Search strategy

The electronic databases Cumulative Index to Nursing and Allied Health Literature (CINAHL), PsycInfo and PubMed were searched between 15 January 2022 and 21 January 2022. Reference lists of identified articles were also searched manually. Moher et al’s (2015) Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) approach was applied to the search.

The formulation of the review question was guided by Stern et al’s (2014) PICOS framework as follows:

The review question formulated was: ‘How do mental health nurses make sense of their experience of being assaulted by patients in inpatient settings according to reports of qualitative research?’.

Search terms were finalised after multiple test searches varying the use of terms, truncations (assault*; nurs*) and Boolean operators (AND, OR) and a review of keywords used across a sample of the retrieved literature. The search terms used are summarised in Table 1.

Table 1.

Search terms

Inclusion and exclusion criteria

The inclusion and exclusion criteria are summarised in Table 2. An earlier scoping review had indicated that there were few qualitative studies focusing on mental health nurses’ personal experiences of assault by patients in inpatient settings, so no limiters regarding publication dates were placed on this review.

Study selection

A total of 10,646 studies were retrieved from the database searches, with a further 19 articles obtained through hand-searching of reference lists. Following the removal of duplicates, 7,522 articles were screened by title and, where it was unclear whether or not to exclude an article based on its title alone, abstracts were read. Of the remaining 728 articles, 699 full texts were screened as 29 full texts were not obtainable. The selection process resulted in 16 articles for inclusion in the review.

Critical appraisal

The authors independently appraised the selected articles using the JBI Critical Appraisal Checklist for Qualitative Research (Lockwood et al 2015). The majority of articles (n=10) lacked clarity regarding their philosophical perspective – for example the ontological and epistemological assumptions underpinning their studies. Most of the articles (n=10) did not situate researchers culturally or theoretically, nor did they address the influence of researchers on the research and vice versa.

Data extraction

Findings relevant to the review question were identified and extracted from each article. The findings were reviewed with reference to their sources and assigned one of three levels of credibility (‘unequivocal’, ‘credible’ and ‘not supported’), as per JBI criteria (Lockwood et al 2020). ‘Unequivocal’ and ‘credible’ findings were included but ‘not supported’ findings were not included.

Perspectives on violence against mental health nurses

Violence and assaults on mental health nurses were considered to be frequent, inevitable, expected and ‘normal’ – particularly in the case of verbal violence, which was believed to be ‘part of the job’ (Currid 2008, Baby et al 2014). Despite these widely expressed views, assaults were often described as unexpected, unpredictable and unpreventable (Tema et al 2011, Yang et al 2016), resulting in nurses feeling shocked and confused (Benson et al 2003, Sim et al 2020).

The consensus view among mental health nurses was that while violence and assaults are frequently experienced, they should not be considered acceptable or ‘part of the job’ (Baby et al 2014, Stevenson et al 2015). There was a perception that the violence they experienced was not always taken seriously by their managers and the police (Baby et al 2014, Hiebert et al 2021), and nurses did not consider themselves to be protected by the law (Dean et al 2021). A perception that, in contrast, assaults on police officers and firefighters were taken seriously and responded to with compassion and support led some nurses to feel a sense of unfairness (Moylan and Cullinan 2011, Moylan et al 2014).

Conceptualisations of what constitutes violence in the mental health inpatient context varied between studies. One study included the explicit finding that such conceptualisations are subjective and influenced by multiple factors (Cutcliffe 1999). For example, if an act was believed to be intentional, deliberate and/or premeditated, nurses were more likely to define it as violence or an assault than if an act was believed to be impulsive, reactive and/or driven by symptoms of mental illness (Cutcliffe 1999, Zuzelo et al 2012). Some nurses linked this intentionality and perception of control to a diagnosis of personality disorder, which implied responsibility and prompted blame (Benson et al 2003, Stevenson et al 2015). When a patient was believed to be mentally ill, for example if they were experiencing psychosis, their behaviour was more likely to be considered unintentional, beyond their control and not conceptualised as violence (Cutcliffe 1999, Hiebert et al 2021).

Personal and professional impact

Being assaulted had the potential to have significant and pervasive effects on mental health nurses’ personal and professional lives. The effect in terms of physical injury was not a prominent theme in the included studies, with nurses stating explicitly that psychological trauma often had a more lasting effect than any physical effects (Yang et al 2016). Nurses reported that their self-esteem, confidence and sense of competency were negatively affected following an assault, with nurses blaming themselves on several occasions (Baby et al 2014, Hiebert et al 2021). The emotional effects were consistently described as being pervasive, with fear, anxiety, frustration, anger, guilt and shame being the most frequently reported (Stevenson et al 2015, Dean et al 2021, Ezeobele et al 2021, Hiebert et al 2021).

Mental health nurses’ sense of agency was also negatively affected, with powerlessness and helplessness recognised as consequences of assault (Sim et al 2020, Ezeobele et al 2021). Nurses experienced a loss of dignity, feeling ‘small’ in the face of verbal abuse (Tema et al 2011) and ‘belittled’ when verbal abuse was sexualised (Moylan et al 2014). One finding was that nurses could become desensitised to violence due to its frequency, ‘except the severe stuff’ (Hiebert et al 2021).

The effects of assaults experienced by male mental health nurses featured in the findings of six of the studies (Benson et al 2003, Tema et al 2011, Zuzelo et al 2012, Moylan et al 2014, Dafny and Beccaria 2020, Sim et al 2020). Male nurses were found to be less likely to express emotions than female nurses, instead ‘downplaying’ them (Benson et al 2003). They often saw themselves, and were seen by their peers, as protective factors against assaults on female colleagues (Zuzelo et al 2012, Dafny and Beccaria 2020). However, many male nurses felt responsible, emotionally drained, fearful of injury and undermined professionally when they were seen as ‘bodyguards’ (Tema et al 2011, Dafny and Beccaria 2020).

Many mental health nurses who had been assaulted felt less able to be empathic, compassionate and person-centred (Cutcliffe 1999, Stevenson et al 2015), often distancing themselves and becoming increasingly task-focused as a means of coping and protecting themselves from a further assault (Currid 2008, Ezeobele et al 2021). This, together with a loss of trust and sense of betrayal, made establishing and maintaining therapeutic relationships with patients increasingly challenging (Kindy et al 2005, Sim et al 2020).

As a result of being assaulted, some mental health nurses experienced a conflict in relation to their role. For example, one nurse spoke of ‘[not feeling] like a nurse’ after an assault (Kindy et al 2005), while others found it challenging to reconcile their feelings of anger and frustration with their caring role (Tema et al 2011).

Effects of being assaulted on mental health nurses’ lives outside of work included disturbed sleep, nightmares, hypervigilance and increased smoking and alcohol consumption (Stevenson et al 2015, Hiebert et al 2021). Nurses’ personal relationships were often negatively affected, with nurses describing themselves as being ‘snappy’ and ‘cranky’ with their family members (Kindy et al 2005, Baby et al 2014) and shouting at their children ‘for nothing’ following an assault (Tema et al 2011).

Response to being assaulted

Mental health nurses frequently reported attempting to cope with the emotional effects of being assaulted by suppressing their feelings and withdrawing from others (Kindy et al 2005, Zuzelo et al 2012). The suppression of emotions was also described as a protective strategy, with many nurses believing that showing emotion in front of patients was risky and would make them feel vulnerable (Zuzelo et al 2012, Lantta et al 2016). Phrases such as ‘moving on’, ‘get past it’, ‘leave work at the door’ (Zuzelo et al 2012) and ‘get back on the bike’ (Baby et al 2014) illustrated nurses’ perspectives on what they believed was important or required following an assault.

Many mental health nurses made sense of their experience by seeking to understand and rationalise patients’ assaultive behaviour. Not taking the assault personally was seen as important in being better able to cope, forgive and move on (Benson et al 2003, Sim et al 2020, Dean et al 2021).

The venting of emotions was referred to as helpful by mental health nurses, but this was typically something done in private; for example, one nurse stated: ‘Nobody knows I’m in there crying. I will vent and let go of stuff’ (Zuzelo et al 2012). Becoming hypervigilant was described as a response to being assaulted, such a response being driven by fear and anxiety (Kindy et al 2005, Yang et al 2016). Being assaulted also prompted some nurses to consider leaving their job or the profession altogether (Ezeobele et al 2021, Hiebert et al 2021).

Learning and behavioural changes were identified by nurses as positive outcomes of reflecting on their experiences of being assaulted (Baby et al 2014, Yang et al 2016), but this was not mentioned frequently or explored in any depth.

Sharing and reporting experiences

Mental health nurses often found it challenging to share and report their experiences of being assaulted and many of them avoided doing so. Reasons why they would avoid sharing their experiences of assault with family members included a desire to protect them and the belief that their families – and society more generally – would be unable to understand and empathise (Sim et al 2020), and also might not believe them (Dafny and Beccaria 2020).

Overwhelmingly, mental health nurses who had experienced assault did not feel supported by their employers and perceived their managers’ responses as critical, invalidating, blaming, punishing and stigmatising (Ezeobele et al 2021, Hiebert et al 2021). For example, one nurse described the people who reviewed incident reports as being ‘only concerned with what the nurse did or didn’t do’ (Hiebert et al 2021), while another believed that they had been seen as ‘the perpetrator of the assault and not as a victim’ (Ezeobele et al 2021). Other nurses mentioned critical comments such as ‘You are responsible for maintaining safety, can’t you do that?’ (Moylan et al 2014).

The absence of a response and/or action following an assault was a common experience, leaving mental health nurses feeling marginalised and ignored (Tema et al 2011, Moylan et al 2014). There was a perceived lack of resolution following an assault (Kindy et al 2005), including patients not being held to account for their actions (Dean et al 2021). Many nurses perceived that others expected them to carry on and continue to care for the person who had assaulted them (Stevenson et al 2015, Yang et al 2016). Nurses felt that personal, direct support that recognised and acknowledged their experiences was important and required, as was being given time to access such support (Stevenson et al 2015, Dean et al 2021).

Findings relating to the reporting of assaults highlighted several explanations for under-reporting. Mental health nurses were less likely to report an assault if they did not consider it to be serious enough (Hiebert et al 2021), since they believed that others expected them to ‘let it go’ and were afraid of being seen as ‘making a big deal’ of their experience (Moylan et al 2014). Nurses also had mixed experiences in relation to support from the police, perceiving that they were discouraged from reporting the incident on some occasions (Stevenson et al 2015) and believing that ‘nothing will be done’ (Dafny and Beccaria 2020). A lack of trust in managers, feelings of shame and fear of blame, criticism and/or stigmatisation made many nurses reluctant to report assaults (Dean et al 2021, Ezeobele et al 2021). Other reasons why nurses did not report assaults were a lack of time and the cumbersome forms involved (Hiebert et al 2021), as well as nurses not recognising the importance of reporting (Dean et al 2021).

Factors affecting violence and assault

Mental health nurses’ perceptions of what contributed to, and could prevent, violence and assault centred on factors relating to the environment, workforce, relationships, restrictive practices and gender.

Nurses mentioned various factors such as insufficient staffing, inadequate training and levels of experience, suboptimal physical security, and lack of space and privacy in the work environment (Kindy et al 2005, Hiebert et al 2021). Developing effective therapeutic relationships and having sufficient time to spend with patients were seen as factors that could reduce the risk of violence and assaults (Lantta et al 2016, Ezeobele et al 2021); however, these factors were also perceived as being compromised by high workloads and excessive paperwork (Kindy et al 2005).

To minimise the risk of mental health nurses being targeted for violence by patients, effective and supportive teamwork based on a shared and consistent approach that reinforces team cohesion was widely seen as important (Yang et al 2016, Dean et al 2021). Nurses also perceived that the failure of teams to hold patients to account and agree on consequences for them contributed to the ongoing issue of violence in the workplace (Dean et al 2021). Patients’ experiences of being restricted, treated against their will and not having their needs met were considered to be precursors to assaults (Currid 2008), as were reductions in medicines (Kindy et al 2005).

Finally, mental health nurses discussed gender as a factor that can influence the risk of assaults. Male nurses were perceived as being assaulted more frequently by male patients (Dafny and Beccaria 2020), while female nurses were viewed as more vulnerable, particularly to verbal abuse (Zuzelo et al 2012, Sim et al 2020) and when few male nurses were present (Tema et al 2011).

Limitations

While the review team collaborated in the final selection of articles, critical appraisal and review of findings, the review was largely undertaken by the primary author (HA), thus increasing the risk of bias. Articles not written in English were excluded, which may have resulted in relevant findings being missed. The number of databases and reference lists searched was limited by time and resources, which may have further limited the findings of the review.