Article Text
PDF
Abstract
Objectives The objectives are to explore the relationship between study stress and anxiety in high school students and the mediating role of physical activity and mobile phone addiction.
Design A cross-sectional study.
Setting 129 high schools were randomly selected in 13 cities of Jiangsu province, China.
Participants High school students aged 16–19 years, age and gender balance. A total of 40 000 questionnaires were distributed, with 32 974 effectively recovered.
Primary and secondary outcome measures Questionnaires were administered offline, covering four parts: General Demographics, Learning Stress Scale, International Physical Activity Questionnaire, Mobile Phone Addiction Scale and Generalized Anxiety Scale-7. Data analysis included path analysis and correlation analysis, along with descriptive statistics, independent sample t-test, correlation analysis and structural equation model.
Results In this study, the proportions of anxiety, high academic pressure, low physical activity level and high mobile phone addiction were 58.18%, 46.48%, 36.40% and 39.26%, respectively. Study stress was positively correlated with anxiety (r=0.130, p<0.01) and mobile phone addiction (r=0.049, p<0.01). Physical activity was negatively correlated with learning stress (r=−0.352, p<0.01), anxiety (r=−0.105, p<0.01) and mobile phone addiction (r=−0.040, p<0.01). The findings were tested by mediating effect analysis that the indirect effect size value of the path ‘learning stress → physical activity level → anxiety path’ was 0.461, 95% CI of Bootstrap (0.367, 0.554), the mediating effect was significant. The indirect effect size value of the path ‘learning stress → mobile phone addiction → anxiety’ was 0.072, 95% CI of Bootstrap (0.042, 0.102), and the mediating effect was significant. The indirect effect size value of the path ‘learning stress → physical activity level → mobile phone addiction → anxiety’ was 0.072, and the 95% CI of Bootstrap (0.226, 0.400), and the mediating effect was significant.
Conclusions High school students’ learning stress can significantly positively predict anxiety levels. High school students learning stress indirectly predicts anxiety through the independent mediating effect of physical activity and mobile phone addiction, as well as the chain mediating effect of physical activity and mobile phone addiction.
- Adolescent
- Cell Phone
- China
- MENTAL HEALTH
Data availability statement
Data sharing not applicable as no datasets generated and/or analysed for this study.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
Through the construction of a chain mediation model, this study explores the relationship between learning stress and anxiety in high school students and its mechanism, as well as the mediating role played by physical activity and mobile phone addiction. At the same time, the influence of sociodemographic, socioeconomic and other related factors is considered.
A convenient sampling method was used to randomly select 129 high school students aged 16–19 (senior one, senior two and senior three) from 13 cities in Jiangsu province to investigate 36 573 questionnaires.
Data are from a gender and age-balanced sample.
Regression analysis and Bootstrap sampling method were used to analyse the data and verify the degree of fit of the intermediary model.
Self-reported outcome variables may have inherent bias and confounding factors.
Introduction
The high prevalence of anxiety among high school students and the frequent occurrence of self-injury and suicide events caused by neurological diseases such as depression have been the focus of national and social attention. In response to the thorny problem of high school students’ anxiety, the ‘Health China 2030’ Planning Outline and Health China Action—Action Program for Children and Youth Mental Health (2019–2022) clearly pointed out that the intervention of mental health problems such as high school students’ anxiety should be strengthened. Moreover, research on the causes of anxiety occurrence and improvement pathways is currently a hot topic.1 2 Learning stress is one of the important reasons for the occurrence of anxiety among high school students, and the study of the relationship and mechanism between them has received attention from the fields of psychology and medicine.3 The purpose of this study is to investigate the mechanism of the internal connection between learning stress and anxiety among high school students and to provide a theoretical and practical basis to improve anxiety through learning stress in practice.
The relationship between learning stress and anxiety
Psychoanalytic and behaviourist viewpoints suggest that the occurrence of anxiety is closely related to the external specific environment that is more stimulating to the individual and the feeling of not being able to cope or control the specific events or things. In response to the academic stress of adolescent students, the ‘educational stressor hypothesis’ holds that the expectation of school performance and the influence of relevant exams make the life prospects of adolescents more dependent on educational performance, thus generating more learning pressure and leading to higher levels of anxiety.4 Learning stress, as a key external specific environment, is an important cause of anxiety.5 6 It was found that learning stress was significantly and positively correlated with anxiety among junior high school students (r=0.38, p<0.05) and college students (r=0.54, p<0.001).7 8 This was closely related to the cognitive environment in which excessive learning stress stimulated cognitive evaluation abilities and personality tendencies.9 The general conceptual model in the aetiology of adolescent psychopathology states that specific life stress (such as learning stress) is associated with specific psychological problems (such as anxiety) through specific mediating processes (such as insecure attachment styles, avoidance, ruminative thinking, neurotic personality, etc).10 However, some studies suggest that appropriate levels of learning stress and anxiety are not entirely negative and that learning stress can be reduced by improving self-efficacy, reassessing anxiety perception, and thereby optimising the experience of the emotional value of stress.11 At present, only a few studies have been conducted with high school students in China as the subjects.12 13 A survey of 503 high school students indicated a significant positive correlation between learning stress and anxiety.14 Based on the shortcomings of the current study, such as the small survey scope and small number of participants, this study intends to investigate a large sample of high school students from 13 cities of Jiangsu province to reveal the relationship between learning stress and anxiety. Hypothesis H1 was proposed: learning stress in high school students positively predicts anxiety.
Mediating role of physical activity
Based on the environment–human matching theory and the cognitive–behavioural cognitive model used by the pathological internet, physical activity is an important means to improve anxiety.15 Due to the high degree of cooperation and high demand for sports among college students, there are many studies on physical activity to improve anxiety taking college students as the research subjects,16 17 while there are few studies taking high school students as the subjects. It was found that physical activity levels were higher among high school males than females, and senior two students had the highest levels of physical activity, while senior three students had the lowest levels.18 A survey found that 51% of senior three had moderate levels of physical activity, 29% had low levels of physical activity and 24% had moderate levels of anxiety, while 12% had severe anxiety.19 The above study only analysed the differences in physical activity levels and anxiety detection rates among high school students of different grades and investigated the relationship between physical activity levels and anxiety. There was a close relationship between learning stress and physical activity, and high learning stress led to an increase in sedentary and static behaviour among high school students.20 21 In addition, some studies pointed out that learning stress was a key factor causing a decrease in physical activity levels,22 but there was a lack of research evidence. Given the above, this study proposed hypothesis H2: physical activity has a mediating effect between learning stress and anxiety in high school students.
The mediating role of mobile phone addiction
Mobile phone addiction refers to a kind of excessive, high-frequency and uncontrollable psychological or behavioural state in which an individual is dependent, impulsive and compulsive to use mobile phones.23 High school students’ high learning stress led to negative behaviours, such as learning aversion and mobile phone addiction.24 The compensatory network usage theory believes that high school students turn to smartphones to circumvent their inner troubles when encountering psychological and social problems.25 In addition, the self-regulation defect model proposes that high school students with psychosocial problems weaken self-regulation and control abilities, leading to uncontrolled use of mobile phones and triggering mobile phone addiction.26 Learning stress is an important psychological and social issue for high school students. Research revealed a significant positive correlation between learning stress and mobile phone addiction among high school students; namely, the greater the learning stress was, the more severe the mobile phone addiction would be.27 And mobile phone addiction would increase the risk of anxiety among teenagers. A survey found a high prevalence of mobile phone addiction among high school students (more than 21% incidence), and severe mobile phone addiction led to reduced interpersonal communication and increased negative emotions like anxiety from the above-mentioned, learning stress-induced mobile phone addiction, which in turn led to the occurrence of anxiety.28 Nevertheless, the research on high school students in the field of sports remains to be revealed. Therefore, hypothesis H3 was put forward: mobile phone addiction has a mediating effect between learning stress and anxiety in high school students.
The chain mediating effect of physical activity and mobile phone addiction
For the two mediating variables in this study, it was found that physical activity negatively predicted mobile phone addiction.29 The social support theory suggests that the execution and experience of physical activities can stimulate adolescents’ sense of social identity and self-identity, making their social behaviour more dynamic and enthusiastic.30 The higher the level of physical activity as an external social environment stimulus was, the more frequent and significantly more self-controlled interpersonal interactions of the individual would be, as well as the significant reduction of static screen behaviour and mobile phone usage.31 32 Therefore, the two variables were in an antagonistic state, and the decreased level of physical activity caused by learning stress led to increased sedentary time and phone dependence, which, therefore, induced mobile phone addiction and anxiety. Therefore, as shown in figure 1, this study proposed hypothesis H4: physical activity and mobile phone addiction play a chain mediating role between learning stress and anxiety.
This chain mediation model diagram in this study.
Methods
Subject
129 senior high schools were randomly selected in 13 cities of Jiangsu province, using a convenient sampling method. A total of 40 000 questionnaires were sent out, and 36 573 questionnaires were collected, with a recovery rate of 91.4%. The effective questionnaires were 32 974, with an effective rate of 90.2%. The age range is 16–19 years. All the senior high school students were enrolled in classes in accordance with the school schedule in the spring semester and were enrolled full time. Their financial sources were provided by their parents for living expenses. The demographic distribution is shown in online supplemental appendix 1. According to the frequency analysis results of each variable, it can be seen that the numerical characteristics of demographic variables basically meet the requirements of the sampling survey.
Supplemental material
Tools
Learning Stress Scale
The Learning Stress Scale, which was developed in 2010, is a tool to measure the degree of students’ learning stress.33 We conducted a survey on 481 middle school students in Xiangtan city, Hunan province, China, and verified its reliability and validity in Chinese middle school students.34 The scale consists of 21 items and 4 dimensions (parental pressure, self-pressure, teacher pressure and social pressure).33 The higher the total score is, the greater the learning pressure is. A score below 0.5 was considered stress free, 0.5–0.59 mild, 0.6–0.69 moderate, and greater than 0.7 high. Cronbach’s α coefficient is between 0.70 and 0.87, and the omega index is 0.81, indicating high scale reliability. Cronbach’s alpha index was derived from the initial questionnaire.
International Physical Activity Questionnaire (IPAQ)
The IPAQ, developed in 2001 by the International Consensus Group on Physical Activity Measurement, is a tool for measuring people’s levels of physical activity in their daily lives. Qu and Li tested the reliability of the questionnaire by systematically sampling 97 college students and repeated surveys 3 days apart to verify the reliability and validity of the questionnaire.35 The scale has seven questions, six of which ask about individuals’ physical activity. The structure of the questions is the same as that of the long volume, and only the activity intensity is reserved. By inquiring about the weekly frequency and daily cumulative time of different intensity activities, the measured energy consumption of physical activity was low, which was more suitable for measuring adolescents’ physical activity levels. Internal consistency reliability r=0.927. The questionnaire divided physical activity levels into high, medium, and low levels. The specific calculation method is shown in online supplemental appendix 2.
Mobile Phone Addiction Scale
The Mobile Phone Addiction Scale, which was compiled by the Chinese in 2013, is an effective tool for studying adolescents’ mobile phone dependence behaviour.36 A stratified random sampling of 2522 freshmen and junior students from a medical university in Hefei, Anhui Province, China, was conducted to verify the reliability and validity of the questionnaire. The scale consists of 13 items, with a 5-level score and 3 dimensions: withdrawal, craving and psychosomatic influence.37 There were six withdrawal symptoms, three cravings and four psychosomatic effects. The higher the score, the higher the degree of phone dependence. A score below 15 is considered as no or low addiction, 15–26 as moderate addiction and above 26 as high addiction. Cronbach’s α coefficient of the scale was 0.891, indicating high reliability of the scale. Cronbach’s alpha index was derived from the initial questionnaire.
Generalized Anxiety Scale-7 (GAD-7)
Adopted by Spitzer RL et al used the Abbreviated Anxiety Symptom Self-Rating Scale, developed according to the Diagnostic criteria of the American Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, is a questionnaire used to assess whether an individual has generalised anxiety symptoms.38 Wang et al by sampling 569 inpatients from West China Hospital of Sichuan University in China for GAD-7 evaluation, the questionnaire was verified to have good reliability and validity.39 The scale measured symptoms within 2 weeks, consisting of 7 items, with a 4-level score. On a scale of 4, 5–9 are classified as mild anxiety, 10–14 as moderate anxiety and 15–21 as severe anxiety.40 Cronbach’s α coefficient was 0.883, indicating high reliability of the scale. Cronbach’s alpha index was derived from the initial questionnaire.
Statistical analysis
Firstly, Excel 2019 was used to collect the raw data, and SPSS V.26.0 was used for descriptive statistics, independent sample t-test, χ2 test, one-way analysis of variance, Person correlation analysis, confirmatory factor analysis, etc. Then Bootstrap method was used to establish and modify the structural equation model. Meanwhile, the SPSS plug-in Process was used to test the intermediary model’s effect size and standardised path coefficient. To achieve better model correction effect.
Common method deviation test
In this study, learning stress, physical activity, mobile phone addiction and anxiety of high school students were measured by questionnaires, which were filled out by the subjects themselves, which were subjective to a certain extent, and the results would be affected by the common method bias. Therefore, the Harman single-factor test was adopted to conduct a common method deviation test on sample data, and exploratory factor analysis was conducted on all questions of the four variables.41 It was found that there were 10 factors with characteristic roots greater than 1, and the total variance interpretation rate of the first factor was 24.78%, less than the critical value of 40%. Therefore, there is no serious common methodology bias in this study.
Results
Descriptive statistics of each variable
As shown in online supplemental appendix 3, the overall score of learning stress was 0.68±0.17, which is above the medium level. None and low learning stress accounted for 14.6% and 11.5% respectively, moderate stress for 27.4%, and high stress for 46.5%. The anxiety score was 5.78±4.47, which was in the lower medium level. 42% had no anxiety, 41% had mild anxiety, 13% had moderate anxiety and 4% had severe anxiety. The score of physical activity level was 1.87±0.76, which was in the lower middle level. 36.4% had a low physical activity level, 40.4% had a moderate physical activity level and 23.2% had a high physical activity level. The overall score of mobile phone addiction was 25.17±9.66, which was above the medium level. The high addiction rate was 39%, the moderate addiction rate was 47% and the no or low addiction rate was 14%.
Differences in demographic variables of each variable
Analysis of gender differences in each variable
Through the analysis of online supplemental appendix 4, it can be shown that there are significant differences in the total level of learning stress in gender (p<0.01), and the learning stress of females is significantly higher than that of males (p<0.01). There was a significant difference in anxiety in gender, and females were higher than males (p<0.01). The physical activity level of males was higher than that of females (p<0.01). A significant difference existed in mobile phone addiction between males and females, and males were higher than females (p<0.01).
Difference analysis of variables in grades
As shown in online supplemental appendix 5, it can be seen that there are significant differences in learning pressure in grades (p<0.01). Multiple comparisons show that the learning pressure in senior two is significantly higher than that in senior three and senior three is higher than that in senior one (p<0.01). There was a significant difference in anxiety in grade one and it was significantly less than that in grade two (p<0.05 or p<0.01). There were significant differences in physical activity among grades (p<0.01). Multiple comparisons showed that the physical activity of senior one was significantly higher than that of senior two and senior three (p<0.01), but there were no significant differences between senior two and senior three (p>0.05). There were significant differences in the grades of mobile phone addiction (p<0.01). Multiple comparisons showed that the differences were significant (p<0.01).
Analysis of differences between urban and rural variables
Through the analysis of online supplemental appendix 6, it shows that there was no significant difference in learning stress between urban and rural students (p>0.05). The difference in anxiety in urban and rural areas was significant and the anxiety in cities was higher than that in towns (p<0.01). There was no significant difference in physical activity levels between urban and rural students (p>0.05). There was a significant difference in mobile phone addiction between urban and rural students (p<0.01).
Difference analysis of variables on an only child or non-only child
According to online supplemental appendix 7, there was no significant difference in learning stress between the only child and non-only child (p>0.05). There was a significant difference in the anxiety of the only child, and the non-only child was significantly higher than the only child (p<0.01). The level of physical activity of the only child was significantly higher than that of the non-only child (p<0.01). There was a significant difference in whether the child was an only child or not, that is, the non-only child was higher than the only child (p<0.01).
Difference analysis of each variable in family structure
From the analysis of online supplemental appendix 8, it can be seen that learning stress had no significant difference in family structure (p>0.05). There were significant differences in anxiety in family structure (p<0.01). Multiple comparisons showed that the level of anxiety in single-parent and reconstituted families was significantly higher than that in both-parent families (p<0.01), but there was no significant difference between single-parent and reconstituted families (p>0.05). The level of physical activity was significantly different in family structure; that is, both-parent families were significantly lower than single-parent families (p<0.05). There was a significant difference in the family structure of mobile phone addiction (p<0.01); that is, both-parent families were significantly lower than single-parent and reconstituted families (p<0.01), but there was no significant difference between single-parent and reconstituted families (p>0.05).
Correlation analysis of each variable
According to table 1, learning stress, physical activity, mobile phone addiction and anxiety are significantly correlated. Study stress was positively correlated with anxiety (r=0.130, p<0.01) and mobile phone addiction (r=0.049, p<0.01). Physical activity was negatively correlated with learning stress (r=−0.352, p<0.01), anxiety (r=−0.105, p<0.01) and mobile phone addiction (r=−0.040, p<0.01).
Correlation analysis among variables
Analysis of chain mediating effect of learning stress and anxiety of senior high school students
Model 6 in SPSS plug-in Process V.4.0 was used to conduct mediating effect analysis, with gender, grade, urban and rural areas, whether or not the only child and family structure as control variables, learning stress as independent variable, anxiety as dependent variable, physical activity and mobile phone addiction as mediating variables. The Bootstrap sampling method was used to test the mediating effect. From table 2, it could be seen that learning stress significantly negatively predicted physical activity (β=−1.525, p<0.01), significantly positively predicted mobile phone addiction (β=2.396, p<0.01), significantly positively predicted anxiety (β=2.409, p<0.01). Physical activity significantly negatively predicted mobile phone addiction (β=−0.363, p<0.01) and significantly negatively predicted anxiety (β=−0.302, p<0.01). Mobile phone addiction significantly positively predicted anxiety (β=0.130, p<0.01).
Regression analysis of each variable in the mediation variable model
It can be seen from table 3 that ind1: the indirect effect size value of the path ‘learning stress → physical activity level → anxiety path’ was 0.461, 95% CI of Bootstrap (0.367, 0.554) did not contain 0, the mediating effect was significant, accounting for 14% of the total effect and physical activity level played a partial mediating role between learning stress and anxiety. Ind2: the indirect effect size value of the path ‘learning stress → mobile phone addiction → anxiety’ was 0.072, 95% CI of Bootstrap (0.042, 0.102) did not contain 0, and the mediating effect was significant, accounting for 2% of the total effect. Mobile phone addiction played a partial mediation role between learning stress and anxiety. Ind3: the indirect effect size value of the path ‘learning stress → physical activity level → mobile phone addiction → anxiety’ was 0.072, and the 95% CI of Bootstrap (0.226, 0.400) did not contain 0, and the mediating effect was significant, accounting for 10% of the total effect. As shown in figure 2, the level of physical activity and mobile phone addiction played a chain mediating role between learning stress and anxiety.
The chain mediation effect model diagram of this study.**P<0.01.
Analysis of chain mediating effect between physical activity level and mobile phone addiction
Discussion and analysis
This study found that learning stress, physical activity and mobile phone addiction all significantly predicted anxiety. High school students faced various pressures, such as the college entrance examinations, and these negative life events were important sources of stress. Therefore, increasing the physical activity levels of high school students in high-learning-stress environments and improving their mobile phone addiction will be beneficial to make them more proactive in facing learning stress and reducing their anxiety levels.
The Stress Buffering Model of Physical Activity believes that as an important external physiological-psychological resource, physical activity can lead to higher physiological functions and psychological effects in individuals, thereby preventing negative emotions such as anxiety.42 Studies have found that high school students’ high learning stress led to increased long-term sedentary or static behaviours, and HPA axis activation led to increased secretion of hormones such as cortisol, which promoted the secretion of the stress hormone epinephrine and triggered anxiety-like behaviours.43 In this study, physical activity had a mediating effect between learning stress and anxiety, suggesting that physical activity was a key factor for learning stress predicting anxiety. Learning stress reduced the sleep quality of high school students through physical activity, promoted individuals’ purposeless, closed-minded and critical attitudes towards attention processing, suppressed ‘reperception’ and automatic, dissociative, and conscious processing of thoughts and emotions, reduced autonomous motivation, self-esteem, and family well-being, and led to anxiety.44 45
Mobile phone addiction served as a critical factor affecting anxiety. High academic pressure significantly predicts mobile phone addiction in high school students. ACE theory suggests that people with higher academic pressure are more likely to use their mobile phones as a way to relieve negative emotions, and then develop addictive behaviours.46 Mobile phone addiction can positively predict anxiety,47 48 and anxiety can positively predict mobile phone addiction.49 Although studies have confirmed that anxiety has a mediating effect between academic stress and mobile phone addiction,50 there is no report on the correlation between academic stress and mobile phone addiction in high school students. According to the theoretical model of social support network, heavy learning pressure reduces the use of social support network of high school students, leading to the decline of subjective social support (emotional experience and satisfaction of being respected, supported and understood, etc). High school students will turn to smart phones to avoid inner troubles and even cause mobile phone addiction.31 High-intensity mobile phone use can not only weaken the individual’s positive emotional experience but also aggravate negative emotions, which can significantly negatively predict anxiety and other negative emotions.51 52
According to social support theory, the execution and experience of physical activity can make adolescents’ social behaviours more energetic and enthusiastic, and then reduce anxiety.30 Studies have confirmed that the level of physical activity can significantly and negatively predict mobile phone addiction, while the level of physical activity can also significantly and negatively predict the level of anxiety in high school students.31 32 The decline in physical activity level of high school students under high pressure will inhibit their self-regulation ability, leading to unrestrained use of smart phones and anxiety.53 In the current study, no study has been reported to explore the chain mediating effect of physical activity and mobile phone addiction on academic stress and anxiety in high school students. In this study, academic stress predicted anxiety through the chain mediating effect of physical activity level and mobile phone addiction. According to the cognitive-behavioural hypothesis proposed by Brand et al, academic pressure as an external stressor of high school students will reduce the level of physical activity, resulting in individuals to use mobile phones to seek internet comfort.54 According to the cognitive information processing theory, the interaction between cognition and emotion of individuals with mobile phone addiction is disordered, leading to the aggravation of negative emotions such as anxiety.54 55
This study demonstrated significant advantages in revealing the mechanism of the impact of academic stress on anxiety among high school students, and provided a unique and innovative scientific basis for mental health interventions through a large sample size and the use of chained mediated effects modelling. However, there are still some shortcomings, which are as follows: (1) the analysis of the mediating effect of learning stress is still to be refined, and future studies need to explore the direct or mediating role of each factor in more depth; (2) the sample is limited to high school students in Jiangsu Province, which is representative, and future studies should expand the scope of the survey to verify the broad adaptability of the results; (3) future studies should pay attention to the network of multiple mediating variables between learning stress and high school anxiety, such as the network of the mediating variables between learning stress and high school student anxiety, and the network of the mediating variables between learning stress and high school anxiety. between learning stress and anxiety in high school students.
Limitation and prospect
The findings of this study have certain theoretical value and practical guidance but also have some limitations. First, learning stress includes four factors, our research lacks direct or mediating role analysis of each factor, and mediating effect analysis should be more refined and precise. Second, this study only included high school students from the background of Jiangsu province, and the representativeness and richness of the sample need to be improved. Future research should focus on other populations in different regions and age groups to further evaluate the adaptability of this result. In addition, there are many mediating variables between learning stress and high school students’ anxiety, such as family intimacy, attachment, and social support. Future research should focus on the construction of multimediating variable networks.
Conclusion
Learning stress positively predicts the anxiety level of high school students. Moreover, anxiety can also be indirectly predicted through the independent mediating effect of physical activity level and mobile phone addiction, as well as the chain mediating effect of physical activity and mobile phone addiction.
Data availability statement
Data sharing not applicable as no datasets generated and/or analysed for this study.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants. The institutional review board of Yangzhou University approved this cross-sectional study (Approval Code: YZUHL2020018). Participants gave informed consent to participate in the study before taking part.
Acknowledgments
The authors thank all participants in the study. Furthermore, we thank Home for Researchers editorial team (www.home-for-researchers.com) for the language editing service.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
XC, PL and XY are joint first authors.
XC, PL and XY contributed equally.
Contributors XC and RY contributed to the study design, while PL, XY, and ZS contributed to the data collection. XC and RY contributed to the study design, while PL, XY, and ZS contributed to the data collection. XY, XZ and WL interpreted the results, whereas XC, PL and ZS drafted the manuscript and edited the language. All authors have read and approved the final manuscript. XC is the study guarantor.
Funding This study was supported by The National Social Science Foundation of China (CLA200279).
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.