Workplace violence inflicted by patients or their family members/visitors and its relationship with suicidal ideation among undergraduate medical students during clinical training in China

Abstract

Background

Workplace violence in healthcare settings is a significant public concern that profoundly impacts healthcare workers. However, there is a dearth of knowledge regarding the prevalence of workplace violence and its correlation with suicidal ideation among undergraduate medical students in China during their clinical training. The objective of this study was to evaluate the prevalence of workplace violence inflicted by patients or their family members/visitors and to assess its association with suicidal ideation among undergraduate medical students.

Method

The snowballing sampling technique was used to recruit Chinese medical students. A question designed by the research team was used to ask medical students about their encounters with workplace violence. Students’ basic demographic information and mental distresses (learning burnout, depression symptoms, anxiety symptoms, alcohol abuse/dependence, excessive daytime sleepiness and history of mental disorders) were also assessed. As appropriate, the data were analysed using descriptive statistics, chi-square tests, independent-sample t-tests and multiple logistic regression.

Results

Out of the 1402 undergraduate medical students who participated, 493 (35.2%) reported having experienced workplace violence inflicted by patients or their family members/visitors, of which 394 (28.1%) were verbal abuse, 14 (1.0%) were physical aggression, and 85 (6.1%) were suffered from both verbal abuse and physical aggression. Furthermore, students exposed to workplace violence were more likely to report suicidal ideation and had a higher prevalence of learning burnout, depression symptoms, anxiety symptoms, alcohol abuse/dependence and excessive daytime sleepiness. Depression symptoms, history of mental disorders, learning burnout and having a partner were significantly associated with suicidal ideation in this population.

Conclusion

The prevalence of workplace violence inflicted by patients or their family members/visitors was high among undergraduate medical students in China. This may be associated with their mental distress and suicidal ideation. Consequently, it is crucial to strengthen workplace safety measures and promptly implement interventions to mitigate the potentially serious consequences.

KEY MESSAGES

• Workplace violence was common among Chinese undergraduate medical students during clinical training and may be associated with suicidal ideation, learning burnout, depression symptoms, anxiety symptoms, alcohol abuse/dependence and excessive daytime sleepiness.

• Depression symptoms, history of mental disorders, learning burnout and having a partner were significantly associated with suicidal ideation among undergraduate medical students exposed to workplace violence.

Keywords:

• Workplace violence
• medical students
• medical education
• suicidal ideation
• burnout
• anxiety
• depression
• mental health

Introduction

Workplace violence is abuse or assault of staff in work-related situations that pose a challenge to their safety, well-being or health [1]. While the precise interpretation and manifestation of workplace violence may differ across nations [2], workplace violence in healthcare settings has become an urgent and significant social problem. This pervasive problem profoundly affects a wide range of healthcare workers, including physicians, nurses and other personnel [3,4]. A recent systematic study of a sample of 42,222 nurses from 13 different countries showed that 58% of the nurses had experienced workplace violence, including verbal abuse, physical violence, sexual harassment, threatening behaviour and bullying [5]. Similarly, a survey of 3016 doctors in 30 provinces in China revealed the prevalence of workplace violence, with 76.2% verbal abuse, 58.3% making difficulties, 40.8% defamation of reputation and 24.1% physical violence [6]. Previous research has demonstrated that the primary perpetrators of workplace violence in healthcare settings are patients or their family members/visitors, who account for the vast majority, ranging from 77.1% to 84.8%, followed by other coworkers [7,8]. To date, workplace violence has had a significant impact on healthcare workers and society as a whole. On one hand, the experience of workplace violence can detrimentally impact on employees’ mental health and quality of sleep, while also exacerbating symptoms of depression and burnout [6,9–11]; on the other hand, workplace violence in healthcare settings is associated with decreased job satisfaction and increased turnover intention of employees [12,13]. Therefore, early detection and prevention are necessary, especially given the global concern about physician shortages [14,15].

Extensive studies have shown that younger age and shorter years of service are associated with an increased risk of workplace violence [16,17]. Researchers have suggested that employees with shorter years of service are more vulnerable to workplace violence and have a higher risk of mental health problems than employees with longer years of service [18]. Therefore, the exposure of medical students to the risk of workplace violence and its adverse effects during clinical training is of concern. A survey among student nurses and doctors in Israel revealed that over 50% of the participants had encountered verbal violence during their clinical rotations [19]. A previous study reported that 68% of medical students in Turkey indicated that they had experienced some form of workplace violence, including 67% experiencing verbal violence and 16% experiencing physical aggression [20]. However, there are fewer studies on workplace violence among medical students in China [21–23], leading to a lack of awareness of this phenomenon and an urgent need for relevant studies.

Notably, student physicians are at a high risk of suicidal ideation and behaviours [24]. According to a prospective cohort study of student physicians (interns) in the United States, the prevalence of suicidal ideation was significantly higher during the internship than during the 2 months before the internship, and 16.4% of the interns had at least one suicidal ideation during their internship [25]. According to model for suicide risk, suicide risk is moderated by a range of factors both at the population and individual levels [26]. Individual risk factors include factors such as genetics, mental distress (e.g. depressive symptoms and anxiety) and traumatic events, which are significantly associated with and contribute to suicide risk. Workplace violence, as a traumatic event [27], has been shown in previous studies to be strongly associated with a high risk of suicide [28]. However, the relationship between workplace violence and suicidal ideation among medical students remains unclear. To bridge this gap, we conducted this web-based cross-sectional study to (1) assess the prevalence of workplace violence experience among medical students in clinical training, (2) assess the association between workplace violence and suicidal ideation among medical students in clinical training and (3) investigate the related factors of suicidal ideation among students who have experienced workplace violence. It is worth noting that the doctor–patient relationship in China is fragile, and many patients do not trust doctors [29], and thus workplace violence among healthcare workers is mainly caused by patients or their family members/visitors [7]. Moreover, the incidence of such workplace violence in China has been increasing in recent years [30]. Therefore, our study primarily targeted workplace violence caused by patients or their family members/visitors.

Materials and method

Study design and participants

The survey was conducted between 20 October 2020 and 5 April 2021 in China via an anonymous online questionnaire. This study is part of a large online study of the mental health of medical students and professionals in China [31,32]. We used the snowballing sampling technique to recruit participants. The inclusion criteria included (1) undergraduate medical students and (2) clinical training (internship) experience. The excluded criteria were as follows: (1) inability to read simplified Chinese text and (2) unwillingness to provide an electronic consent form.

Sample size

A sample size of 1056 produces a two-sided 95% confidence interval (CI) with a width equal to 0.060 when the sample proportion is 0.400. The sample proportion is based on the 40% of Chinese medical students exposed to workplace violence reported in a previous study [33]. The sample size was calculated using PASS 15.0.5.

Data collection and quality control

The data were collected through a Chinese online survey platform (www.wjx.cn) and a social media platform (WeChat). First, we select 40 medical students and medical university teachers from our affiliated hospital as the ‘original delivers’. They shared the survey link and invited their classmates or students to participate in the survey via WeChat group and Moments. Click on the survey link to participate in the study. In addition, participants were also encouraged to share the survey link via their WeChat group and Moments. Only participants who responded to all the questions were allowed to submit the questionnaire, thus, there were no missing data in this study. To ensure the validity of the questionnaire, the researcher predetermined the criteria for determining an invalid questionnaire: (1) incorrect answers to the general knowledge question; (2) all the answers were the same throughout the questionnaire; and (3) there was inconsistency in the logical consistency check.

Questionnaire development

The online questionnaire was written in Chinese. The questionnaire consisted of three sections. After clicking on the link, respondents were informed of the content and purpose of the study and asked to give consent in the first section. Respondents who chose the option ‘I have read the informed consent form and volunteer to participate in this survey’ were redirected to the second section. The second section lists questions related to sociodemographic characteristics. The last part contained measures of workplace violence; suicidal ideation; and mental health characteristics.

Sociodemographic characteristics

We used a questionnaire designed by our research team to collect the sociodemographic characteristics of all participants, including age, gender, relationship status and monthly income.

Workplace violence caused by patients or their family members/visitors

Medical students’ exposure to workplace violence caused by patients or their family members/visitors was assessed using a research team-designed question, ‘Have you ever experienced workplace violence (verbal abuse or physical aggression) from patients or their family members/visitors?’ They could choose one of the following options: ‘no’, ‘verbal abuse’, ‘physical aggression’ or ‘both verbal abuse and physical aggression’. Those who answered ‘no’ were considered not to have experienced workplace violence, while those who answered the other options were considered to have experienced workplace violence. This question was designed by two experts in our team, who are specialized psychiatrists in our country and have been focusing on the study of mental states/disorders in medical student/healthcare worker. Before starting the final questionnaire dissemination, it was piloted with 20 medical students. We then interviewed psychiatrists and these 20 students, all of whom indicated no difficulty in understanding and responding to this questionnaire. In addition, we chose the Workplace Violence Scale, a commonly used scale in China, as the standard [34]. A total of 130 medical students were recruited to answer this research team-designed question as well as the Workplace Violence Scale, and the specificity and sensitivity of this research team-designed question were calculated. The results showed specificity = 72% and sensitivity = 81%, indicating that the validity of this question to measure workplace violence is acceptable [35]. In addition, a similar question was used in a previous study to detect workplace violence among physicians [36].

Suicidal ideation

According to a previous study [37], suicidal ideation among students was measured by the question, ‘Have you ever thought about suicide in the past 12 months?’ They could choose one of the following options: ‘no’ and ‘yes’.

Mental health characteristics

Learning burnout of the students was measured using the Learning Burnout of Undergraduates Scale (LBUS) [38], with a total score >60 indicating learning burnout [39]. LBUS is highly reliable with a reliability coefficient (Cronbach’s α) of 0.915. The eight-item Patient Health Questionnaire (PHQ-8) was used to assess depression symptoms [40], and the seven-item Generalized Anxiety Disorder Scale (GAD-7) [41] was used to assess students’ anxiety symptoms, with scores >10 indicating the presence of moderate or severe depression symptoms (PHQ-8) or anxiety symptoms (GAD-7) [42,43]. The reliability coefficients (Cronbach’s α) for the GAD-7 and PHQ-8 showed excellent internal consistency at 0.911 and 0.882, respectively. Daytime sleepiness was measured using the Epworth Sleepiness Scale (ESS), with higher total scores indicating greater excessive daytime sleepiness [44]. The internal reliability of the ESS was high, with a Cronbach’s α of 0.796. The Alcohol Use Disorders Identification Test-Consumption (AUDIT-C) was used to identify students with alcohol abuse/dependence [45], with a score of ≥3 for women and ≥4 for men indicating alcohol abuse/dependence [46]. The reliability of AUDIT-C was acceptable, with a Cronbach’s α of 0.820. The history of mental disorders was assessed using the question, ‘Have you ever been diagnosed with mental disorders by a doctor?’ Participants who answered ‘yes’ were recorded as having a history of such a disorder.

Statistical analysis

Standard descriptive statistics were used to characterize the undergraduate medical students who responded. Categorical data were displayed as frequencies and percentages. The chi-square test or Fisher’s exact test was used for group comparisons. The normality of continuous variables was checked by using the Shapiro–Wilk normality test, Kolmogorov–Smirnov normality test and QQ-plot. The only continuous variables in our study conformed to a normal distribution, therefore we expressed it as the mean ± standard deviation (x ± s) and used the independent-sample t-test to compare the two groups. Multiple logistic regression was used to screen for independent variables associated with suicidal ideation among students exposed to workplace violence. Variables with p < .1 in the univariate analysis were included in the multiple logistic regression model. Multi-collinearity between the independent variables was determined using variance inflation factor (VIF), with VIF >5 indicating multi-collinearity. In addition, we performed Receiver Operating Characteristic (ROC) curves and calculated the Area Under the Curve (AUC) to determine the discriminatory capacity of our model to distinguish between medical students exposed to workplace violence with and without suicidal ideation. AUC values between 0.7 and 0.8 was generally considered acceptable [47]. All the statistical analyses were performed using SPSS for Windows (Version 26; IBM, Armonk, NY, USA). All tests were two-tailed, with a significance level of 0.05.

Ethical considerations

This study was approved by the ethics committee of the Second Xiangya Hospital of Central South University (LYF2020075). All participants were voluntary and informed consent was obtained through an online survey platform (www.wjx.cn). Immediately after completing the survey, participants will receive a report on their mental health status and recommendations to promote their mental health status. In addition, participants will be provided with contact information for our team of mental health professionals to contact us for assistance.

Results

Sociodemographic characteristics and the prevalence of workplace violence among medical students in clinical training

Of the 1440 returned questionnaires, 38 were considered invalid and were excluded from the analysis. Thus, 1402 questionnaires were included in the final analysis, with an efficiency rate of 97.3%. The sociodemographic characteristics of the participants are summarized in Table 1. There were 375 (26.7%) males and 1027 (73.3%) females. Their ages were mainly 19–21 years old (69.2%), their relationship status was mainly single (75.9%), and their monthly income was mainly 616–1310 Chinese Yuan [CNY] (52.3%). Out of these 1402 medical students, 493 (35.2%) medical students reported experiencing workplace violence inflicted by patients or their family members/visitors, of which 394 (28.1%) were verbal abuse, 14 (1.0%) were physical aggression, and 85 (6.1%) were both verbal abuse and physical aggression.

Table 1. Demographic characteristics and risk factors of medical students with and without workplace violence exposure.

Characteristics	Without workplace violence exposure (participants [%])	With workplace violence exposure (participants [%])	χ^2/z	p Value
Age (years)			6.335	.096
16–18	33 (3.6%)	10 (2.0%)
19–21	634 (69.7%)	337 (68.4%)
22–24	235 (25.9%)	137 (27.8%)
>24	7 (0.8%)	9 (1.8%)
Gender			0.988	.320
Male	251 (27.6%)	124 (25.2%)
Female	658 (72.4%)	369 (74.8%)
Relationship status			0.963	.326
Single	698 (76.8%)	367 (74.4%)
Partnered	211 (23.2%)	126 (25.6%)
Monthly income (CNY)			6.547	.088
<615	425 (46.8%)	231 (46.9%)
616–1310	289 (31.8%)	156 (31.6%)
1311–2086	164 (18.0%)	76 (15.4%)
>2086	31 (3.4%)	30 (6.1%)
Suicidal ideation	149 (16.4%)	134 (27.2%)	23.094	<.001
Learning burnout	278 (30.6%)	207 (42.0%)	18.375	<.001
Depression symptoms	110 (12.1%)	121 (24.5%)	35.958	<.001
Anxiety symptoms	66 (7.3%)	65 (13.2%)	13.242	<.001
Alcohol abuse/dependence	85 (9.4%)	69 (14.0%)	7.053	.008
Excessive daytime sleepiness	22.526 ± 4.653	22.805 ± 4.875	−1.056	.291
History of mental disorders	28 (3.1%)	22 (4.5%)	1.776	.183

The prevalence of suicidal ideation among medical students exposed to workplace violence versus those not exposed to workplace violence

Of the 1402 medical students, 283 (20.2%) reported suicidal ideation in the last 12 months. The prevalence of suicide ideations was significantly higher among students who have experienced workplace violence (n = 134/493, 27.2%) than among those who have not (n = 149/909, 16.4%) (χ² = 23.094, p < .001). This difference remained significant after controlling for characteristics including age, relationship status, monthly income, anxiety symptoms, depression symptoms, learning burnout, excessive daytime sleepiness, a history of mental disorder and alcohol abuse/dependence (odds ratio [OR]: 1.427, 95% confidence interval [95% CI]: 1.055–1.931, p = .021).

Comparison of sociodemographic, mental health characteristics between medical students exposed to workplace violence versus those not exposed to workplace violence

Compared to students who have not experienced workplace violence, those who have experienced workplace violence have significantly higher prevalence of learning burnout, depression symptoms, anxiety symptoms and alcohol abuse/dependence (all ps < .05) (Table 1). However, these two groups had no significant differences in sociodemographic characteristics, excessive daytime sleepiness and history of mental disorders (all ps > .05). In addition, we compared differences in the sociodemographic and mental health characteristics between students who have experienced different types of workplace violence and those who have not. Specifically, compared to students who have not experienced workplace violence, students who have experienced verbal abuse were more likely to report suicide ideation, learning burnout, depression symptoms, anxiety symptoms and being a female (all ps < .05). Students who have experienced physical aggression were more likely to report alcohol abuse/dependence and excessive daytime sleepiness than students who have not experienced workplace violence (all ps < .05). Students who have experienced both verbal abuse and physical aggression were more likely to report suicide ideation, learning burnout, depression symptoms, anxiety symptoms, alcohol abuse/dependence and being a male compared to students who have not experienced workplace violence (all ps < .05).

The related factors of suicidal ideation among medical students exposed to workplace violence

The univariate analyses of students who have experienced workplace violence demonstrated that suicidal ideation was significantly associated with the status of having a partner (p = .003), monthly income (p = .001), learning burnout (p < .001), depression symptoms (p < .001), anxiety symptoms (p < .001), alcohol abuse/dependence (p = .035) and history of mental disorder (p < .001), as shown in Table 2. We then performed a multiple logistic regression and included variables with p < .1 in the univariate analysis in this model (Table 3). The results of the Hosmer–Lemeshow goodness-of-fit test showed that χ² = 6.661, p = .574, indicating that the model fit was good. For medical students who have experienced workplace violence, suicidal ideation was significantly associated with depression symptoms (OR: 2.432, 95% CI: 1.392–4.247, p = .002), history of mental disorders (OR: 3.519, 95% CI: 1.249–9.915, p = .017) and learning burnout (OR: 2.380, 95% CI: 1.492–3.798, p < .001), as well as those with a partner (vs. participants who were single; OR: 1.880, 95% CI: 1.155–3.058, p = .011). Finally, the combination of PHQ-8, LBUS, gender and a history of mental disorders had a higher AUC value of 0.743 to distinguish medical students who have experienced workplace violence with and without suicidal ideation (Supplementary Figure S1).

Table 2. Demographical characteristics and risk factors of the medical students with workplace violence exposure who had or did not have suicidal ideation.

Characteristics	Without suicidal ideation (participants [%])	With suicidal ideation (participants [%])	χ^2/z	p Value
Age (years)			9.316	.019
16–18	5 (1.4%)	5 (3.7%)
19–21	258 (71.9%)	79 (59.0%)
22–24	91 (25.3%)	46 (34.3%)
>24	5 (1.4%)	4 (3.0%)
Gender			0.591	.442
Male	87 (24.2%)	37 (27.6%)
Female	272 (75.8%)	97 (72.4%)
Relationship status			8.760	.003
Single	280 (78.0%)	87 (64.9%)
Partnered	79 (22.0%)	47 (35.1%)
Monthly income (CNY)			15.845	.001
<615	169 (47.1%)	62 (46.3%)
616–1310	125 (34.8%)	31 (23.1%)
1311–2086	51 (14.2%)	25 (18.7%)
>2086	14 (3.9%)	16 (11.9%)
Learning burnout	123 (34.3%)	84 (62.7%)	32.367	<.001
Depression symptoms	58 (16.2%)	63 (47.0%)	50.174	<.001
Anxiety symptoms	31 (8.6%)	34 (25.4%)	23.884	<.001
Alcohol abuse/dependence	43 (12.0%)	26 (19.4%)	4.469	.035
Excessive daytime sleepiness	22.967 ± 4.787	22.373 ± 5.097	1.203	.230
History of mental disorders	8 (2.2%)	14 (10.4%)	15.462	<.001

Table 3. Factors associated with suicidal ideation in medical students who have been exposed to workplace violence.

Characteristics	B	Wald statistic	p Value	OR	95% CI	VIF
Relationship status	0.631	6.452	.011	1.880	1.155–3.058	1.006
Depression symptoms	0.889	9.752	.002	2.432	1.392–4.247	1.517
History of mental disorders	1.258	5.670	.017	3.519	1.249–9.915	1.068
Learning burnout	0.867	13.226	<.001	2.380	1.492–3.798	1.138

Sensitivity analysis

We have modified the measurement of depression symptoms, anxiety symptoms, alcohol abuse/dependence and learning burnout. Total scores for PHQ-8, GAD-7, AUDIT-C and LBUS were treated as continuous data without using cutoff values to treat them as categorical variables. A similar result was also found in this sensitivity analysis shown in Supplementary Table S1.

Discussion

Main findings

This survey indicated that more than a third of medical students have experienced workplace violence inflicted by patients or their family members/visitors during their clinical training. We found that medical students who have experienced workplace violence were more likely to report suicidal ideation than those who have not experienced it. Depression symptoms, history of mental disorders, learning burnout and having a partner were significantly associated with suicidal ideation among medical students who have experienced workplace violence.

Medical students were at a high risk of workplace violence in clinical training

Our findings indicated that workplace violence experienced by medical students during their clinical training warrants serious attention. The most common form of workplace violence experienced by medical students during clinical training was verbal violence, which aligns with other studies [6,48,49]. In addition, the overall prevalence of workplace violence among medical students was 35.2%, which is lower than in other countries (ranging from 47% to 68%) [19,20, 50]. This difference may be attributable in part to the different samples and methods used to assess workplace violence. For example, one study of US internal medicine trainees used the validated Workplace Violence in the Health Sector Survey Questionnaire to test for workplace violence, including verbal violence, physical, emotional and sexual violence [50], whereas another study of Israel nursing and medical students used a self-administered scale to measure workplace violence, including only verbal, physical and sexual violence [19]. Our study, however, focused only on physical and verbal violence. Furthermore, the present study focused only on workplace violence caused by patients or their family members/visitors, whereas some previous studies have included workplace violence from senior residents and others [20,50]. Although the proportion of medical students who experienced workplace violence in this study was lower than the previously reported prevalence of workplace violence among Chinese physicians [16,51], screening for workplace violence and targeted intervention for medical students is also important and necessary in China, as violence is unacceptable and reprehensible regardless of when, where or by whom it occurs. It is worth noting that the number of undergraduate medical students enrolled in 2020 has exceeded 2 million [52], which means that a large number of students could be potential victims of workplace violence and its adverse consequences and should therefore be given more attention. In addition, we call for more research on workplace violence to focus not only on medical students but also on other healthcare workers, such as doctors and nurses, who may also be potential victims of workplace violence caused by patients or their family members/visitors. The main reason for workplace violence in healthcare settings in China may be the tension between doctors and patients due to poor doctor–patient communication [6,51]. On the one hand, there is a serious shortage of human resources for health in China, with only 0.14 general practitioners and 0.25 nurses per 1000 people [53]. As a result, doctors rarely have enough time to fully communicate with patients about their condition and treatment options [54]. On the other hand, many doctors put more emphasis on providing excellent treatments than on communicating with patients [55]. This may lead to doctors ignoring the emotional needs of their patients and lacking the necessary humanistic care for them, which ultimately leads to doctor–patient conflicts and workplace violence [56]. Therefore, we propose a number of recommendations that may help to reduce the occurrence and risk of workplace violence. In terms of medical education, it is important to train more healthcare workers to address the shortage of human resources for health. Additionally, emphasizing the importance of doctor–patient communication in medical education can also be beneficial. At the hospital management level, improving workplace safety measures such as installing fences, security cameras and metal detectors and hiring security guards is essential [57]. It is also important to enhance the ability of healthcare workers to protect themselves and to assess, prevent and respond to workplace violence [58,59]. At the national legal level, making it a felony to assault a healthcare worker may also help reduce the risk of workplace violence.

Medical students exposed to workplace violence at a high risk of suicidal ideation

The present study suggested that the prevalence of suicidal ideation is significantly higher in medical students exposed to workplace violence than in those without such exposure. Similarly, a national survey of general surgery residents demonstrated that residents who reported verbal or physical abuse were more likely to have suicidal ideation [60]. One possible reason for this is that workplace violence negatively impacts the mental health of affected residents (e.g. leading to emotional exhaustion, depressive symptoms, anxiety symptoms, etc.) [61,62], which is significantly associated with suicidal ideation [63,64]. Lack of social support may also increase the risk of suicidal ideation among medical students exposed to workplace violence [65,66]. Previous studies have reported that a large percentage of healthcare workers are unaware of the support services available to them when they experience workplace violence [67,68]. The present study also found that the prevalence of mental distress (learning burnout, depression symptoms, anxiety symptoms and excessive daytime sleepiness) was higher among medical students who had been exposed to workplace violence, which was consistent with previous studies conducted among doctors and nurses [9,10]. Therefore, efforts such as establishing workplace violence prevention procedures, education on coping with workplace violence and timely screening and intervention (e.g. treatment by mental health professionals) of mental health issues among medical students exposed to workplace violence are urgently needed to prevent the adverse consequences of workplace violence.

Learning burnout, depression symptoms, a history of mental disorder and having a partner are related factors for suicidal ideation among medical students who reported workplace violence exposure

In the present study, learning burnout has been recognized as an independent risk factor for suicidal ideation among medical students who reported workplace violence exposure. A recent review of suicidal ideation among medical students also demonstrated that burnout was significantly associated with suicidal ideation [69]. Notably, a longitudinal cohort study of medical students in the United States suggested that burnout might be reversible [37]. Thus, screening and interventions for burnout may significantly reduce the risk of suicidal ideation among medical students. Furthermore, among medical students exposed to workplace violence, students with suicidal ideation were more likely to have depression symptoms and a history of mental disorders. Similarly, previous studies have demonstrated that depression symptoms and mental diseases are associated with an increased risk of suicidal ideation in medical students [69–72] and the general population [63,73,74]. To our surprise, having a partner was an independent risk factor for suicidal ideation among medical students who had been exposed to workplace violence, which was different from the findings of earlier studies. For example, studies of Norwegian and Italy medical students both reported that being in a relationship was significantly associated with a lower level of suicidal ideation [75,76]. However, a study of Serbia medical students demonstrated no significant association between relationship status and suicide risk [77]. The different study samples may have caused this difference. we focused on medical students who had been exposed to workplace violence, whereas other studies were focused on medical students and did not specifically consider whether these students had experienced workplace violence [75–77]. However, to the best of our knowledge, we are the first study to examine the associated factors of suicidal ideation among medical students who had been exposed to workplace violence, so more research is necessary to examine the relationship between suicidal ideation and relationship status. In addition, work–life conflict, a risk factor for suicidal ideation [78], was common among healthcare workers in China, especially those with a partner [79]. Thus, medical students exposed to workplace violence may have difficulty in obtaining support from their partner and face work–life conflict, which is related to their suicidal ideation. Meanwhile, pressure from family/partner, such as economic burden and family/partner responsibilities, may increase their likelihood of suicidal ideation [80]. However, because this study did not directly assess work–life conflict and pressure from family/partner among the medical students, we can only speculate that work–life conflict and pressure from family/partner may account for the existence of an association between relationship status and suicidal ideation. Further studies are needed to examine this association in detail.

Limitations of the study

There are several limitations to this study. First, the cross-sectional design adopted in this study limited the examination of causality, suggesting the need for further longitudinal studies. Second, workplace violence and suicidal ideation were both measured through one self-report question, which made responses more subjective; thus, more discriminative measurement tools are needed to examine these issues more objectively. Third, we only examined verbal abuse and physical aggression inflicted by patients or their family members/visitors. In future studies, other forms of workplace violence (e.g. sexual harassment and damage to reputation) and workplace violence exerted by others (e.g. supervisors and colleagues) should also be included to make the results more comprehensive. Fourth, only questionnaires from participants who answered all the questions were included in the study. When confronted with questions that made participants feel uncomfortable, they may have refused to answer and thus not participated in this study, which may have led to selection bias. So future studies should allow participants to choose not to answer but still ‘answer’ the submitted questions, for example, by providing the option ‘I do not wish to answer’. Fifth, all participants were anonymous and completed this survey through self-reporting, which may have made responses subjective. Therefore, potential bias may have been created, which needs to be improved in future studies. Finally, this study was conducted at the late stage of the COVID-19 pandemic. Although the quarantine was lifted in most areas, it is unclear whether this increased the susceptibility of the participants or whether it affected the likelihood of workplace violence, which requires further research.

Conclusion

In summary, this study suggested that workplace violence was not uncommon among medical students in China, which highlighted the need for workplace violence prevention for medical students during clinical training. In addition, medical students with exposure to workplace violence were more likely to report suicidal ideation and mental distress than those without such exposure. Thus, timely screening and prevention procedures are essential to prevent adverse consequences of workplace violence. It is worth noting that workplace violence and adverse consequences for other healthcare workers are also of great concern in addition to medical students. Therefore, future works are warranted to focus on workplace violence among different healthcare workers and facilitate the development of relevant interventions.

Author contributions

QXW and TQL contributed to all aspects of the study. XW, QXW, PP and WFZY contributed to the drafting of the manuscript. QXW, TQL, YYT, YHL and JST participated in the conception and design of the study. All authors contributed to survey development and data acquisition. XW, PP, QXW, YHL, QJW and YZH contributed to data analyses and result interpretation. All authors critically revised the manuscript and agreed to be responsible for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Ethical approval

The study was approved by the ethics committee of the Second Xiangya Hospital of Central South University.

Acknowledgements

We give our greatest gratitude to all the participants and volunteers who made contributions to the online survey.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Additional information

Funding

This study was supported by the Provincial Natural Science Foundation of Hunan [Grant No. 2020JJ4795] to TQL and the Scientific Research Launch Project for new employees of the Second Xiangya Hospital of Central South University to QXW.