characteristics

In this population, there were 12 individuals with low blood pressure, accounting for 5%; 121 individuals with normal blood pressure, accounting for 50.4%; 87 individuals with Prehypertension, accounting for 36.3%; and 20 individuals with hypertension, accounting for 8.3%. The detection rate of low total power (TP) was 30.4%, indicating that the overall autonomic nervous system regulation ability of this population is poor; the detection rate of low frequency power (LF) was 15%, suggesting weakened sympathetic nerve activity; the detection rate of high LF was 20.8%, indicating excessive sympathetic nerve activity. The detection rate of high frequency power (HF) was 35%, excessive parasympathetic activity, which is related to the physiological demands of the body during relaxation and recovery, suggests a tendency toward symptoms such as depression, fatigue, bradycardia, and lowered blood pressure (14), as shown in Table 2.

Characteristics of blood pressure and HRV frequency domain indicators in different blood pressure populations (n%).

Item Feature n(%)
Blood Pressure Low blood pressure 12(5.0)
Normal blood pressure 121(50.4)
Prehypertension 87(36.3)
High blood pressure 20(8.3)
TP Low 73(30.4)
Normal 108(45.0)
High 59(24.6)
LF Low 36(15.0)
Normal 154(64.2)
High 50(20.8)
HF Low 9(3.8)
Normal 147(61.3)
High 84(35.0)

The research results show that there are significant differences between the different blood pressure groups and all HRV indices (p < 0.001). Effect size analysis indicates that blood pressure has a moderate-to-large impact on TP and HF (η2 = 0.118) and a small-to-moderate impact on LF (η2 = 0.082), as shown in Table 3. Within-group comparisons reveal that, compared with the normal-blood-pressure group, the pre-hypertension and hypertension groups exhibit a significant reduction in HRV, as illustrated in Figures 1–3.

Comparison of HRV frequency domain indicators for different blood pressure characteristics (−x ± s).

Feature n TP LF HF
Low blood pressure 12 7.22 ± 1.57 5.71 ± 1.54 6.33 ± 1.91
Normal blood pressure 121 7.63 ± 0.96 6.33 ± 1.10 6.49 ± 1.26
Prehypertension 87 6.90 ± 1.12 5.64 ± 1.32 5.54 ± 1.48
High blood pressure 20 6.65 ± 0.57 5.51 ± 0.98 5.24 ± 1.25
F 10.47 6.896 10.511
p <0.001 <0.001 <0.001

Comparison of different blood pressure levels and TP.

Bar chart showing TP levels across four blood pressure categories: low, normal, prehypertension, and high. Bars represent mean values with error bars. Significant differences are indicated with p.adj less than 0.05.

Comparison of different blood pressure levels and LF.

Bar chart showing LF values across four blood pressure categories: Low, Normal, Prehypertension, and High. Each bar has error bars, with values around 6 to 7. A bracket indicates significance with P.adj less than 0.05.

Comparison of different blood pressure levels and HF.

Bar graph showing HF levels across four blood pressure categories: low, normal, prehypertension, and high. Each bar includes error bars. A p-value less than 0.05 indicates statistical significance. Single-factor analysis of heart rate variability in different blood pressure groups Demographic and sociological characteristics

The research results show that there are significant differences in the frequency domain indicators of HRV, such as TP, LF, and HF, among different age groups (p < 0.05). Specifically, TP values decrease with age, indicating a reduction in the overall activity level and regulatory capacity of the autonomic nervous system as people age. There are also significant differences in HRV frequency domain indicators between individuals with different marital statuses and educational levels (p < 0.05), with unmarried individuals showing better autonomic nervous system regulation compared to married individuals. TP and LF also show significant differences among individuals with different monthly family incomes (p < 0.05), while gender does not show significant differences in any HRV indicator (p > 0.05), as shown in Table 4.

Analysis of the impact of demographic and sociological characteristics on heart rate variability (−x ± s).

Item Feature n TP LF HF
Gender Male 71 7.29 ± 1.33 6.11 ± 1.47 5.89 ± 1.69
Female 169 7.25 ± 1.02 5.93 ± 1.14 6.09 ± 1.35
t 0.228 1.043 −0.967
p 0.82 0.298 0.335
Age 18–40 years old 136 7.70 ± 0.92 6.48 ± 1.05 6.53 ± 1.21
41–65 years old 79 6.74 ± 0.92 5.43 ± 0.99 5.36 ± 1.32
≥66 years old 25 6.52 ± 1.54 5.03 ± 1.07 5.42 ± 0.22
F 30.681 32.826 22.019
p <0.001 <0.001 <0.001
Marital status Unmarried 81 8.01 ± 0.73 6.74 ± 0.93 6.96 ± 1.03
Married 146 6.91 ± 1.09 5.63 ± 1.19 5.59 ± 1.40
Divorced 7 6.95 ± 0.40 5.73 ± 0.47 5.55 ± 0.53
Widowed 6 6.09 ± 1.57 4.52 ± 1.90 4.62 ± 2.38
F 25.172 20.853 22.076
p <0.001 <0.001 <0.001
Education level Primary School 10 6.78 ± 1.10 5.54 ± 1.26 5.30 ± 1.07
Junior High School 25 6.62 ± 1.41 5.20 ± 1.52 5.33 ± 1.85
High School, Vocational School 79 6.80 ± 0.94 5.50 ± 1.09 5.42 ± 1.25
College, Bachelor’s Degree 126 7.72 ± 0.96 6.47 ± 1.07 6.61 ± 1.28
F 18.217 16.921 16.845
p <0.001 <0.001 <0.001
Monthly household income No Income 4 7.52 ± 0.32 6.21 ± 0.69 6.25 ± 0.27
Below 1,000 Yuan 14 7.42 ± 1.05 6.29 ± 1.23 6.03 ± 1.53
Between 1,000 and 3,000 Yuan 70 7.26 ± 1.09 5.96 ± 1.28 5.98 ± 1.31
Between 3,000 and 5,000 Yuan 98 7.45 ± 1.12 6.18 ± 1.20 6.31 ± 1.49
More than 5,000 Yuan 54 6.87 ± 1.15 5.55 ± 1.24 5.57 ± 1.53
F 2.601 2.606 2.333
p 0.037 0.037 0.057
Behavioral lifestyle

The research results show that there are significant differences in HRV indicators among individuals with different drinking habits, dietary preferences, and exercise patterns (p < 0.05). Specifically, individuals who consume alcohol have higher values of LF and HF compared to those who do not drink alcohol, indicating that alcohol consumers exhibit higher sympathetic and parasympathetic nervous activity than non-drinkers. Smoking and water drinking habits do not show significant differences in HRV indicators (p > 0.05), as shown in Table 5.

Analysis of factors influencing heart rate variability by behavioral lifestyle (−x ± s).

Item Feature n TP LF HF
Smoking No 177 7.30 ± 1.13 5.99 ± 1.26 6.11 ± 1.49
Yes 63 7.16 ± 1.09 5.94 ± 1.22 5.80 ± 1.36
t 0.907 0.303 1.537
p 0.366 0.762 0.127
Drinking No 171 7.02 ± 1.12 5.73 ± 1.24 5.73 ± 1.47
Yes 69 7.86 ± 0.87 6.60 ± 1.04 6.78 ± 1.13
t −5.544 −5.533 −5.37
p <0.001 <0.001 <0.001
Exercise Less than once a month 23 7.32 ± 1.30 5.91 ± 1.41 6.20 ± 1.91
2–3 times a month 62 6.95 ± 1.25 5.64 ± 1.40 5.64 ± 1.56
2–3 times a week 70 7.15 ± 1.01 5.82 ± 1.11 5.91 ± 1.31
3–5 times a week 61 7.56 ± 0.96 6.36 ± 1.07 6.34 ± 1.28
Almost every day 24 7.59 ± 1.06 6.41 ± 1.16 6.42 ± 1.36
F 3.088 3.731 2.482
p 0.017 0.006 0.045
Diet preference Light taste 121 7.20 ± 1.13 5.94 ± 1.30 5.92 ± 1.49
Heavy salt 53 7.67 ± 1.03 6.40 ± 1.16 6.59 ± 1.25
Likes high-calorie food 66 7.05 ± 1.10 5.71 ± 1.13 5.78 ± 1.47
F 5.128 4.77 5.374
p 0.007 0.009 0.005
Drinking water habits Drinking water on an empty stomach in the morning 62 7.28 ± 1.13 5.99 ± 1.24 6.01 ± 1.54
No habit of drinking water on an empty stomach, but drinking water in the afternoon 44 7.34 ± 1.02 6.04 ± 1.21 6.20 ± 1.18
Drinking water irregularly 80 7.39 ± 1.15 6.13 ± 1.26 6.17 ± 1.52
Only drink water when thirsty” 54 7.00 ± 1.13 5.70 ± 1.24 5.70 ± 1.46
F 1.464 1.331 1.396
p 0.225 0.265 0.245
Physiological indicators

The research results show that there are significant differences (p < 0.05) in HRV indicators among patients with diabetes, hyperlipidemia, coronary heart disease, and poor sleep quality. Patients with coronary heart disease have significantly lower values of TP, LF, and HF compared to those without the disease, indicating weaker activity and regulatory capacity of the autonomic nervous system. There is a significant difference between BMI categories and HF (p < 0.05), but no significant difference with TP and LF (p > 0.05). Specifically, individuals who are obese or overweight have slightly lower parasympathetic activity compared to those with normal body weight, as shown in Table 6.

Analysis of factors affecting heart rate variability by physiological indicators (−x ± s).

Item Feature n TP LF HF
BMI classification Underweight 22 7.84 ± 1.40 6.25 ± 1.47 6.87 ± 1.94
Normal weight 115 7.17 ± 1.03 5.88 ± 1.13 5.99 ± 1.35
Overweight 88 7.24 ± 1.12 6.01 ± 1.32 5.91 ± 1.42
Obese 15 7.28 ± 1.16 6.17 ± 1.32 5.77 ± 1.42
F 2.288 0.745 2.839
p 0.079 0.526 0.039
Diabetes No 216 7.33 ± 1.09 6.04 ± 1.22 6.12 ± 1.42
Yes 24 6.65 ± 1.25 5.44 ± 1.33 5.19 ± 1.58
t 2.538 2.106 2.776
p 0.017 0.044 0.01
Hyperlipidemia No 158 6.81 ± 0.95 5.39 ± 0.98 5.51 ± 1.32
Yes 82 8.13 ± 0.88 7.11 ± 0.87 7.03 ± 1.17
t −10.755 −13.942 −9.129
p <0.001 <0.001 <0.001
Coronary heart disease No 218 7.37 ± 1.09 6.08 ± 1.23 6.15 ± 1.42
Yes 22 6.24 ± 0.92 4.97 ± 0.94 4.87 ± 1.33
t 5.381 4.107 4.278
p <0.001 <0.001 <0.001
Sleep quality Poor 106 7.03 ± 1.22 5.75 ± 1.34 5.76 ± 1.59
Good 134 7.45 ± 1.00 6.16 ± 1.34 6.24 ± 1.32
t −2.968 −2.591 −2.509
p 0.003 0.01 0.013
Physiological factors

Research results show that anxiety, cognitive impairment, and various HRV indicators have significant differences (p < 0.05). Specifically, individuals with anxiety states have higher values of LF and HF, indicating increased sympathetic and parasympathetic nervous activity, leading to an imbalance in autonomic nervous system regulation. There is no significant difference between depression and HRV indicators (p > 0.05), as shown in Table 7.

Analysis of factors influencing heart rate variability by psychological factors (−x ± s).

Item Feature n TP LF HF
Depression No 233 7.27 ± 1.12 5.99 ± 1.24 6.04 ± 1.47
Yes 7 6.89 ± 1.11 5.67 ± 1.56 5.55 ± 1.20
t 0.895 0.546 1.057
p 0.404 0.604 0.328
Anxiety No 105 7.03 ± 1.22 5.75 ± 1.34 5.56 ± 1.59
Yes 135 7.45 ± 1.00 6.16 ± 1.14 6.24 ± 1.31
t −2.941 −2.478 −2.52
p 0.004 0.014 0.013
Cognitive impairment No memory problems 203 7.40 ± 1.02 6.14 ± 1.15 6.17 ± 1.35
Possible memory issues 37 6.50 ± 1.31 5.13 ± 1.40 5.25 ± 1.78
t 3.951 4.141 3.01
p <0.001 <0.001 0.004
Impact factors of heart rate variability in different blood pressure groups Variable importance ranking

Taking the various frequency domain indicators of HRV (TP, LF, and HF) as dependent variables, and the factors that significantly influenced HRV in the univariate analysis of different blood pressure groups as independent variables, a random forest model was constructed using the Random Forest package in R studio software. 20% of the samples were used as the test set, and 80% as the training set. The Importance function was applied to rank the variables, where %Inc. MSE represents the increase in mean squared error (Mean Squared Error). The larger the %Inc. MSE value, the greater the importance of the variable in the influencing factors (16). The variable assignment situation of the random forest model is shown in Table 8. The results showed that the highest importance variables for TP, LF and HF were all high lipid levels, as detailed in Figure 4.

Variable assignment of random forest model for HRV influencing factors.

Related variables Assignment
Age X1 18–40 years = 1,41–65 years = 2,≥66 years = 3
Marital status X2 Unmarried = 1, Married = 2, Divorced = 3, Widowed = 4
Education level X3 Primary school = 1, Junior high school = 2, High school, technical school = 3, College, undergraduate = 4
Monthly household income X4 No income = 1, Below 1,000 yuan = 2,1,000–3,000 yuan = 3,3,000–5,000 yuan = 4, Above 5,000 yuan = 5
Drinking X5 No = 0, Yes = 1
Exercise X6 Less than once a month = 1, 2–3 times a month = 2, 1–2 times a week = 3, 3–5 times a week = 4, About once a day = 5
Dietary preference X7 Light taste = 1, Heavy salt intake = 2, Likes high-calorie foods = 3
Blood pressure X8 Low blood pressure = 1, Normal blood pressure = 2, Prehypertension = 3, High blood pressure = 4
Diabetes X9 No = 0, Yes = 1
Hyperlipidemia X10 No = 0, Yes = 1
Coronary heart disease X11 No = 0, Yes = 1
BMI classification X12 Underweight = 1, Normal weight = 2, Overweight = 3, Obese = 4
Sleep quality X13 Poor = 1, Good = 2
Anxiety X14 No = 0, Yes = 1
Cognitive impairment X15 No memory problems = 0, Possible memory problems = 1

Importance ranking of TP influencing factors.

Dot plot showing the importance of various factors in increasing mean squared error (%IncMSE). Hyperlipidemia and marital status have the highest %IncMSE around 40, followed by other factors like alcohol consumption, blood pressure, and age. Variable selection

Based on the results of Figure 4, the glment function in R Studio was used to incorporate variables that showed significant differences in the univariate HRV analysis into LASSO regression models, as illustrated in Figure 5. The left and right vertical dashed lines represent lambda.min (the penalty coefficient λ corresponding to the minimum cross-validation error) and lambda.1se (the penalty coefficient λ corresponding to the minimum cross-validation error plus one standard error), respectively. Mean-squared error (MSE) is a core metric for measuring the discrepancy between a model’s predictions and the true values; log(λ) denotes the natural logarithm of the penalty coefficient λ. Within the [lambda.min, lambda.1se] interval, the model’s error variation is minimal (15). When λ equals 0.02419 (the left dashed line, lambda.min), the TP model achieves the smallest error, with 10 variables selected for TP. When λ equals 0.01973, the LF model attains the smallest error, also with 10 variables selected for LF. When λ equals 0.03348, the HF model reaches the smallest error, with 12 variables selected for HF. These selected variables were then entered into multivariate linear regression models (see Figures 5, 6 for details).

Feature variable screening based on LASSO regression – TP.

Line graph showing mean-squared error versus log of lambda (λ) values. The mean-squared error remains stable between log λ of negative six to negative two, then increases sharply. Red points represent data, with error bars illustrating variability.

Lasso regression analysis variable interpretation plot – TP.

Line graph depicting the coefficients of a model as a function of the logarithm of lambda. Multiple lines show how coefficients shrink toward zero as lambda increases from negative six to negative one. Multifactorial analysis of heart rate variability in different blood pressure populations

To develop the final predictive model, we performed a multiple stepwise linear regression analysis using IBM SPSS Statistics software. All effective variables identified by the LASSO regression were entered as candidate independent variables. In the regression dialog box, the method was set to “Stepwise.” SPSS’s stepwise regression automatically selects variables based on the p-value of the partial F-test, with the default entry criterion of p ≤ 0.05 and removal criterion of p ≥ 0.10, which aligns with the objectives of this study. Ultimately, only those variables that remained significant (p < 0.05) after adjusting for the other variables were retained in the final multiple linear regression model. The results indicated that factors such as age, marital status, alcohol consumption, exercise, blood pressure, and hyperlipidemia all affect autonomic nervous system balance in this population, as shown in Table 9.

Multivariate linear regression analysis of factors affecting heart rate variability.

Dependent variable Independent variable β SE Beta t p VIF
TP Constant 8.566 0.214 40.02 <0.001
Hyperlipidemia 1.152 0.104 0.489 11.075 <0.001 1.042
Marital status −0.431 0.093 −0.244 −4.654 <0.001 1.466
Drinking 0.407 0.113 0.165 3.599 <0.001 1.124
Blood pressure −0.256 0.07 0.164 −3.636 <0.001 1.093
Age −0.279 0.089 −0.169 −3.142 0.002 1.544
LF Constant 6.786 0.239 28.383 <0.001
Hyperlipidemia 1.537 0.104 0.587 14.719 <0.001 1.043
Age −0.425 0.089 −0.231 −4.751 <0.001 1.551
Marital status −0.382 0.093 −0.194 −4.113 <0.001 1.466
Drinking 0.252 0.117 0.092 2.144 0.033 1.201
Exercise 0.119 0.045 0.109 2.653 0.009 1.110
Blood pressure −0.179 0.071 −0.103 −2.504 0.013 1.118
HF Constant 7.678 0.298 25.742 <0.001
Hyperlipidemia 1.3 0.148 0.423 8.755 <0.001 1.041
Marital status −0.707 0.114 −0.307 −6.219 <0.001 1.090
Blood pressure −0.423 0.100 −0.209 −4.247 <0.001 1.080
Drinking 0.654 0.156 0.203 4.202 <0.001 1.045
Discussion

According to the results of demographic studies and HRV research, the frequency domain indicators of HRV decrease with age, indicating that the overall activity level and regulatory capacity of the autonomic nervous system weaken as people age. This is consistent with the findings of Wang (14), who studied 483 civil servants in a direct-controlled organ in Changchun. The results showed that TP, LF, and HF values decreased with age, indicating that the regulatory capacity of the autonomic nervous system weakens as people age. Jiang et al. (17) pointed out that age becomes one of the risk factors for autonomic nervous dysfunction. As age increases, the body’s autonomic nervous control function degenerates, leading to a shift in the balance between the sympathetic and parasympathetic nerves toward the sympathetic side. Wu (18) also showed that various indicators of HRV decrease with age, indicating that the autonomic nervous system is affected and overall health declines. Marital status is another factor affecting HRV. Unmarried individuals have higher values of TP, LF, and HF compared to married or divorced individuals, indicating that unmarried individuals have better autonomic nervous regulation ability.

Research results on lifestyle behaviors and HRV show that alcohol consumption has a significant impact on HRV. Karpyak et al. (19) found that high alcohol intake is associated with a decrease in HRV. Chen et al. (20) discovered that HRV is influenced by physical activity, and the intensity of physical activity is positively correlated with the sympathetic nervous system. Fu et al. (21) analyzed the effects of Tai Chi and fitness exercises, both moderate-intensity physical activities, on HRV in patients with primary hypertension. The results showed that these two exercises can effectively improve the autonomic nervous function of patients with primary hypertension.

The physiological indicators and HRV research results show that blood pressure has a significant impact on HRV. Jarczok et al. (22) found through a health assessment of 9,550 working adults that HRV may be associated with an increased risk of cardiovascular events. This study confirmed for the first time the correlation between reduced HRV and elevated cardiovascular risk, suggesting that HRV could potentially become a simple and novel indicator for assessing cardiovascular risk (23). The research results of Fang et al. (24) indicate that a decrease in HRV indicators is significantly associated with an increased risk of cardiovascular events. Yue et al. (25) found that both patients with masked hypertension and those with primary hypertension had significantly lower HRV parameters compared to the normal population, but there was no significant difference between the two groups. Virtanen et al. (6) studied and found that all absolute measurements of HRV in hypertensive patients were relatively lower compared to the healthy population. Zhao (26) found in a study on HRV and prehypertension that individuals with excessive sympathetic tone had a higher detection rate of prehypertension than those with normal tone, and elevated HRV significantly increased the risk of developing prehypertension. Moreover, related clinical research has shown that the sympathetic nervous system can promote angiogenesis and tumor progression in various cancers (27).

Furthermore, this study has several limitations. First, the sample was sourced from a single urban center, which limits the external validity of the findings when generalizing to other populations. Second, the sample size was relatively limited, particularly within the hypotension and hypertension subgroups, which may affect the statistical power of complex multivariate models and increase the risk of Type II errors. Therefore, analyses focusing on these subgroups should be considered exploratory. Finally, although we included various demographic and behavioral factors in our analysis, we did not comprehensively measure and control for known confounders affecting HRV, such as detailed caffeine intake and specific medication history, which might have influenced the precision of the results. Despite these limitations, this study, utilizing modern statistical methods like Random Forest and LASSO regression on a systematically quality-controlled community sample, identified hyperlipidemia as the most significant factor influencing HRV, with other notable factors including blood pressure, marital status, alcohol consumption, age, and physical exercise. These findings have important public health implications. Based on the above findings, we recommend the following measures for the community population to maintain autonomic nervous system health: advocate for regular physical exercise and develop personalized exercise plans; strengthen the management of blood pressure and blood lipids, reduce the intake of high-calorie and fried foods, and control alcohol consumption; simultaneously, promote public awareness and knowledge regarding the maintenance of autonomic nervous system health. Future studies should expand the sample size, include more blood pressure abnormality subgroups, and conduct long-term follow-up to clarify the causal relationship between HRV and cardiovascular events, thereby providing stronger evidence for the early prevention of cardiovascular diseases.

 Data availability statement

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

 Ethics statement

The studies involving humans were approved by Ethics Committee of the Affiliated Hospital of Changchun University of Traditional Chinese Medicine (Approval number 75 CCYFYLL-SQ-2024). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

 Author contributions

YG: Data curation, Visualization, Formal analysis, Validation, Methodology, Software, Writing – review & editing, Investigation, Writing – original draft. JL: Writing – review & editing, Conceptualization, Supervision, Project administration, Data curation. FG: Project administration, Writing – review & editing, Data curation, Investigation. HS: Validation, Writing – review & editing, Investigation. YN: Validation, Investigation, Writing – review & editing. XW: Funding acquisition, Supervision, Conceptualization, Resources, Project administration, Writing – review & editing, Validation, Methodology.

 Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

 Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

 Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

 References JiangD YuG TianJ QianY ZhongJ LiuX . Evaluation of heart rate variability monitoring on autonomic nervous function in children with hyperthyroidism. Chin J Pract Pediatr. (2002) 17:152–4. VybiralT BrygRJ MaddensME BhasinSS CroninS BodenWE . Effects of transdermal scopolamine on heart rate variability in normal subjects. Am J Cardiol. (1990) 65:604–8. doi: 10.1016/0002-9149(90)91038-8, 2309630 KjellgrenO GomesJA. Heart rate variability and baroreflex sensitivity in mvocardial infarction. Am Heart J. (1993) 125:204–15. QuH LiuR. Research progress on exercise therapy for cardiovascular disease prevention and treatment based on heart rate variability analysis. Chin J Gerontol. (2009) 12:3309–11. LiuX MiaoD WangW XiaoW LiuF JingX . Study on heart rate variability in normal young males under different stress situations. Chin J Behav Med. (2002) 11:679–80. VirtanenR JulaA KuuselaT HeleniusH Voipio-PulkkiLM. Reduced heart rate variability in hypertension: associations with lifestyle factors and plasma renin activity. J Hum Hypertens. (2003) 17:171–9. doi: 10.1038/sj.jhh.1001529 GuoY SunH WangZ HanY WangX. Study on heart rate variability and influencing factors in an elderly population in Changchun City. Chin J Gerontol. (2024) 44:5867–71. doi: 10.3969/j.issn.1005-9202.2023.24.056 HanY WangZ SunH . Factors affecting sleep quality in elderly people with different traditional Chinese medicine constitutions. Chin J Gerontol. (2023) 43:6119–21. doi: 10.3969/j.issn.1005-9202.2024.23.052 LiuR. Regression analysis of constipation constitution factors in the elderly and construction of a risk constitution adjustment scheme. Nanning: Guangxi University of Chinese Medicine (2022). HaoL HuX HanY JiaJ. Translation and reliability and validity analysis of the English version of the subjective cognitive decline questionnaire. Chin Gen Pract. (2019) 22:3238–45. doi: 10.12114/j.issn.1007-9572.2019.00.045 WangL LuK WangC ShengL HuD DingR. Reliability and validity analysis of GAD-2 and GAD-7 in anxiety screening in cardiovascular outpatient clinics. Sichuan Ment Health. (2014) 27:198–201. doi: 10.3969/j.issn.1007-3256.2014.03.002 XuY WuH XuY. Application of the patient health questionnaire depression scale (PHQ-9) in community elderly population: reliability and validity analysis. Shanghai J Psychiatry. (2007) 5:257–259+276. XiaoS FangX SunX. Observation on the influence of three factors on heart rate variability. Beijing Biomed Eng. (2002) 21:276–8. WangX. Analysis of heart rate variability and influencing factors among public officials in Changchun City (Doctoral dissertation). Jilin University (2011). Chinese Hypertension Prevention and Treatment Guidelines Revision Committee. Chinese guidelines for hypertension prevention and treatment 2010. Chin J Cardiovasc Dis. (2011) 39:579–616. YuMingxuan WangXing. Research on the characteristics of heart rate variability changes before and after acute exercise in female college students with hidden obesity. Abstracts of the 13th National Congress of the Chinese Society of Sports Sciences—Wall Poster Session (physical fitness and health branch) (part III). Shanghai, China: Shanghai Yangsi High School; Shanghai Sport University, (2023), 163–164. JiangG GongL GanL. Relationship between urinary microalbumin creatinine ratio and autonomic nervous function in hypertensive patients. China ChronDis Prev Control. (2019) 27:863–6. doi: 10.16386/j.cjpccd.issn.1004-6194.2019.11.017 WuM. Study on heart rate variability and influencing factors in different population characteristics (Master’s thesis). Shenyang Medical College, Shenyang, China. (2018). KarpyakVM RomanowiczM SchmidtJE LewisKA BostwickJM. Characteristics of heart rate variability in alcohol-dependent subjects and nondependent chronic alcohol users. Alcohol Clin Exp Res. (2014) 38:9–26. doi: 10.1111/acer.12270, 24117482 ChenLidan PengLinyu LiaoJingwen. Study on the related risk factors of autonomic nervous system in overweight or obese adults/Chinese Society of Sports Sciences. Proceedings of the 12th National Conference on Sports Sciences – Wall Poster Session (Physical Fitness and Health Branch). Guangzhou Sport Institute, (2022):3. FuJ ChenZ LiuL FangH MaY CuiM . Effects of tai chi and fitness exercises on the autonomic nervous system in patients with essential hypertension. Chin J Hyper. (2024) 32:763–71. doi: 10.16439/j.issn.1673-7245.2024.08.010 JarczokMN KoenigJ WittlingA FischerJE ThayerJF. 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 Clin Med. (2019) 8:1940. doi: 10.3390/jcm8111940, 31717972 HuZ YaoY. Historical development and research progress of heart rate variability. Pract Electrocardiol J. (2023) 32:355–9. doi: 10.13308/j.issn.2095-9354.2023.05.010 FangSC WuYL TsaiPS. Heart rate variability and risk of all-cause death and cardiovascular events in patients with cardiovascular disease: a meta-analysis of cohort studies. Biol Res Nurs. (2020) 22:45–56. doi: 10.1177/1099800419877442, 31558032 YueWW YinJ ChenB ZhangX WangG LiH . Analysis of heart rate variability in masked hypertension. Cell Biochem Biophys. (2014) 70:201–4. doi: 10.1007/s12013-014-9882-y, 24639105 ZhaoL. Research on risk factors and health management strategies for Normal high blood pressure in medical students (Master’s thesis/Doctoral dissertation). Jilin University, Changchun, China. (2016). LiF XuP LiD. Research progress on the interaction between the autonomic nervous system and cancer. J Nanjing Med Univ. (2024) 44:1408–18. doi: 10.7655/NYDXBNSN240434

Edited by: Deep Shikha, Swami Rama Himalayan University, India

Reviewed by: Rohan Bodapati, University of Wisconsin-Madison, United States

Gagan Kaur, Uttar Pradesh University of Medical Sciences, India