NHS Staff Experiences of Violence: Literature Review

NHS Inpatient Staff Experiences of Violence: A Systematic Review of the Literature



For decades there has been an interest in the link between mental health and violence. Research has shown that, within psychiatric populations, violence is more likely to occur if the patient has a history of violence, is impulsive or hostile, has been admitted to hospital involuntarily, or has been in hospital for a considerable length of time. Verbal aggression is the most common form of abuse that healthcare staff are exposed to, followed by physical assault. Management techniques for violence include seclusion, restraint, medication, de-escalation and communication. Globally, there is some evidence that exposure to aggression can lead to high levels of burnout, stress and even symptoms of Post-Traumatic Stress Disorder (PTSD) in healthcare staff. In addition, it is suggested that the staff-patient interaction is central to the development of violence in inpatient units, and that violence can lead to negative interactions and a hostile ward environment. This in turn can lead to staff burnout, and poor quality of patient care. This systematic review of the literature aims to explore the experience of violence on National Health Service (NHS) inpatient staff, and to establish staff experience of post-incident support.


A search of the literature focusing on staff experience of violence identified 5080 studies. Following the removal of duplicates and papers that did not meet the inclusion criteria, eight studies were included in the review and subject to quality assessment.


Studies varied in terms of quality and design, and were a mixture of qualitative and quantitative research. Four themes emerged from the studies: Distress and Emotional Impact, Work Environment and Competence, Therapeutic Relationship and Experience of Support. Overall, staff exposed to violence reported a variety of feelings, including guilt, anger and shame, feeling less competent in their work and viewing their work environment as dangerous, wanting to withdraw from spending time with the patients as a result of a ruptured therapeutic relationship, and finding that support was inconsistent and of poor quality.


This review highlights the importance of exploring staff experiences of violence, particularly as this seemed to have a negative impact both on the staff personally, and on the therapeutic relationship. Furthermore, staff experience of support needs to be explored in greater detail, as this is an area where staff reported feeling dissatisfied. More qualitative research needs to be conducted in this area, using a larger number of participants and ensuring validation methods are used to improve methodological rigour.


The link between violence and mental health has been an area of interest in research for decades, and as a result, there is a wealth of literature that explores how, why, and when violence is likely to occur, what form the violence is likely to take, and how best to manage violence when it occurs. The research suggests that, within psychiatric populations, violence is more likely to occur if the patient has a history of violence, is impulsive or hostile, is in hospital involuntarily, or has been in hospital for a considerable length of time (Cornaggia, Beghi, Pavone & Barale, 2011; Sanghani et al., 2017). Verbal aggression towards staff is the most common form of abuse, followed by physical assault (Kowalczuk & Krajewska-Kułak, 2017; Omerov, Edman & Wistedt, 2002). The management of aggression and violence differs across a variety of settings, though seclusion, restraint, medication, de-escalation and communication are often used as techniques to try to reduce the impact of violence (Jalil, Huber, Sixsmith & Dickens, 2017).

There is some evidence that, globally, exposure to violence and aggression from patients both in general hospital wards and specific psychiatric settings can lead to high levels of burnout, staff stress and in some cases, symptoms of Post-Traumatic Stress Disorder (PTSD) (Hilton, Ham & Dretzkat, 2017 & Nijman, Bowers, Oud & Jansen, 2005). These findings are important when considered in line with the proposed model of violence towards healthcare staff by Arnetz and Arnetz (2001). This model suggests that the patient-staff interaction is central to the development of violence in inpatient units, and that this interaction is influenced by the immediate ward environment, a proposal supported by Cornaggia et al. (2011). Arnetz and Arnetz (2001) suggested that violence from patients had a negative effect on staff, who would further develop negative attitudes towards their work tasks and the patients themselves, creating a negative environment. Furthermore, this negative environment causes staff to feel more on guard around the patients, spending less time with them and therefore not meeting the needs of the patients. This then causes the patients to feel frustrated, resulting in an increase in violence in an attempt to communicate with staff when all other forms of communication have failed. Arnetz and Arnetz (2001) conclude that this development of a negative ward environment and higher risk of violence leads to poor patient satisfaction with care, and an increase risk of staff burnout, issues highlighted in the aforementioned literature.

Given the high levels of burnout that are linked to violence and aggression, and the current economic climate, the systematic review aims to explore the experience of violence on National Health Service (NHS) inpatient staff.


  1. To establish the experience of violence on NHS inpatient staff, and what impact this might have on them.
  2. To establish whether NHS inpatient staff report any experience of post-incident support, and if so, their opinions on this.


Literature Search

A database search of Web of Science, Medline, CINHAL, psycINFO and psycARTICLES was conducted on 26th October 2017. The grey literature was also searched during this time, using OpenGrey. In addition, reference lists of articles included in the review were searched for any further articles that may be of relevance to the review. A search strategy identifying references pertaining to the impact of violence on inpatient staff was used. No date limits were set on the searches, to allow for maximum inclusion of potential articles. The full search strategy, and search results, are listed in Table 1.

Table 1 – References Identified Through Database Searching

Study Selection

First, titles were screened for eligibility. Titles were coded as either meeting, not meeting or potentially meeting inclusion criteria. Next, the abstracts of the titles identified as meeting or potentially meeting the inclusion criteria were screened. Abstracts were again coded as meeting, not meeting or potentially meeting inclusion criteria. Finally, full texts articles were accessed for the abstracts that either met or potentially met the inclusion criteria. Full text articles were coded as either meeting or not meeting the inclusion criteria.

Inclusion Criteria

  • Studies that include staff members working in inpatient mental health units
  • Studies conducted in the NHS
  • Qualitative studies where staff report their experience of violence
  • Quantitative studies using outcome measures exploring the experience of violence

Exclusion Criteria

  • Studies that focus solely on the type or frequency of violence
  • Studies focusing on staff or patient interventions for managing violence
  • Studies focusing on staff training
  • Studies conducted outside of the NHS
  • Studies conducted in community settings
  • Non-English texts

Quality of Studies

An appraisal tool of primary research studies from a variety of fields (Kmet, Lee & Cook, 2004) was used to quality assess the full texts selected for review. Criteria listed in this tool are scored as either not applicable, criteria not fulfilled, criteria partially fulfilled or criteria fulfilled (scoring -, 0, 1, or 2 respectively). The final score is calculated as the total sum/total possible sum, resulting in a score falling between 0 and 1.


A total of 5080 studies were identified from the database searches. 181 were removed due to being non-English texts, 1384 were removed as duplicates, and 3506 did not meet the inclusion criteria. The final review included eight studies (Figure 1).

Study Design and Quality

Studies included in the review varied in terms of quality, with ratings ranging from 0.56 – 0.94 (Tables 2-3). On the whole, qualitative studies (Table 2) displayed methodological strength in describing clear objectives, having an evident and appropriate study design, using analytic methods that are justified and appropriate, reporting results in sufficient detail and ensuring that the conclusions were supported by the results. The quantitative studies with the most methodological strength (Reininghaus, Craig, Gournay, Hopkinson & Carson, 2007; Wildgoose, Briscoe & Lloyd, 2003) gained higher scores for demonstrating an estimate of variance in their results, where others failed to do this. In addition, the paper by Reininghaus et al., (2007) was the only quantitative study that reported on the quality of the outcome measures used, discussing their internal consistency, validity and reliability, and ensuring that they were designed for use with the target population. Therefore, this paper demonstrated the highest methodological strength. Generally, quantitative studies displayed methodological weakness in describing participant selection and characteristics, using well defined and robust outcome measures and reporting an estimate of variance in the results. The paper with the lowest rating (Whittington & Wykes, 1992) and therefore which displayed the most methodological weakness, failed to report any estimate of variance in the results, and showed further lack of detail in study participant characteristics, and in relating the conclusions to the results. Furthermore, this study lacked detail in outlining the aims and objectives of the research and in describing the study design.

Figure 1 – Flow Chart of Study Selection

Quantitative studies (Table 3) showed methodological strength in, generally, having clear objectives and study designs, describing the context of the study and describing the data analysis in a systematic way. Methodological weakness came from connecting the research to a clear theoretical framework, describing the sampling strategy and data collection, using verification procedures to improve credibility and discussing reflexivity of the account. The highest scoring qualitative paper, and therefore the one which displayed the most methodological strength, used verification procedures that included triangulation, theoretical memoing and member-checking in order to reduce research bias and improve transparency in the final results (Jeffery & Fuller, 2016). Neither of the remaining two qualitative papers described using any validation methods in their papers, and so equally scored lower for methodological quality.

Study Characteristics

Details of the studies included in the review (Table 4) demonstrate a preference for quantitative surveys (five articles) over qualitative interviews (three articles). Of the eight studies included in the review, six focused on the emotional and psychological impact of violence on inpatient staff (Crabbe, Alexander, Klein, Walker & Sinclair, 2002; Currid, 2009; Jeffery & Fuller, 2016; O’Brien, Tariq, Ashraph & Howe, 2013; Reininghaus et al., 2007; Wildgoose et al., 2003;), one focused on the nature of the reactions expressed by staff (Whittington & Wykes, 1992) and one explored both the impact of the incident on the therapist and the effect this had on the therapeutic relationship (Jussab & Murphy, 2015). Due to the small number of studies included in the review, a narrative synthesis was used to analyse the results. Four common themes emerged from the main conclusions of the eight studies, and are discussed in more detail below.

Distress and Emotional Impact

Participants who had been exposed to violence were often left with an ‘emotional hangover’ following the incident (Jeffery & Fuller, 2016). Common emotions that were experienced following an incident included guilt, anger, fear, vulnerability, fatigue, irritability and emotional exhaustion (Crabbe et al., 2002; Jeffery & Fuller, 2016; Jussab & Murphy, 2015; O’Brien et al., 2013; Whittington & Wykes, 1992). The impact of these emotions varied, with some studies finding statistical significance for questionnaires exploring psychological distress, caseness and Post-Traumatic Stress Disorder (PTSD) symptoms (Reininghaus et al., 2007; Whittington & Wykes, 1992; Wildgoose et al., 2003). Furthermore, participants in some studies reported having to take time off from work as a result of the impact of the incident, ranging from two days to one month (O’Brien et al., 2013; Wildgoose et al., 2003).

Work Environment and Competence

Many studies identified that, following an incident, participants were concerned about their own competence in working with their clients, felt that they had been to blame for incidents occurring, and further feared that their colleagues would view them as being incompetent in their work (Currid et al., 2009; Jussab & Murphy, 2015). Experiences of violence also impacts on how staff perceive the ward environment, with some having concerns over their future safety (Jeffery & Fuller, 2016) and being significantly more likely to perceive their work environment as more dangerous (Reininghaus et al., 2007).

Therapeutic Relationship

Three of the eight studies identified that, as well as having a personal impact, violence can have a negative effect on the therapeutic relationship. Participants in these studies reported avoiding the client for several days after the incident, rejecting and withdrawing from the client due to feeling angry towards them, and not engaging with the client out of fear that they may be putting themselves at risk by attempting to engage (Currid, 2009; Jeffery & Fuller, 2016; Jussab & Murphy, 2015).

Experience of Support

Five of the eight articles included in this review explored the importance of support following an incident. The consistency, quality and usefulness of the support offered varied across studies, with some participants being given the opportunity to talk about the incident immediately, and others having to wait until the next shift (Whittington & Wykes, 1992). Support was seen as important in managing the emotional impact of the event, and those with an unsupportive manager were significantly more likely to experience psychological distress after the event (Jeffery & Fuller, 2016; Jussab & Murphy, 2015; Reininghaus et al., 2007). Furthermore, participants were not encouraged to spend some time away from the ward immediately after the incident, and advice to go to general hospital or to take some time off work was rarely given (Whittington & Wykes, 1992). Overall, these studies found that participants were not satisfied with the support they received, which at times led to participants feeling that their management, and the organisation itself, does not care for the staff (Crabbe et al., 2002, Jeffery & Fuller, 2016; Whittington & Wykes, 1992).

Table 2 – Quality Rating Scores for Quantitative Articles, in Chronological Order
Table 3 – Quality Rating Scores for Qualitative Articles, in Chronological Order
Table 4 – Summary of Articles Reviewed Exploring the Experience of Violence of NHS Inpatient Staff, in Chronological Order


Of the eight articles included in this review, five used quantitative methodologies and three used qualitative methodologies. Exploring staff experiences of violence was the primary aim of seven of the eight articles. Despite not being the primary aim of the study by Whittington and Wykes (1992), staff experiences of violence were still explored. Four common themes emerged across the eight articles, concerning distress and emotional impact, work environment and competence, the therapeutic relationship and the experience of support.

Individuals who had been exposed to violence reported feelings ranging from guilt and anger, to fear and emotional exhaustion (Crabbe et al., 2002; Jeffery & Fuller, 2016; Jussab & Murphy, 2015; O’Brien et al., 2013; Whittington & Wykes, 1992). Furthermore, some individuals reported symptoms pertaining to psychological distress, caseness and PTSD (Reininghaus et al., 2007; Whittington & Wykes, 1992; Wildgoose et al., 2003). This, in line with reports of some individuals taking time off from work as a result of the incident (O’Brien et al., 2013; Wildgoose et al., 2003) supports the suggestion that emotional distress and the experience of violence can lead to staff burnout and exhaustion (Hilton et al., 2017).

Some individuals who had experienced violence reported concerns over their feelings of competence, and were significantly more likely to perceive their work environment as being dangerous (Currid, 2009; Jeffery & Fuller, 2016; Jussab et al., 2015). Given the proposed model of violence by Arnetz and Arnetz (2001), which highlights the importance of the ward environment on staff-patient interactions, this is an important finding. Staff who feel that their work environment is more dangerous may withdraw from patients as they fear for their own safety, further adding to patient dissatisfaction and increasing the likelihood of violence as a means of communication.

In addition, three of the eight studies identified that the experience of violence can have a negative impact on the therapeutic relationship (Currid, 2009; Jeffery & Fuller, 2016; Jussab et al., 2015). This can further contribute to the risk of violence occurring, as patients feel that their needs are not being met, and resort to violence as a method of communication (Arnetz & Arnetz, 2001). This in turn can make the ward feel more dangerous, leading to further withdrawal, further violence and a greater risk of burnout and sickness.

Finally, five articles in this review explored the importance of support following an incident of violence. Despite staff feeling that support was important in allowing them to manage the emotional impact of the violence (Jeffery & Fuller, 2016; Jussab et al., 2015), the majority were unsatisfied with the support they received, leading to feelings of being uncared for by management and the organisation itself (Crabbe et al., 2002; Jeffery & Fuller, 2016; Whittington & Wykes, 1992). This can have negative implications on the NHS in the current economic climate, if those staff then choose to leave their place of work due to feeling unsupported.

The articles included in this review varied in their level of assessed quality, which indicates that the results should be interpreted with caution. Many quantitative studies did not describe participant selection and characteristics, use well defined and robust outcome measures or report an estimate of variance in the results, whilst two of the three qualitative studies did not report the use of member checking or triangulation methods, which would improve the validity and transparency of the reported results.

The critical appraisal tool by Kmet et al., (2004) was selected for use in this review as it can be applied to primary research from a variety of fields, allowing for quantitative and qualitative articles to be assessed using the same tool, giving a universal score between 0-1. This allows for comparisons in quality to be compared, as articles are scored and summed according to rules that apply to both quantitative and qualitative studies. However, there is no guidance for which scores would constitute a low, medium or high quality study, leaving interpretation of the scores subject to the individual examining them. Therefore, a study may be classified as high quality by one person, but only medium quality by another, and so on.

Whilst every attempt was made to search for and include all relevant articles, by using a variety of databases, searching the grey literature and hand searching reference lists, some papers may have been missed from the review. This is of particular importance where searching for NHS inpatient environments is concerned, as many studies failed to state if research was carried out in these particular settings, meaning that they would be excluded from the review at an early stage.

From the limited literature included in this review, there is emerging evidence that the experience of violence can have negative emotional impacts on the staff who are exposed to it, leading to feelings of incompetence, fear of the ward environment and a ruptured therapeutic relationship. Furthermore, support is seen as a highly important factor in the management of the emotional impact, but this is inconsistently offered, and staff are generally displeased with the level and quality of support they experience, which can lead to further emotional distress. However, the number of participants included in each study was small, meaning results should be interpreted with caution.

Future research needs to ensure that more detailed qualitative research is carried out in order to continue to explore inpatient staff experiences of violence in the NHS. This research should aim to include a larger number of participants that the studies included in this review, and should aim to address the methodological issues identified, particularly those concerning the validity and transparency of the results. In addition, the support offered to staff, the quality of this and the impact this has needs to be explored further, as it was not a primary aim in the literature included in this review, but emerging evidence has revealed its importance.


This review has demonstrated that there is a dearth of research exploring the experience of violence towards NHS inpatient staff. However, from the limited research that has attempted to explore this area, themes of emotional distress, concerns around competence and safety, and damage to the therapeutic relationship have emerged. Post-incident support has also been identified as being important for managing emotional distress, although this if often lacking and of poor quality when received. Future research should qualitatively explore the impact of violence on staff, and how support is viewed and received. Future research should also aim to address some of the methodological weaknesses in this area, in order to improve the quality of the results produced.


Arnetz, J.E., & Arnetz, B.B. (2001). Violence towards health care staff and possible effects on the quality of patients care. Social science and medicine, 52, 417-427.

Cornaggia, C.M., Beghi, M., Pavone, F., & Barale, F. (2011). Aggression in psychiatry wards: a systematic review. Psychiatry research, 189, 10-20.

Crabbe, J., Alexander, D.A., Klein, S., Walker, S., & Sinclair, J. (2002). Dealing with violent and aggressive patients: at what cost to nurses? Irish Journal of Psychological Medicine, 19(4), 121-124.

Currid, T. (2009). Experiences of stress among nurses in acute mental health settings. Nursing Standard, 23(44), 40-46.

Hilton, N.Z., Ham, E., & Dretzkat, A. Psychiatric Hospital Workers’ Exposure to Disturbing Patient Behavior and Its Relation to Post-Traumatic Stress Disorder Symptoms. The Canadian journal of nursing research, 49(3), 118-126.

Jalil, R., Huber, J.W., Sixsmith, J., & Dickens, G.L. (2017). Mental health nurses’ emotions, exposure to patient aggression, attitudes to and use of coercive measures: Cross sectional questionnaire survey. International journal of nursing studies, 75, 130-138.

Jeffery, D., & Fuller, P. (2016). Witnessing violence: what are the experiences of mental health nurses? Mental Health Practice, 20(2), 15-21.

Jussab, F., & Murphy, H. (2015). “I Just Can’t, I Am Frightened for My Safety, I Don’t Know How to Work With Her”: Practitioners’ Experiences of Client Violence and Recommendations for Future Practice. Professional Psychology: Research and Practice, 46(4), 487-497.

Kmet, L.M., Lee, R.C., & Cook L.S. (2004). Standard quality assessment criteria for evaluating primary research papers from a variety of fields. Retrieved 1st December 2017 from Canada: Alberta Heritage Foundation for Medical Research: https://www.biomedcentral.com/content/supplementary/1471-2393-14-52-s2.pdf

Kowalczuk, K., & Krajewska-Kułak, E. (2017). Patient aggression towards different professional groups of healthcare workers. Annals of agricultural and environmental medicine, 24(1), 113-116.

Nijman, H., Bowers, L., Oud, N., & Jansen, G. (2005). Psychiatric nurses’ experiences with inpatient aggression. Aggressive behavior, 3, 217-227

O’Brien, A., Tariq, S., Ashraph, M., & Howe, A. (2013). A staff self-reported retrospective survey of assaults on a psychiatric intensive care ward and attitudes towards assaults. Journal of Psychiatric Intensive Care, 10(2), 93-99.

Omerov, M., Edman, G., & Wistedt, B. (2002). Incidents of violence in psychiatric inpatient care. Nordic journal of psychiatry, 56(3), 207-213.

Reininghaus, U., Craig, T., Gournay, K., Hopkinson, P., & Carson, J. (2007). The High Secure Psychiatric Hospitals’ Nursing Staff Stress Survey 3: Identifying stress resistance resources in the stress process of physical assault. Personality and Individual Differences, 42, 397-408.

Sanghani, S.N., Marsh, A.N., John, M., Soman, A., Lopez, L.V., & Russ, M.J. (2017). Characteristics of Patients Involved in Physical Assault in an Acute Inpatient Psychiatric Setting. Journal of Psychiatric Practice, 23, 260– 269.

Whittington, R., & Wykes, T. (1992). Staff strain and social support in a psychiatric hospital following assault by a patient. Journal of Advanced Nursing, 17, 480-486.

Wildgoose, J., Briscoe, M., & Lloyd, K. (2003). Psychological and emotional problems in staff following assaults by patients. Psychiatric Bulletin, 27, 295-297.

Cite This Work

To export a reference to this article please select a referencing stye below:

Related Services

View all 

Female student working on a laptop

Literature Review Service
From £124
Male student reading book

Dissertation Writing Service
From £136
Female student reading and using laptop to study

Leave a Reply