Network analysis of resilience, anxiety and depression in clinical nurses

Abstract

Background

Resilience is a protective feature against anxiety and depression disorders. However, the precise relationship and structure of resilience and anxiety and depression remain poorly understood. This study sought to investigate the link among resilience’ components and anxiety as well as depression.

Methods

1,279 clinical nurses were recruited. 10-item Connor-Davidson Resilience Scale, Generalized Anxiety Disorder 7, and Patient Health Questionnaire 9 were employed to evaluate resilience, anxiety, and depression, respectively. The regularized partial-correlation network was generated utilizing data from cross-sectional survey and the bridge expected influence index was utilized to quantify bridge components.

Results

The rates of anxiety and depression within clinical nurses were 67.3% and 67.2%, accordingly. Four strongest bridge edges appeared in the resilience-anxiety network, like “Adapt to change”- “Fear that something might happen”, and “Stay focused under pressure”- “Uncontrollable worry”. Two strongest bridge edges appeared in the resilience-depression network, like “Adapt to change”- “Concentration difficulties” and “Stay focused under pressure”- “Fatigue”. “Adapt to change” was recognized as bridging nodes in both the resilience-anxiety network and the resilience-depression network.

Conclusions

Interventions targeting the bridge component “Adapt to change” within resilience, may mitigate the intensity of anxiety and depression symptoms among clinical nurses.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12888-024-06138-8.

Introduction

Nurses are an indispensable part of healthcare systems and their well-being, psychological health and work performance have a crucial influence on the quality of hospital care. However, they encountered several pressures, including a demanding workload imposed by the stringent demands of the strained healthcare system, night work, work-family conflict, rendering them more exposed to mental disorders [1, 2]. The emergence of coronavirus disease 2019 (COVID-19) has culminated in considerable psychological issues for healthcare professionals in China, particularly nurses [3, 4]. Previous evidence has demonstrated that depression and anxiety were prevalent mental disorders among clinical nurses [5–7]. According to a recent survey of clinical nurses from 30 provinces in China, 55.5% clinical nurses reported depression symptom and 41.8% reported anxiety symptom [8]. Anxiety and depression have been strongly linked with numerous adverse outcomes, involving reduced healthcare quality, staff attrition, heightened risk of medical errors, and lower patient satisfaction [9–11]. Moreover, depression and anxiety also affect nurses’ quality of life and well-being [12]. Both psychiatric disorders have detrimental effects on individuals as well as the healthcare system due to their significant incidence and adverse repercussions in clinical nurses. Thus, more research should focus on these two psychiatric disorders in nurses. Efforts are needed to explore effective measures and interventions for reducing anxiety and depression among nurses.

Resilience is regarded as a changing, developmental, and psychosocial process in which persons subjected to prolonged adversity or possibly detrimental experiences achieve beneficial psychological adaption over time [13]. Resilience exerts protective influence on the person’s mental and physical well-being, facilitating their capacity to handling adversity or stressful situations efficiently. Resilience is not merely an individual feature; it encompasses the interplay of inside resilience features and outside environment factors [14]. Previous studies have found that individuals with higher resilience levels could protect them against psychiatric disorders [15–17]. Much early research has examined that the preventive impact of resilience on anxiety and depression symptoms among distinct population such as patients with cardiovascular diseases [18], medical staff [19], university students [20], and pregnant women [21]. Resilience was also demonstrated as a key psychological intervention target to decrease anxiety and depression symptoms among clinical nurses [17].

However, previous research examining the connection between resilience and anxiety, depression is frequently evaluated collectively based on the aggregate scores of assessments, which ignored different components within these psychological variables and obscured the fine-grained connection among them [22]. Each item within the measures reflects a different characteristic of the psychological variables. The strength and nature of the interactions between resilience items and those syptoms of depression and anxiety may vary. Prior correlation evidence failed to identify intervention targets at a symptom level, largely due to lack of analysis approach that measures the relative significance of items [23]. Network analysis (NA) is a powerful approach for displaying complex connections among individual’ s symptom of mental disorders [24], supporting that the emergence and progression of psychological conceptions or mental diseases are sustained by the dynamic interplay of symptoms. Compared to traditional statistical correlation approach, NA enables the visualization of intricate interactions among psychological constructs at a more granular level (e.g., components and symptoms), thereby mitigating spurious associations arising from the multitude of constructs. Moreover, NA also can provide centrality for each node within the network to reflect the degree of importance of each node. Core symptoms in mental disorders may serve as useful and efficient targets for treatment interventions.

In recent years, the NA method has been increasingly applied in the psychology field. Numerous studies have utilized NA to explore the pathway mechanisms underlying comorbid mental disorders, pinpointing symptoms that are more impactful regarding their impacts [25–27]. Additionally, NA has been extensively applied to various psychological constructs, including personality, suicide ideation, burnout, and resilience [23, 28–30]. A previous study explored the connections at the item level between resilience, burnout, and coping using NA [30]. The findings highlighted distinct expected influences of resilience items on burnout and coping, identifying the item “I am able to adapt when changes occur” key bridge components between these constructs, which provided potential targets for intervention [30]. Researchers can better understand the mechanisms underpinning component-to-symptom interaction and provide more accurate intervention targets at a finer level with the use of NA. Furthermore, Network Intervention Analysis (NIA) has been employed to analyse the treatment-induced sequence of changes in symptoms and/or variables, facilitating the identification of those affected directly or indirectly by certain therapies [31–33]. It also assists health psychologists in detecting risky behaviours aiding in primary as well as secondary prevention, and assessing treatment efficacy [34]. Appling NA to analyse the finer connection between resilience and anxiety, depression could provide precise targets for nurse mangers to develop more effective intervention to prevent and reduce anxiety and depression among clinical nurses, ultimately promoting nurses’ psychological well-being and improving patient care quality.

No previous research has explored how resilience components connect with symptoms of anxiety and depression. This study intended to fill the existing gap and build upon prior research regarding resilience in the context of anxiety and depression by modelling the correlations between resilience components and the symptoms of anxiety and depression using a NA approach. Thus, our study aimed to analyse the unique connections between resilience and anxiety as well as between resilience and depression, and to find the essential nodes that connect the resilience cluster and anxiety and depression cluster.

Methods

Study design and participants

A multi-site, cross-sectional design using an anonymous online survey was conducted. Clinical nurses were recruited by convenience sampling from five tertiary hospitals in Baoding City, Hebei Province, China, from November 2023 to January 2024. The criteria for inclusion were: (a) individuals who are over 18 years of age and of Chinese nationality; (b) enrolled or registered, employed full-time; Nurses with less than one year of work experience and nursing students were excluded. R powerly package was used to evaluate sample size. The lower and upper bounds of the candidate sample size range were set to 300 and 1000, sensitivity was set to 0.6, and power was set to 0.8. According to the recommendation of R powerly package, the sample size for both networks was 300 [35]. A total of 1600 online questionnaires were obtained. 310 questionnaires were deleted because of too short response times (completion times of less than 2 sec per item) [36] and 11 questionnaires were excluded due to work experience less than one year. Finally, 1279 valid participants were included in this study (an effective rate of 79.9%).

Data collection

This online survey was administrated utilizing the Wenjuanxing platform, a well-known website in China (https://www.wjx.cn/). The initial page of the questionnaire delineated the study’s introduction, objectives, and informed consent. The questionnaire was distributed via the social network WeChat. Participants initially consented to partake, subsequently completing all questions and were permitted to submit their responses once to prevent duplicate submissions. The questionnaire required approximately 15 to 20 min for completion.

Measures

Demographic information

A demographic information form was designed to achieve the aim, comprising questions related to demographic and job-related characteristics, such as, age, gender, marital status, education level, working years, and job title.

Resilience

The 10-item Connor-Davidson Resilience Scale (CD-RISC-10) was utilized to assess resilience. CD-RISC-10 is developed based on choosing 10 items from the original 25-item version, which captures resilience’s core characteristics [37]. The CD-RISC-25 has been extensively employed to assess the resilience level of nurses, initially created by Connor and Davidson [38], and then adapted to Chinese by Yu and Zhang [39]. The CD-RISC-10 version assesses the capacity to endure situations including change, personal difficulties, stress. Responses to the items vary from 0 (“not true at all”) to 4 (“true nearly all the time”). The overall score runs from 0 to 40, with elevated scores implying more resilience capacity. The validity and reliability of the CD-RISC-10 among Chinese population have been thoroughly documented [40, 41]. The Cronbach’s α coefficient for CD-RISC-10 in this research was 0.957.

Anxiety

The Chinese Version of Seven-item Generalized Anxiety Disorder (GAD-7) was applied to measure anxiety symptoms [42]. Nurses were requested to document how often of symptoms they encountered in the preceding two weeks. A Likert scale was utilized to evaluate the items ranging from 0 (not at all) to 3 (nearly every day). Higher GAD-7 scores reflect greater severity of anxiety symptoms. A score higher than 5 is considered indicative of anxiety symptoms [43]. The GAD-7 scale is extensively used in China and has strong validity and reliability [44]. This study observed a Cronbach’s α coefficient of 0.954.

Depression

Depression was evaluated employing the Chinese Version of nine-item Patient Health Questionnaire (PHQ-9) [45]. Nurses were requested to document how often of symptoms they encountered in the preceding two weeks. Item frequency ranges from 0 (not at all) to 3 (nearly every day). Higher PHQ-9 scores suggest greater severity of depression symptoms. A score higher than 5 is considered indicative of depression symptoms [45]. This scale acquired great validity and reliability and has been widely utilized within the Chinese population. The PHQ-9 had strong internal consistency in this research, with a Cronbach’s coefficient of 0.945.

Data analysis

First, the data were analysed in the SPSS 26.0 package program. We conducted a descriptive analysis to analyse the characteristics of the participants (continuous variables: means and standard deviations (SD); categorical variables: frequencies and percentages).

Then, the R package qgraph was employed to construct two network models based on Spearman rho correlation, namely Resilience-anxiety and Resilience-depression network. The Gaussian Graphical Model was tested to ascertain these undirected networks’ network structure (42). Edges signify the partial connections among two nodes after statistically adjusting for all other nodes in the network. We obtained a sparse network that reflects the true structure of the network by using the least absolute shrinkage and selection operator regularization algorithm [46, 47]. The parameter value for the Extended Bayesian Information Criterion was established at 0.5 to equilibrate sensitivity and specificity [48]. The Fruchterman-Reingold algorithm was employed to lay out both networks established [49]. Positive correlations within each network were drawn by blue edges, while negative correlations were depicted by red edges.

This study included two predetermined communities: RISC (resilience) and symptom communities (anxiety and depression symptoms). To assess the relative significance of specific nodes in elucidating cross-community co-occurrence, we computed the bridge expected influence centrality (BEI), which is the aggregate of the edges linking a certain node to all nodes in the opposing community [50]. BEI is considered a more appropriate centrality to determine bridge nodes in a network characterized by both positive and negative relationships [51]. A higher positive BEI value indicates an increased capacity for activation of other communities, whereas a lower negative value signifies a larger capacity for deactivation [51]. The above-described procedures were executed using the R package networktools.

Bootstrapping approaches were adopted through the R package bootnet to guarantee the networks’ accuracy and stability. We applied non-parametric bootstrapping (with 2,000 samples) to estimate the 95% confidence interval for all edges within the network to verify the accuracy of edge weights [52]. To assure stability of BEI centrality, the correlation stability (CS) coefficient was obtained by a case-dropping subset bootstrap method (with 2,000 bootstrap samples) [52]. It is recommended that the CS-coefficient be preferably above 0.5 and should not fall below 0.25 [52]. Bootstrap difference tests (2,000 samples) were conducted to determine if there are significant differences between edge weights or between node BEIs. Bootstrap difference testing (2,000 samples) was run to find out if significant differences exist between edge weights or node BEIs [52].

Results

Descriptive statistics

The sample comprised 1279 clinical nurses. The sample averaged 34.1 years of age (SD = 6.4), and the average duration of working was 11.6 years (SD = 6.8). Table 1 provides the demographic and job-related features for the sample. The current study found 67.3% of clinical nurses experienced anxiety and 67.1% experienced depression. Network node means and SD are shown in Table 2.

Table 1.

Demographic characteristics of the participants (N = 1279)

	n	%
Age
18 years∼	321	25.1
30 years∼	729	57.0
40 years∼	192	15.0
50 years∼	37	2.9
Gender
Male	86	6.7
Female	1193	93.3
Education level
Junior college or less	80	6.3
Undergraduate or more	1199	93.7
Marriage
Married	1008	78.8
Single (Unmarried/Divorced/ Widowed)	271	21.2
Working years
<5	173	13.5
5∼	359	28.1
≥10	747	58.4
Night work
Yes	818	64.0
No	461	36.0
Professional title
Junior	501	39.2
Middle	629	49.2
Senior	149	11.6
Monthly income (CYN)
<5000	111	8.7
5000∼	291	22.8
7000∼	553	43.2
≥9000	324	25.3

Table 2.

Abbreviation, mean scores, and standard deviations for items of resilience, anxiety and depression (N = 1279)

Number	Items	Abbreviation	Mean	SD
RISC1	Able to adapt to change	Adapt to change	2.79	0.81
RISC2	Can deal with whatever comes	Deal with whatever comes	2.54	0.85
RISC3	Tries to see humorous side of problems	See humorous side	2.71	0.86
RISC4	Coping with stress can strengthen me	Stress strengthens me	2.72	0.87
RISC5	Tend to bounce back after illness or hardship	Bounce back	2.58	0.90
RISC6	Can achieve goals despite obstacles	Achieve goals	2.77	0.84
RISC7	Can stay focused under pressure	Stay focused under pressure	2.50	0.88
RISC8	Not easily discouraged by failure	Not discourage by failure	2.66	0.85
RISC9	Thinks of self as strong person	Strong person	2.74	0.85
RISC10	Can handle unpleasant feelings	Handle unpleasant feelings	2.61	0.85
GAD1	Feeling nervous, anxious, or on edge	Nervousness or anxiety	1.17	0.80
GAD2	Not being able to stop or control worrying	Uncontrollable worry	1.07	0.83
GAD3	Worrying too much about different things	Excessive worry	1.10	0.85
GAD4	Trouble relaxing	Trouble relaxing	1.07	0.88
GAD5	Being so restless that it is hard to sit still	Restlessness	0.73	0.80
GAD6	Becoming easily annoyed or irritable	Irritability	1.07	0.83
GAD7	Feeling afraid as if something awful might happen	Fear that something might happen	0.82	0.84
PHQ1	Little interest or pleasure in doing things	Anhedonia	1.03	0.82
PHQ2	Feeling down, depressed, or hopeless	Depressed or sad mood	0.94	0.80
PHQ3	Trouble falling/staying asleep or sleeping too much	Sleep difficulties	1.06	0.92
PHQ4	Feeling tired or having little energy	Fatigue	1.23	0.87
PHQ5	Poor appetite or overeating	Appetite changes	0.97	0.88
PHQ6	Feeling bad about yourself—or that you are a failure or have let yourself or your family down	Feelings of worthlessness	0.80	0.83
PHQ7	Trouble concentrating on things, such as reading the newspaper or watching television	Concentration difficulties	0.79	0.84
PHQ8	Moving or speaking so slowly that other people could have noticed, or the opposite— moving around a lot more than usual, being fidgety or restless	Psychomotor agitation/retardation	0.72	0.83
PHQ9	Thoughts that you would be better off dead or hurting yourself in some way	Suicidal ideation	0.52	0.76

The resilience-anxiety network

Figure 1(A) shows clinical nurses’ resilience-anxiety network model. The resilience-anxiety network comprised 79 edges, with 15 of them were bridging different communities. Four strongest bridge edges were RISC1 “Adapt to change”-GAD7 “Fear that something might happen” (edge weight = -0.05), RISC7 “Stay focused under pressure”-GAD2 “Uncontrollable worry” (edge weight = -0.03), RISC10 “Handle unpleasant feelings”-GAD7 “Fear that something might happen” (edge weight = -0.03), and RISC1 “Adapt to change”-GAD5 “Restlessness” (edge weight = -0.03). Table S1 (in additional file 1) presents all resilience-anxiety network edge weights. The bootstrapped 95% CI is relatively narrow, suggesting acceptable edge weight accuracy (Fig. S1 in additional file 1). Fig. S2 (in additional file 1) exhibits edge weight bootstrapped difference test results.

Fig. 1.

Network structure of resilience-anxiety and centrality index. (A) Network construction of different components of resilience and anxiety in clinical nurses; (B) Bridge expected influence centrality index of the network. Note: Blue edges represent positive correlations, and red edges represent negative correlations. The thickness of the edge reflects the magnitude of the correlation

The BEI for each resilience-anxiety network node is depicted in Fig. 1B. RISC1 “Adapt to change” (BEI = -0.08) and GAD-7 “Fear that something might happen” (BEI = -0.08) had the greatest BEIs of their communities and were recognized as resilience-anxiety bridges. Node BEI had a CS-coefficient of 0.44, indicating a stable network (Fig. S3 in additional file 1). Fig. S4 (in additional file 1) displayed the results of the bootstrapped difference test of BEI.

The resilience-depression network

Figure 2(A) displays the resilience-depression network model for clinical nurses. The resilience-anxiety network consists of 98 edges, with 23 of them were bridging different communities. Two strongest bridge edges were RISC1 “Adapt to change”-PHQ7 “Concentration difficulties” (edge weight = -0.05) and RISC7 “Stay focused under pressure”-PHQ4 “Fatigue” (edge weight = -0.04). Table S2 (in additional file 1) describes the specifics of the edge weights inside of the network. A relative narrow bootstrapped 95% CI revealed acceptable edge weight accuracy (Fig. S5 in additional file 1). The bootstrapped difference test of edge weights is illustrated in Fig. S6 (in additional file 1).

Fig. 2.

Network structure of resilience-depression and centrality index. (A) Network construction of different components of resilience and depression in clinical nurses; (B) Bridge expected influence centrality index of the network. Note: Blue edges represent positive correlations, and red edges represent negative correlations. The thickness of the edge reflects the magnitude of the correlation

Figure 2B depicts the BEI for all node in the resilience-depression network. RISC1 “Adapt to change” (BEI = -0.08) and PHQ1 “Anhedonia” (BEI = -0.08) had the greatest BEIs of their communities and were recognized as resilience-depression bridges. The CS-coefficient of node BEI was 0.44, suggested adequately stability (Fig. S7 in additional file 1). Fig. S8 (in additional file 1) presents the results of the bootstrapped difference test of BEI.

Discussion

The incidences of anxiety and depression were 67.3% and 67.1% among clinical nurses in this study, respectively. The reported prevalence rates of anxiety and depression in the current study were high, aligning with previous studies conducted among Chinese clinical nurses [53], suggesting serious mental health problems among nurses. This study firstly applied NA to investigate the interrelationship between resilience and anxiety, as well as between resilience and depression among clinical nurses. Our findings provide new theoretical viewpoints that not only comprehend the understanding of the links between resilience and anxiety/depression but also have direct implications for developing more targeted interventions for clinical nurses.

In the resilience-anxiety network, “Adapt to change” was found to be negatively related to “Fear that something might happen”. Due to the complexity of the work environment and practices in the post-COVID-19 epidemic era, clinical nurses who are able to adapt to change may adopt a positive coping approach in the face of uncertainty instead of worry and fear. Consequently, they are less prone to fearing that something awful might happen [54, 55]. “Stay focused under pressure” was negatively associated with “Uncontrollable worry”, stress and coping theory suggests individuals that adopt different coping strategies when dealing with stress [56]. Clinical nurses who focus on the tasks or problems faced with stress, they are more likely to positively and effectively cope with stress, which means they are less susceptible to ruminate or worry [57]. “Handle unpleasant feelings” was negatively connected with “Fear that something might happen”. Previous studies have reported that emotion regulation is strongly linked to negative mental health [58, 59]. Individuals with a high level of emotion regulation are often better at addressing obstacles as well as accepting reality, and are therefore less susceptible to fear. “Adapt to change” was negatively connected with “Restlessness”. Individuals who are more adaptable tend to be emotionally stable and better able to maintain calm, while those who are less adaptable may be more prone to such anxiety symptoms [55].

In the resilience-depression network, “Adapt to change” was negatively correlated with “Concentration difficulties”. Adaptable individuals tend to exhibit greater flexibility in navigating new situations and challenges, allowing for smoother transitions in directing their attention towards new tasks [60]. In contrast, individuals struggling with adaptation may experience more anxiety or restlessness, potentially hindering their ability to concentrate. “Stay focused under pressure” was negatively related to “Feeling tired or having little energy”. Individuals are able to maintain focuse under stress, indicating the ability to handle stress effectively and solve problems efficiently, thus reducing the likelihood of experiencing fatigue or low energy. Additionally, people who feel tired or lack energy often experience reduced motivation and difficulty in concentrating [61].

This study’s networks illustrate that bridge nodes offer a unique insight into the co-occurrence of resilience with anxiety and depression, highlighting the distinct roles of various resilience components in the emergence and persistence of these conditions. Targeting the bridge component could culminate in the inactivation of the propagation pathway and a decrease in co-occurrence [51]. Interestingly, our study found the BEI indices of RISC1 “Adapt to change” was the highest both in the resilience-anxiety network and resilience-depression network, identifying it as bridge node. This also means that RISC1 “adapt to change” is the most influential component of resilience in connection with anxiety and depression. Our finding is similar to prior research, which has demonstrated that “Adapt to change” was significantly correlated with negative mental health disorders, such as anxiety and depression” [62, 63]. Additionally, Sharpley’s study identified “Adapt to change” as a bridging component between the network of resilience and depression, consistent with our results [63]. Nurses often encounter stressful and unsatisfactory circumstances in clinical workforce due to substantial workloads and patients with intricate care requirements. The ability to adapt to change is crucial for clinical nurses dealing with the complexities of nursing practice. Therefore, intervention focusing on the capacity to “adapt to change” could prevent and reduce anxiety and depression among clinical nurses.

The present study offers several significant implications for addressing mental disorders among clinical nurses. First, the prevalence of anxiety (67.3%) and depression (67.1%) symptoms remains alarmingly high among clinical nurses. This underscores the urgent need for timely screening and targeted mental health interventions. Implementing such measures is crucial to maintain clinical nurses’ well-being and preserve the quality of care. Second, our study not only offers a fine-grained comprehension of the connections underlying resilience with anxiety/depression, but also lays the groundwork for future psychological intervention aimed at fostering resilience among clinical nurses. Third, “adapt to change” was identified as the most influential component of resilience linked to both anxiety and depression. It is recommended to develop intervention focused on promoting the capacity to “adapt to change” among clinical nurses and then evaluate the effectiveness of these interventions in real-world clinical settings. Intervention strategies should focus on developing clinical nurses’ skills in flexibility, emotional regulation, and problem-solving under changing workforce. Tailored measures, including cognitive behaviour therapy (CBT) and mindfulness, can be effective in fostering adaptability to change. For example, CBT-based programs could be designed to equip clinical nurses with techniques to manage stressors at work, reframe negative thoughts, and maintain emotional stability during change. Moreover, mindfulness-based stress reduction programs could be implemented to teach nurses how to stay present and focused, mitigating the emotional impact of unexpected change and enhancing adaptability. Additionally, from a nurse manager’ s perspective, promoting transformational leadership is essential. By fostering an open and communicative work environment, nurse managers can help clinical nurses feel supported to confront challenges and seek help, thereby enhancing their adaptability to change.

Limitations

Despite these implications, this research also involves some limitations. First, the cross-sectional design employed in this study limited the ability to determine the causal connections between resilience and anxiety, depression. Longitudinal investigations of these variables using NA are needed. Second, the sample was exclusively from Baoding city in northern China, and thus generalizing the findings to more developed cities remains to be considered. Finally, our study did not empirically confirm the proposed intervention targets, necessitating further research to validate them before they can be applied in the real world.

Conclusion

This study firstly applied NA to explore the correlations between resilience and anxiety, as well as between resilience and depression in clinical nurses. The visual network structure offered a nuanced depiction of the connection pathways linking resilience to anxiety and resilience to depression. The BEI comparison facilitated the identification of connections among these linkages and suggested “adapt to change” of the resilience as a potential target for interventions aimed at reducing anxiety and depression. This study offers a dependable reference for the implementation of psychological treatments.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

COVID-19

Coronavirus Disease 2019

NA

Network Analysis

CD-RISC

The Connor-Davidson Resilience Scale

GAD-7

Generalized Anxiety Disorder 7-Item

PHQ-9

Patient Health Questionnaire 9-Item

SD

Standard Deviations

BEI

Bridge Expected Influence

CS

Correlation Stability

CBT

Cognitive Behaviour Therapy

NIA

Network Intervention Analysis

Author contributions

Y.Z., W.G., and H.L. originally designed the study concept and idea. W.G., H.L., X.Y., and J.W. collected data. Y.Z. and X.Z. conducted the statistical analysis for the study. Y.Z. wrote the initial draft. X.Z. prepared the tables. Y.Z., X.Y., and J.W. prepared the figures. W.G., X.Y., and J.W. contributed to the amendment of the draft and suggestions for data analysis. All authors reviewed and approved the final manuscript.

Funding

This work was supported by the Xingtai Science and Technology Bureau (Key Research and Development Plan Project, No.2023ZC123).

Data availability

Data in the current study is available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was executed in compliance with the Declaration of Helsinki and received approval from the Medical Institutional Review Board of Baoding No.1 Central Hospital (approval number: [2023]127). Informed consent to participate in this study was obtained from all participants.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Clinical trial number

Not applicable.