The effects of workplace violence on the job stress of health care workers: buffering effects of wellbeing HRM practices

Abstract

Contemporary research has focused on antecedents, incidences, and consequences of workplace violence in the healthcare sector. However, little is known about the role of HRM in supporting healthcare workers to cope with negative emotions arising from workplace violence. Grounded in the conservation of resources theory, this study examines the moderating effects of wellbeing HRM practices (WBHRM) on the relationship between workplace violence and job stress. We also examine the mediation effects of job stress on the relationship between workplace violence and quality of care. Data were gathered from 254 and 225 (Study 1) aged care workers (nurses and personal care assistants) in Australia and 136 healthcare workers (doctors and nurses) in China (Study 2) across multiple waves. We use a new measure of WBHRM in this paper to test our hypotheses. Results show that the presence of WBHRM practices moderated the relationship between workplace violence and job stress. We also found that job stress mediates the relationship between workplace violence and quality of care. The findings contribute to the HRM literature by providing an innovative WBHRM approach to support health care workers to cope with job stress after they experience workplace violence.

Keywords:

• Wellbeing HRM
• health care workers
• workplace violence
• job stress
• quality of patient care

Introduction

Despite the high rates of violence against healthcare workers across aged care facilities and hospitals (Acquadro et al., 2018; Ramacciati et al., 2019) there remains limited research on how organisations manage the emotional effects of workplace violence against these workers (Jia et al., 2020; Kobayashi et al., 2020; Pariona-Cabrera et al., 2020). The purpose of this study is to examine workplace violence against doctors, nurses and personal care assistants (PCAs) in Australian aged care facilities and Chinese hospitals. We develop the means for measuring ways to manage and mitigate workplace violence and its impact on job stress and quality of care across different organisational and national contexts. The World Health Organisation (2003, p. 1) defines workplace violence as ‘incidents where employees are abused, threatened, assaulted or subject to other offensive acts or behaviours in circumstances related to their work’. There are two types of violence: first, physical violence includes biting, kicking, stabbing, pinching; and second, psychological violence occurs when one person threatens another with verbal aggression, humiliating comments, intimidation, or insults (ILO, 2020). Healthcare workers suffer the highest rates of injury due to violence at work. This accounts for rates that are five times higher as compared with employees from different industries (U.S. Bureau of Labor Statistics, 2018). In Australia, more than 60% of health care professionals report exposure to verbal or physical violence in 2021 (WorkSafe Victoria, 2022). A recent Chinese study conducted by Tian et al. (2020) found that 56.4% (N = 2078) of hospital healthcare workers (doctors and nurses) had experienced workplace violence in the previous 12 months. Zhao et al. (2018) revealed that between 2000 and 2015, 290 incidents of violence lead to serious injury or death of nurses in the Chinese healthcare context. These statistics are alarming and call for urgent action to be taken by organisations to mitigate the adverse effects of workplace violence on healthcare workers.

Previous studies have focused on high performance work systems (HPWS) to improve employee performance (Ang et al., 2013; Cooke et al., 2019). However, there is growing evidence that HPWS is associated with increased employee burnout and job stress (Bartram et al., 2012; Cooper et al., 2019). Wellbeing human resource management (WBHRM) aims to reduce the negative effects of job stress and further enhance employee performance (Guest, 2017). WBHRM practices involve ‘five sets of HRM practices that are designed to promote wellbeing and positive employment relationship’ (Guest, 2017, p. 30). The first set on investment in employees involves training and development; fosters the enhancement of learning and knowledge of new skills. The second set on provision of engaging work involves job quality, gives autonomy to decide how to undertake a job. The third set on the provision of a positive social and physical environment involves occupational health and safety, gives assurance to employees to feel safe and secure at work. The fourth set on employee voice involves participation in decision-making and gives employees the opportunity to voice their ideas and share their knowledge with others. The fifth set on organisational support involves work-life balance, which promotes a balancing approach between work and non-work demands. WBHRM practices can enhance levels of trust, fairness and security, which may contribute to managing the effects of negative work events such as workplace violence and enhance employee wellbeing (Guest, 2017). Employee wellbeing is associated with increased individual performance (Van Veldhoven & Peccei, 2015) and in the context of nursing, high quality of care (Cooke and Bartram, 2015). A study conducted by Cooper et al. (2019) of 561 Chinese employees found that WBHRM practices increase the development of resilience and employee performance. The findings highlight that implementing day-to-day WBHRM practices, particularly training interventions, job quality, use of teams and supportive leadership (WBHRM practices) may contribute to the development and support of resilient workforce, as well as employee wellbeing and performance. We argue that a well-rounded ‘diet’ comprising five sets of WBHRM practices (Guest, 2017) may be critical to manage and mitigate the effects of workplace violence (Pariona-Cabrera et al., 2020).

Some governments and international organisations have examined ways to better manage violence in healthcare organisations through legislation, policy edicts and administrative guidelines (Congress Gov, 2021; ILO, 2020). For example, the USA approved the Workplace Violence Prevention for Health Care and Social Services Act to address incidents of violence against workers. The International Labour Organisation (ILO, 2020) recently launched a report that highlights the role of a comprehensive OHS framework for addressing violence at work. Moreover, Shao et al. (2023) have suggested the implementation of some HRM practices such as security measures and anti-violence training, which are necessary to support healthcare workers. To further extend this line of research, we argue that it is critical that healthcare organisations take measures to reduce the negative effects of violence against employees’ wellbeing through WBHRM practices.

Workplace violence has a significant negative impact on healthcare workers because it affects their psychological and physical health, leading to financial and social costs to employees, organisations and society (Safe Work Australia, 2017). At the employee level, the quality of work-life is negatively impacted because of the presence of distressing events such as violence at work, which may decrease employees’ wellbeing and performance (Baby et al., 2018). There has been a call for further research regarding ‘the use of HRM practices to promote employee wellbeing on managing the negative consequences of workplace violence and subsequent effects on quality of care’ (Pariona-Cabrera et al., 2020, p. 9). In this study, we respond to this call by examining the role of WBHRM practices to manage and mitigate the negative emotional effects of workplace violence on doctors, nurses and PCAs.

Drawing on conservation of resources (COR) theory, we seek to examine the relationship between workplace violence and job stress and the mediating effects of job stress on the relationship between workplace violence and the quality of care. We also test the moderating effects of WBHRM practices on the relationship between workplace violence and job stress. We conducted two studies: Study 1 examines first, the relationship between workplace violence and job stress; second, the mediating role of job stress on the relationship between workplace violence and quality of care; and third, the moderating role of WBHRM on the relationship between workplace violence and job stress. Study 1 comprises 254 and 225 nurses and PCAs union members employed in aged care facilities across the state of Victoria, Australia. In Study 2, we recruited 136 doctors and nurses from Chinese hospitals to replicate findings of Study 1. In this study, given that we examine both aged care facilities and hospitals, we use the term ‘quality of care’ to encompass both ‘quality of resident care’ and ‘quality of patient care’.

This article makes two contributions to the HRM literature. First, using COR theory (Hobfoll, 1989) and Guest’s (2017) conceptualisation of WBHRM, we unpack the theoretical process through which WBHRM practices reduce the effects of workplace violence on the job stress of health care workers and subsequent quality of patient care. We do this through an examination of the role of WBHRM as an organisational resource to buffer the negative effects of workplace violence on employees’ job stress. The buffering role of WBHRM practices provides a richer understanding of the theoretical mechanisms through which accumulation and protection of wellbeing resources supports the mental health of workers. Second, we developed a new measure for WBHRM based on the conceptual work of Guest (2017) to systematically analyse how bundles of WBHRM practices can be implemented to help workers manage job stress. This new measure of WBHRM comprised of five sets of HRM practices (Guest, 2017) provides novel insights for HRM scholars and practitioners to improve the wellbeing of workers in a systematic, measurable and holistic way.

Theoretical framework and hypotheses development

Workplace violence, job stress and quality of care

Based on COR theory, job stress is a critical response to the threat or experience of loss (Hobfoll, 2001). Job stress is defined ‘as a reaction to a difficult work environment in which key resources are threatened with loss or when key resources are lost, or when there is a failure to gain key resources following significant effort’ (Hobfoll et al., 2018, p. 104). COR theory includes different types of resources such as object resources, personal resources, energy resources and conditional/organisational resources (Hobfoll & Shirom, 2000). Object resources comprise security cameras, job equipment, and security personnel. Personal resources involve skill sets, efficacy beliefs and personal traits. Energy resources contain experience, abilities and knowledge. Conditional/organisational resources include social support, job autonomy, and work environment (Hobfoll, 2011). In this paper, we propose when doctors, nurses and PCAs face challenging work situations such as violence at work, they may draw on conditional/organisational resources reserved from their work (Schat & Kelloway, 2000). The conditional/organisational resources may include a bundle of WBHRM practices such as investment in employees (e.g. training and development), provision of engaging work (e.g. job quality), creation of a positive social environment (e.g. OHS), employee voice (e.g. participation in decision making) and organisational support (e.g. work-life balance) (Guest, 2017). However, if there is a threat of a net loss of these resources or a lack of resource gain, doctors, nurses and PCAs may experience job stress (Itzhaki et al., 2018).

WBHRM practices as conditional/organisational resources play a critical role in managing the effects of workplace violence (Salanova et al., 2010). For example, when doctors, nurses and PCAs increase access to a pool of resources (WBHRM practices), they may enhance coping and adaptation skills, which may result in the increase of sense of control to manage adverse work situations and negative outcomes (Hobfoll et al., 2018). WBHRM practices can be linked together as resource caravans, which may operate as a system to curb the adverse effects of workplace violence on job stress. Therefore, this may impact on positive organisational outcomes including employee satisfaction and employee engagement (Bakker & Demerouti, 2007; Halbesleben et al., 2014). The next section (Hypothesis 3) further explains the role of resource caravans.

In contrast, the resource-draining presence of a work stressor (e.g. workplace violence) leading to job stress can substantially minimise healthcare workers’ performance affecting their quality of care. This occurs when the level of job demands is greater than the level of replenished job resources. Clearly, if doctors, nurses and PCAs cannot counteract major stressful conditions and have fewer resources to offset resource loss (Hobfoll, 2011), they may experience a depletion of energy and professional efficacy (Salanova et al., 2010). Subsequently, this may undermine their motivation and job satisfaction, affecting quality of care (Magnavita et al., 2020). Previous studies have shown that incidents of violence may lead to negative emotional outcomes such as job stress which can impact quality of care (Magnavita et al., 2020). For example, a study by Acquadro et al. (2018) of 108 Italian hospital employees and 96 volunteers revealed that employees exposed to physical or verbal violence experienced greater job stress as compared with employees that did not experience violence. In a cross-sectional study of 275 Italian hospital nurses, Magnavita et al. (2020) found that exposure to workplace violence is significantly associated with job stress/strain. The researchers emphasise that violence at work is significantly associated with lower employee job satisfaction and reduced quality of patient care.

Despite the existing evidence, Shablon et al. call for additional research to examine the adverse effects of violence on job stress and quality of care, particularly in aged care settings. Healthcare employees in inpatient geriatric care are most likely to be affected by daily physical and verbal attacks, particularly from patients with dementia (Rodwell et al., 2015). Based on the preceding literature, we argue that workplace violence is detrimental to psychological health of aged care and hospital workers because it may deplete their emotional and physical capacity which can lead to negative performance outcomes such as reduced quality of care. Therefore, we propose the following hypotheses:

• Hypothesis 1: Workplace violence predicted job stress

• Hypothesis 2: Job stress mediates the relationship between workplace violence and quality of care

WHRM practices as a resource moderating the effect of workplace violence on job stress

In this paper, we underpin our examination of the moderating effect of WBHRM practices on the relationship between workplace violence and job stress using COR theory. Drawing on this theory, Hobfoll (2001, p. 353) argues that ‘in the wake of severe stress, individuals seek to both repair the damage and to mobilise resources for further resource accumulation and protection’. In this study, we employ this concept of accumulation and protection of resources (Hobfoll, 2011), in which healthcare workers actively seek to accumulate resources and protect themselves from feelings of job stress using WBHRM practices. The five sets of WBHRM practices can be considered as conditional/organisational resources (COR theory). The implementation of these practices can enhance adaptation and coping skills and overall wellbeing of employees (Salanova et al., 2010) because these resources can protect employees’ self-esteem and emotional stability (Schaufeli & Bakker, 2004). Similarly, these resources can be linked together to create resource caravans to protect healthcare workers from the negative effects of violence on their mental health (Hobfoll, 2011). The benefit of these resources occurs when employees have access to a pool of resources that can increase their sense of control to manage challenging work events and undesirable outcomes. This process is underpinned by accumulation and integration of resources to create resource caravans (Salanova et al., 2010) to protect against resource depletion such as job stress (Hobfoll, 1989). The next section explores the role of resource caravan passageways.

Resource caravans are created and sustained within resource caravan passageways, that is, ‘the environmental conditions that support, foster, enrich, and protect the resources of individuals, segment of workers, and organisations in total’ (Hobfoll, 2011, p. 119). Importantly, resource caravans involve organisational support, stability and safety that can create and sustain organisational ecologies. We argue that the bundle of HRM practices (all five sets of WBHRM practices) act as a resource caravan as they are mutually reinforcing and provide a comprehensive suite of support to healthcare workers. Organisations can create passageways in which resources can be supplied, protected, shared, fostered and pooled (Hobfoll, 2011). For example, investing in healthcare workers (first set of WBHRM practices) that promotes development of skills and capacities can foster greater learning and development of self-efficacy and a sense of organisational support (Guest, 2017), which are critical aspects of building resources to alleviate the effects of workplace stressors (Halbesleben et al., 2014). Similarly, provision of engaging work involves information sharing (second set of WBHRM) helps employees connect to co-workers by making the knowledge they have available to others through two-way dialogue. This environmental condition can enrich and foster resources for employees by developing new ideas and ways to cope with adverse work situations, which may thereby contribute to resource caravans. Consequently, in response to adverse work situations such as workplace violence, doctors, nurses and PCAs strive to obtain and retain resources to develop their resource caravan passageways, replenish resource loss and counteract resource depletion (Salanova et al., 2010).

WBHRM practices can also help satisfy healthcare workers’ need for resources. When these practices are implemented, employees can gain resources to replace and reinforce other resources during negative work events such as workplace violence (Hobfoll et al., 2018). For example, provision of positive social and physical environment (third set of WBHRM) that prioritise employees’ health and safety can contribute to feelings of support from their managers. This adds accumulation and protection of wellbeing resources, allowing employees to feel protected against resource depletion such as feelings of job stress. Similarly, provision of employee voice (fourth set of WBHRM) empowers employees to participate in decision-making. Opportunities for expression of voice can empower to employees to complete their job tasks in a way they prefer, such autonomy may contribute to the management of job stress. This allows healthcare workers to obtain, conserve and redirect resources where they are needed. Therefore, Hobfoll (2011, p. 119) reiterates that ‘an ecological approach [resource caravan passageway] looks to motivate individuals to maximize their performance and application of shared resources to bring the lowest up to a higher standard’.

Another aspect of COR theory refers to ‘resource gain that increases in salience in the context of resource loss. That is, when resource loss circumstances are high, resource gains become more important and they gain in value’ (Hobfoll et al., 2018, p. 105). For example, the fifth set on organisational support involves developmental performance management, which enables two-way communication addressing goals setting, development planning and supportive feedback. By promoting this ongoing process, employees can enhance adaptation and coping skills through resource gain, which are well documented to enhance employee wellbeing. WBHRM practices may contribute greater resources to support doctors, nurses and PCAs experiencing stressful events (Hobfoll, 2001). Specifically, Hobfoll et al. (2018) suggest that resources can help individuals cope with job stress by reinforcing positive aspects of the self in stressful times.

Moreover, ‘when resources are outstretched or exhausted, individuals enter a defensive mode to preserve the self’ (Hobfoll et al., 2018, p. 106). In this case, individuals have two options: they may conserve resources (evolutionary strategy) or they may search for adaptation strategies (exploratory strategy). When nurses experience feelings of job stress because of adverse work experiences, they may show a defensive withdrawal to rethink or wait for help (evolutionary strategy) or seek adaptation or new coping strategies (WBHRM practices, exploratory strategy). WBHRM practices provide alternative strategies to cope with negative work events to increase their feelings of self-determination (Dollard & Bakker, 2010).

Emerging research has linked the role of resources and how to address workplace violence and its effect on employees’ mental health. For example, Kobayashi et al. (2020) asserted that the implementation of protective factors, including training and development, can help in enhancing healthcare workers’ self-efficacy skills to cope with the negative emotional effects of violence. Similarly, Kwak et al. (2020) examined the impact of workplace violence on quality of life among 399 clinical nurses in Korea. Researchers suggested that the provision of organisational-level interventions reduces exposure to emotional labour and workplace violence. Effective strategies need to combine prevention and health promotion interventions. Despite the growing interest in the scholarship of workplace violence, little is known about the role of WBHRM in supporting healthcare workers to cope with job stress arising from workplace violence. Consistent with extended literature, we expect that WBHRM practices will moderate the relationship between workplace violence and job stress:

• H3: WBHRM practices moderate the relationship between workplace violence and job stress, such that when WBHRM is at a high rather than low level, the positive relationship between workplace violence and job stress will be weaker.

Purpose of the current research

We conducted two studies: In Study 1, we investigated the relationship between workplace violence and job stress, the mediating effects of job stress on the relationship between workplace violence and quality of care. We also examined the moderating role of WBHRM on the relationship between workplace violence and job stress on nurses and PCAs union members employed in aged care facilities across the state of Victoria, Australia. In Study 2, we recruited doctors and nurses from Chinese hospitals to replicate findings of Study 1. Study 1 comprises 254 and 225 nurses and PCAs union members employed in aged care facilities across the state of Victoria, Australia. Study 2 involves 136 doctors and nurses employed in Chinese hospitals This was undertaken to examine the generalisability of our WBHRM scale and its utility in the management and mitigation of the negative effects of workplace violence among different worker cohorts, different health care contexts and countries. We selected Australian and Chinese health care contexts because health care workers are experiencing high levels of workplace violence. Moreover, to date, health care organisations in these contexts have not been able to develop effective HRM mitigations strategies or adequately manage the effects of workplace violence on the mental health of workers.

Study 1

Methods

The purpose of this first study was to determine whether a relationship exists between workplace violence and job stress, as well as how job stress mediates the relationship between workplace violence and quality of care. We hypothesised that incidents of violence will increase job stress and the presence of job stress will mediate the impact of workplace violence on quality of care. We also hypothesised that WBHRM will moderate the relationship between workplace violence and job stress.

Participants and procedures for Study 1

Online surveys were sent to members of a large nursing union employed in aged care facilities across the State of Victoria, Australia. We received 254 usable responses in the first wave and 225 useable responses in the second wave. Respondents completed the online survey during the COVID 19 pandemic between February–March 2020 for the first wave and May–June 2020 for the second wave. Approximately 72% of nurses and PCAs worked part-time. Employees were nurses (57%) followed by PCAs (24%). Nurses’ and PCAs’ ages ranged from 22 to 70 years, with a mean of 50. Nearly 95% of the sample speak English. Various education levels were represented in the sample, with the largest group holding certificate (33.1%), diploma (27.2%), and bachelor’s degree (25.2%).

Measures for Study 1

All responses were rated on a 5-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree). The workplace violence measure, however, was rated on a frequency of never, 1 time, 2–3 times, or 4 or more times. Workplace violence, job stress, and WBHRM practices were collected in Time 1, with the first survey. Quality of care was collected in Time 2, with the second survey.

Workplace violence was measured with 11 items from Rogers and Kelloway (1997). This measure consists of two dimensions: first, ‘physical violence at work (e.g. have you been hit, kicked, grabbed, shoved or pushed by anyone while you’ve been at work?); second, psychological or verbal violence at work (e.g. have you been yelled at or shouted at while you’ve been at work?)’ (Rogers & Kelloway, 1997, p. 65). The reliability of each dimension through internal consistency using Cronbach’s alpha was α = 0.88 for physical violence at work and psychological or verbal violence at work was .90.

Job stress was measured with ‘depression, anxiety and stress’ (DASS) scales that contain 21 items (Henry & Crawford, 2005). For the purpose of this study, we used the stress subscale that consists of seven items. A sample item of stress is ‘I tend to over-react to situations’ The Cronbach’s alpha for job stress was .90.

Quality of care was measured based on the scale developed by Bartram et al. (2014), who adapted the quality of care measure from the Victorian Patient Satisfaction Monitor (VPSM) to investigate perceptions of healthcare workers related to quality of care. The developed measure consisted of a total of 16 items. ‘This includes perceptions of courtesy, helpfulness, responsiveness, willingness to listen, the provision of information by staff, and perceptions of safety, privacy, and respect for patients’ (Bartram et al., 2014, p. 2407). The measure contains the word ‘patients’ but this was replaced with ‘residents’, as this word is also more familiar to respondents working in aged care facilities. The reliability coefficient was acceptable for this scale at .94.

WBHRM practices are aligned with the five sets of WBHRM practices as developed by Guest (2017). The five sets of practices comprise ‘investing in employees (e.g. training and development), providing engaging work (e.g. job quality), providing positive social and physical environment (e.g. OHS) promoting employee voice (e.g. participation in decision making), and providing organisational support (e.g. work-life balance)’ (Guest, 2017, p. 31).

WBHRM practices were measured using different validated scales aligned to the five sets of practices. This involves 10 sub-scales: recruitment and selection (Edgar & Geare, 2005), training and development (Edgar & Geare, 2005), occupational health and safety (OHS) (Edgar & Geare, 2005), equal employment opportunity (Edgar & Geare, 2005), job quality (Zacharatos et al., 2005), performance management (Bartram et al., 2007), opportunities for promotion (Price & Mueller, 1981), participation in decision making (Vanyperen et al., 1999), rewards (Vandenberg et al., 1999), and work-life balance (Nitzsche et al., 2014). These sub-scales are aligned with the purpose of this study which is focused on WBHRM practices. Please see Appendix A for all items in the scale.

An EFA on the items revealed that after items that cross-loaded and with low (<.40) loadings were dropped, a final set of 35 items loaded on five factors with Eigenvalues greater than one.

Results – Study 1

Prior to testing our hypotheses, we conducted a CFA to verify the factor structure of the variables.

The CFA provided mixed results (CFI = .90; IFI = .90; RMSEA = .09), indicating that the model has a less than ideal, but a somewhat acceptable fit.

While the chi-square difference test indicated that each model was significantly different from another as seen in Table 2, the principle of parsimony forces researchers to abandon complex models that are tweaked to observed data in favour of simpler models (Vandekerckhove et al., 2014). Therefore, our hypothesised four-factor model, on balance with consideration of the theory that underpins it, is acceptable.

Table 2. Descriptive statistics in Study 1 (Australian sample).

	Mean	SD	1	2	3	4
1. Workplace violence	2.82	0.69	(0.94)
2. Job stress	2.56	0.93	0.21**	(0.90)
3. Quality of care	4.65	0.45	−0.20	−0.13	(0.80)
4. WBHRM	2.55	0.74	0.29**	−0.41**	0.07	(0.87)

N = 254. *p < .05, **p < .01, ***p < .001.

Table 2 reports descriptive statistics for the Study 1 variables (Australian sample)

Hypothesis 1 predicted that workplace violence predicts job stress. Initial evidence of the relationship was seen in the correlation table (Table 1; r = .02, p <.001). Results of the OLS regression analysis are presented in Table 3.

Table 1. Comparison of model fit indices and chi-square differences of alternate and baseline models (Australian sample).

Model	χ ^2	df	χ ^2 diff	IFI	CFI	RMSEA
Hypothesised four-factor model (WBHRM, job stress, violence, and quality of care)	6670.35***	854		.80	.80	.09
Three-factor model: (violence, WBHRM, as separate latent variables, with job stress and quality of care loaded on one latent variable)	8932.00***	860		.66	.66	.11
Latent variable CMV test	6303.10***	832	367.25***	.73	.73	.10

***p < 0.001.

Table 3. Regression analysis: workplace violence predicting job stress in Study 1 (H1 – Australian sample).

Variables	b	s.e.	β
Constant	1.74**	0.24
Workplace violence	0.28***	0.83	0.21
R^2	0.45***
ΔR^2	0.45***

N = 254, *p < 0.05, **p < 0.01, ***p < 0.001.

Results of the OLS regression analysis indicate support for H1, which shows that workplace violence (b=.28 se = 0.83, p<.001), is significant predictor of job stress. H1, therefore, is supported.

Hypothesis 2 was tested through mediated regression analysis. This hypothesis postulated that job stress mediates the relationship between workplace violence and quality of care. Figure 1 presents the result of the mediated regression (Australian sample).

Figure 1. Job stress mediates the relationship between workplace violence and quality of care in Study 1 (H2 – Australian sample).

Figure 1 shows the direct effect of workplace violence on stress was statistically significant (b=.21, p<.001), indicating that nurses and PCAs scoring higher on violence are more likely to experience job stress than those scoring lower on the measure. The direct effect of job stress on quality of care was statistically significant (b=−0.14, p<.001), indicating that nurses and PCAs scoring higher on job stress are more likely to lower their quality of care than those scoring lower on the measure. The indirect effect was tested using bootstrapping procedures. The indirect effect of workplace violence via stress was statistically significant: .01, 95% CI= (−0.04 to −0.02), indicating that the mediational effect occurred in which workplace violence influences quality of care through job stress. Therefore, hypothesis 2 is supported.

Hypothesis 3 predicted that WBHRM practices would moderate the relationship between workplace violence and job stress, such that when the perception of WBHRM was high, job stress is lower. Table 4 presents the results.

Table 4. Moderated regression analysis: WBHRM moderating the relationship between workplace violence and job stress in Study 1 (H3 – Australian sample).

Variables	Step 1			Step 2			Step 3
	b	s.e.	β	b	s.e.	β	b	s.e.	β
Constant	2.01	.17		3.23	.26		4.37	.61
Workplace violence	.19***	.06	.14	.07***	.06	.05	−0.33***	.05	−0.27
Well-being oriented HRM practices				−0.33***	.05	−0.27	−0.74***	.20	−0.62
Workplace violence x HRM							.15*	.07	.38
R^2	.02***			.09***			.98*
ΔR^2	.02***			.07**			.08*

N = 254, ^†p < .10, *p < .05, **p < .01, ***p < .001.

Step 2 of Table 5 introduces the moderator (WBHRM) and Step 3 introduces the workplace violence x HRM interaction term. The interaction term is significant (b =.15, s.e. =.07 p <.05). Figure 2 presents the interaction between workplace violence and HRM.

Figure 2. WBHRM moderating violence-job stress relationship (Australian sample).

Table 5. Comparison of model fit indices and chi-square differences of alternate and baseline models (Chinese sample).

Model	χ ^2	df	χ ^2 diff	IFI	CFI	RMSEA
Hypothesised four-factor model (WBHRM, job stress, violence, and quality of care)	2213.07***	863		.80	.80	.09
Three-factor model: (violence, WBHRM, as separate latent variables, with job stress and quality of care loaded on one latent variable)	2663.41***	865		.65	.65	.11
Latent variable CMV test	1798.61***	832	367.25***	.80	.80	.13

***p < 0.001.

Figure 2 illustrates the significant workplace violence x HRM interaction term (Step 3 of Table 5). Following the procedure outline by Aiken et al. (1991), we tested the significance of the simple slopes. As WBHRM increases, the impact violence has on stress decreases, with a negative significant relationship. This result is consistent with the hypothesis. Therefore, Hypothesis 3 is supported.

Study 2 – China

Method

Following a similar approach to the Australian study, three waves of surveys were sent to doctors and nurses working at two hospitals in China. The survey questionnaire was translated into Chinese from English by the bilingual members of the team and two independent researchers, using the back-translation procedure (Brislin, 1970). Paper-based questionnaires were distributed to doctors and nurses working at two Chinese hospitals. Prospective respondents were assured that their responses were anonymous. A total of 136 questionnaires were returned, yielding 68% response rate. Of the 136 respondents, approximately 85.3% were female and 11% male. The majority (94.1%) worked full time, followed by 5.9% part-time/job sharing. Approximately 51.5% were nurses and 48.5% were doctors. The majority of respondents (66.9%) worked in public hospitals, followed by private hospitals (22.8%), and non-for profit (10.3%).

Measures for Study 2

The same measures of workplace violence, job stress, quality of care, and WBHRM practices from the Australian study (Study 1) were used in the Chinese study. Workplace violence was collected in Time 1, job stress and WBHRM practices were collected in Time 2, and quality of care was collected in Time 3.

Results for Study 2

Mirroring the approach taken with the first study (Australian sample), we conducted a CFA to verify the factor structure of the variables in the study.

The CFA results indicated that each model was significantly different to each other, however, given our theoretical framework, the four-factor model was chosen due to its acceptable fit measures.

Table 6 presents the descriptive statistics for the Chinese sample.

Table 6. Descriptive statistics in Study 2 (Chinese sample).

	Mean	SD	1	2	3	4
1. Workplace violence	1.45	0.54	(0.92)
2. Job stress	2.54	0.80	0.43***	(0.90)
3. Quality of care	4.02	0.50	−0.10	−0.13*	(0.80)
4. WBHRM	3.41	0.60	0.25*	−0.15*	0.35***	(0.87)

N = 123. ^†p < .10, *p < .05, **p < .01, ***p < .001.

***Significant at the 0.001.

Hypothesis 1 predicted that workplace violence predicts stress. Results of the OLS regression used to test this hypothesis are presented in Table 7.

Table 7. Regression analysis: Workplace violence predicting job stress in Study 1 (H1 – Chinese sample).

Variables	b	s.e.	β
Constant	1.61**	0.17
Workplace violence	0.64***	0.11	0.43
R^2	0.19***
ΔR^2	0.19***

N = 254, *p < 0.05, **p < 0.01, ***p < 0.001.

Table 8. Moderated regression analysis: WBHRM moderating the relationship between workplace violence and job stress in Study 2 (Chinese sample).

Variables	Step 1			Step 2			Step 3
	b	s.e.	β	b	s.e.	β	b	s.e.	β
Constant	1.61	.17		2.62	.47		5.23	1.14
Workplace violence	.64***	.11	.43	.61***	.11	.42	−1.22^†	.05	−0.27
Well-being oriented HRM (WBHRM) practices				−0.28*	.17	−0.17	−1.01**	.31	−0.61
Workplace violence x WBHRM							.51**	.20	1.00
R^2	.19***			.22*			.25*
ΔR^2	.19***			.03*			.03*

N = 123, ^†p < .10, *p < .05, **p < .01, ***p < .001.

Hypothesis 2 predicted that job stress mediates the relationship between workplace violence and quality of care. Figure 3 presents the results of the mediated regression used to test this hypothesis.

Figure 3. Job stress mediates the relationship between workplace violence and quality of care in Study 2 (H2 – Chinese sample).

Figure 3 shows the direct effect of workplace violence on stress was statistically significant (b=.64, p<.001), indicating that doctors and nurses scoring higher on violence are more likely to experience job stress than those scoring lower on the measure. The direct effect of job stress on quality of care was statistically significant (b=−0.06, p<.001), indicating that doctors and nurses scoring higher on job stress are more likely to lower their quality of care than those scoring lower on the measure. The indirect effect was tested using bootstrapping procedures. The indirect effect of workplace violence via stress was statistically significant: .01, 95% CI= (−0.10 to −0.08), indicating that the mediational effect occurred in which workplace violence influences quality of care through job stress. Therefore, hypothesis 2 is supported.

Hypothesis 3 postulated that WBHRM practices moderate the relationship between workplace violence and job stress. Table 6 presents the results of the moderated regression.

Table 8 shows the statistical analysis of study 2.

Step 2 of Table 6 introduces the moderator (WBHRM) and Step 3 introduces the workplace violence x HRM interaction term. The interaction term is significant (b =.51, s.e. =.20 p <.05). Figure 3 presents the interaction between workplace violence and HRM.

Figure 4 illustrates the significant workplace violence x HRM interaction term (Step 3 of Table 5). Following the procedure outline by Aiken et al. (1991), we tested the significance of the simple slopes. As WBHRM increases, the impact workplace violence has on job stress decreases, with a negative significant relationship. This result is consistent with the hypothesis. Therefore, Hypothesis 3 is supported in the Chinese sample.

Figure 4. WBHRM moderating violence-job stress relationship (Chinese sample).

Discussion

This study examined the role of WBHRM practices in managing job stress of doctors, nurses and PCAs in the context of Australian and Chinese healthcare organisations. To do this, we tested three hypotheses underpinned by COR theory (Hobfoll, 2001). All hypotheses were supported in both Study 1 and Study 2. By testing our hypotheses using various clinician groups across two different cultural and healthcare settings, we provide further validation of our WBHRM practices measure. First, based on Hypothesis 1, we found that workplace violence was positively related to job stress. When doctors, nurses and PCAs experience incidents of violence, a negative emotional state (job stress) can occur as a response to adverse work events, as this places significant demands on job resources (Hobfoll, 2001). These results extend the findings of previous research on the adverse effects of workplace violence (Itzhaki et al., 2018; Kobayashi et al., 2020; Pariona-Cabrera et al., 2023). The results also provide further evidence that workplace violence has detrimental consequences for healthcare workers’ mental health. Second, consistent with Hypothesis 2, we found that job stress mediated the relationship between workplace violence and quality of care. The presence of job stress reduces doctors, nurses’ and PCAs’ quality of care after they experience workplace violence. Our results suggest mediational effects of job stress on workplace violence and quality of care relationships (Acquadro et al., 2018). These results reinforce the importance of the management of the effects of violence on job stress to mitigate harmful consequences, which can negatively affect perception of quality of care.

Third, consistent with Hypothesis 3, we found that WBHRM practices moderated the relationship between workplace violence and job stress. HRM practices as conditional/organisational resources (COR theory) play a critical role in managing the effects of workplace violence by combining and accumulating various HRM practices as job resources to create resource caravans (Salanova et al., 2010). HRM practices act as a vehicle to buffer the negative impact of workplace violence on doctors, nurses and PCAs’ mental health. When healthcare workers have access to job resources such as training and development, job quality, occupational health and safety, participation in decision making and work-life balance (Guest, 2017), they can better manage the negative impact of workplace violence on their mental health. Consistent with our reasoning, positive perceptions of doctors, nurses and PCAs should increase with the implementation of WBHRM practices, thus reducing feelings of job stress. As WBHRM practices are employee centred (Guest, 2017), they act as job resources providing essential support to healthcare workers, which can positively impact on their emotional state (Alvaro et al., 2010). Importantly, these findings reinforce our approach in which WBHRM practices contribute to supporting healthcare workers to cope with their negative emotions (job stress), arising from workplace violence. In doing so, when nurses and PCAs recognise caring behaviours from managers through the implementation of these HRM practices, they perceive the employment relationship as favourable towards their mental health and organisational performance (Van De Voorde et al., 2012).

Contributions to theory

This study makes two contributions to HRM literature. First, the study contributes to the HRM literature by conceptualising WBHRM as an organisational resource underpinned by COR theory (Hobfoll, 2011). WBHRM practices can act as a buffer against the negative impact of violence on job stress. HRM practices decrease the negative (depleting) effects of job stress by protecting against resource loss, recovering from loss, and potentially gaining additional resources (Bakker & Demerouti, 2007). WBHRM practices provide an advanced approach to a greater theoretical certainty about the processes by which accumulation and protection of wellbeing resources will support employees and curtail the negative effects of job stress. For example, when organisations implement WBHRM practices, employees can enhance adaptation and coping skills and consequently improve their wellbeing (Salanova et al., 2010). Through creating resource gain, WBHRM resources help to protect employees against feelings of job stress. WBHRM bundles add to resource caravans, alleviate resource drains and consequently increase employees’ performance (Hobfoll, 2011). Therefore, by having access to a pool of resources (conditional/organisational resources COR theory), employees restore threatened resources and increase their sense of control to manage adverse work events such as workplace violence and undesirable outcomes such as low quality of care.

Second, this study unpacks the process through which the five sets (bundles) of WBHRM practices buffer the negative emotional effects of violence on employees’ wellbeing. This provides a more comprehensive approach to what Guest (2017, p. 32) refers to as ‘a well-rounded ‘diet’ of HRM’. The first set of WBHRM practices training and development, fosters the enhancement of learning and enthusiasm, which are important elements of building resources. Employees will acquire and create more resources through the process of growing and developing further skills, abilities and competencies, which subsequently may help in decreasing job stress, as they develop deeper resource caravans (Hobfoll, 1989). The second set on job quality, promotes role clarity and autonomy, as well as supports a variety of tasks, opportunities, and job engagement. This allows employees to increase feelings of self-efficacy and belief in their ability to cope with the effects of negative work events and promote an effective course of action. Consequently, this provides resources to protect employees against feelings of job stress. The third set on occupational health and safety, contributes to having a safe work environment in which employees feel protected and supported by their employers. When organisations provide safe work conditions, employees may experience less stress (Shea et al., 2018). The fourth set on participation in decision making, promotes individuals voice, opinions and new ideas. When organisations support a positive and open dialogue, employees are motivated to actively find opportunities to share and learn new approaches. This represents important resource caravans for employees to increase their individual resources in response to resource drain such as job stress (Hobfoll, 2011). The final set on work-life balance, promotes shared goals and mutual respect, which enhances collaboration and opportunities for employee’s growth and development. By having perceptions of fairness and transparency during a consultation process, employees can replenish and/or maintain mental and physical resources, influencing their wellbeing. Therefore, WBHRM practices delivered as resource bundles or conditional/organisational resources (COR theory) will provide employees with the necessary tools to strengthen their emotional regulation competencies and consequently coping strategies to manage and mitigate job stress.

Practical implications

Findings from our study have practical implications for key stakeholders in the healthcare sector. In aged care facilities and hospitals, the recurrent incidents of violence negatively impact doctors, nurses and PCAs and increase their job stress and consequently decrease their in-role performance (quality of care). Hence, it would be beneficial for these organisations to focus on the prevention of workplace violence, and mitigation of the negative effects of violence at work (Pariona-Cabrera et al., 2022). First, we suggest that aged care facilities and hospitals implement WBHRM practices to manage job stress Organisations should adopt WBHRM practices as a systematic approach across both healthcare settings. This can be promoted through knowledge sharing and staff meetings of aged care and hospital workers, including the development of virtual/online education resources. Direct managers should frequently engage with floor staff through regular meetings, feedback sessions and one-on-one and group discussions that facilitate open communication to gain a better understanding of the management of the negative emotional effects of workplace violence.

Second, our findings underscore the important role of aged care and hospital executive/leadership in the implementation of WBHRM practices. Senior management has formal power to influence and implement WBHRM practices that support employees’ wellbeing. Senior management should develop appropriate HRM structures, policies and procedures, and promote effective communication between managers and HRM departments. This can help in implementing formalised WBHRM practices that enhance doctors, nurses and PCAs’ wellbeing and subsequently drive positive in-role performance such as quality of care.

Finally, we recommend that managers implement WBHRM practices to better support doctors, nurses and PCAs. Managers should promote high levels of training and development in managing negative emotions at work. This may include workshop sessions combined with complementary practices such as feedback, varied content and duration (Baby et al., 2018). Similarly, managers should promote job quality through opportunities for control, skills use and variety at work. This promotes a degree of autonomy which emphasises how employees are valued by their managers (Zacharatos et al., 2005). Moreover, managers should focus on the provision of safe work conditions. A caring environment contributes to helping employees to better cope with job stress (Er & Sökmen, 2018). Employee participation in decision-making promotes open communication at work as healthcare workers share their ideas and views, which in turn, potentially influences their wellbeing. Finally, work-life balance initiatives provide employees the opportunity to self-manage their work schedule, enhancing their mental health. Therefore, WBHRM practices should be promoted as a bundle to enhance employees’ wellbeing at work and positively impact employment relationships.

Limitations and directions for future research

First, we investigated the role of WBHRM practices to mitigate the effects of workplace violence through quantitative data. Future research could include qualitative studies that provide voice to participants and gain deeper insights into other ways to manage and mitigate the emotional effects of violence. Also, future studies may include withdrawal behaviours after health care workers experience adverse work situations such as workplace violence. Second, data were collected from nurses and PCA members of the ANMF and employed in aged care facilities across the State of Victoria, as well as doctors and nurses from two Chinese hospitals. Generalisation of the findings may be limited to the healthcare industry and other states in Australia and China that have similar characteristics. Future studies could examine the effects of violence on healthcare workers’ wellbeing in aged care facilities and hospitals across all Australian states and territories and other countries. Third, sample size of both studies, especially the Chinese sample were relatively small. However, the samples of the two studies were satisfactory to conduct the factor analysis and test moderation and mediation effects (Aguinis, 2004). However, it would be important to consider larger sample sizes, in different healthcare organisations across different contexts to enhance generalisability of findings (Field, 2013).

Conclusion

Our paper provides clear evidence that workplace violence negatively affects healthcare workers’ job stress and quality of care in both Australia and China. Findings enhance understandings of the role of WBHRM practices to manage and mitigate workplace violence and its impact on employees’ wellbeing across healthcare organisations. Through testing a new measure of WBHRM practices, we demonstrated the efficacy of this new measure to buffer the effects of workplace violence on employees’ job stress in Australian aged care and Chinese and hospital facilities. We encourage further research that adopts this new measure of WBHRM practices to examine its efficacy in mitigating various forms of workplace aggression, violence and other workplace stressors on employees and work teams in diverse organisational and industry settings.

Acknowledgments

The authors wish to thank all participants (e.g., registered nurses, enrolled nurses, endorsed enrolled nurses, personal care workers, the OHS coordinator, and the OHS officer) in this study; without their contribution, it would not have been possible to undertake the research.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

The datasets generated during the current study are available from the corresponding author on reasonable request.

Additional information

Funding

There has been no funding support for the research paper.

Appendix A

Well-being HRM measure items

Dimension 1: investing in employees

1. Only the best people are hired to work in my unit

2. Employees of my unit are involved in the hiring of their peers

3. My immediate manager encourages me to extend my abilities

4. My unit has provided me with training opportunities enabling me to extend my range of skills and abilities

5. Promotions are regular

6. There is very good opportunity for advancement

7. There is a good chance to get ahead

Dimension 2: providing engaging work

1. I have lots of opportunity to decide how to do my work

2. If a problem emerges within my work I can take action to remedy it

3. I have enough information to do my job well

4. It is easy for me to communicate my thoughts to management

5. I always feel safe working in my Unit in these conditions

Dimension 3: positive social and physical environment

1. My unit does what it can to ensure the well-being of its employees

2. My unit spends enough resources on health and safety-related matters

3. My unit spends enough resources on EEO awareness and EEO related training

4. My unit supports employees with the balancing of work and family responsibilities

5. Management in my unit are supportive of cultural differences in this organization

6. Men and women have the same employment opportunities in my unit

7. EEO is promoted in my unit

8. There is a strong link between how well I perform my job and the likelihood of my receiving recognition and praise

9. There is a strong link between how well I perform in my job and the likelihood of receiving a raise in pay/salary

10. There is a strong link between how well I perform my job and the likelihood of my receiving high performance appraisal ratings

11. There is a strong link between how my team performs and the likelihood of my receiving a raise in pay/salary

Dimension 4: promoting employee voice

1. My unit tries to find out how employees are feeling

2. My unit measures how its managers are performing

3. I have significant say in decision making at work

4. I participate in decision making with regard to work-related issues

5. I participate in decision making with regard to the nature of my work activities

Dimension 5: promoting organizational support

1. My unit values measures to promote employee work-life balance

2. My unit supports employees in balancing their professional and private lives

3. My unit’s management sets a good example of work-life balance

4. My unit’s management is trained to promote the work-life balance of employees

5. In my unit, employees are informed about programs promoting work-life balance

6. In my unit there is a performance management system to ensure that staff are competent and accountable for their work

7. In my unit there is a performance management system to ensure that future growth and development needs are identified