Workplace violence (WPV) against healthcare workers (HCWs) is a notable and widespread healthcare issue with a high global prevalence (1). WPV is an umbrella term that is defined by the World Health Organization (WHO) as “incidents involving work-related abuse, threats or assaults among HCWs including physical, sexual, verbal, psychological abuse and workplace harassment.” (1) The World Medical Association (WMA) has condemned the increase in cases of WPV perpetrated against HCWs and has labeled the surge in violence over the last decade as a healthcare emergency. Further, the WMA has called on the WHO and member states to take urgent action against WPV (2). The WHO has similarly acknowledged the impact of WPV on the physical and psychological health of HCWs and their subsequent job motivation, potentially compromising patient care and putting healthcare provision at risk (3).

Liu et al. recently quantified rates of WPV in healthcare and concluded that 62% of HCWs have been exposed to WPV. Liu et al. further noted that violence varied by geographic location and hospital department, with an increased prevalence in Asia and North America, and in the Emergency Department (ED) and psychiatric wards (4). Moreover, in their review articles exploring WPV specifically within an ED setting, Aljohani et al. postulated that 77% of ED staff members had been exposed to WPV, and Nikathil et al. suggested a rate of 36 violent patients for every 10,000 attendances (5, 6).

Aljohani et al. further explored the breakdown of WPV by role in their review and meta-analysis. They found that the majority of ED workers involved in WPV were nurses (55.7%) followed by physicians (36.5%), and that the primary instigators of WPV were family members (52%) followed by patients themselves (27%).

Many factors can lead to exposure to WPV, the causes of which have been extensively studied. The majority of studies note that nurses are the group most likely to be victims of abuse. Studies have also shown that increasing age, and a longer duration of employment are risk factors (7, 8). The association with gender and WPV is less clear, with some studies reporting men to experience higher rates of WPV, whereas a systematic review of studies in the African healthcare setting showed females to be at higher risk of WPV (8, 9).

Recent literature suggests that during the COVID-19 pandemic, there has been an alarming increase in WPV. For example, Chiroco et al. reported that WPV during this period was excessively high (10). Similarly, McGuire et al. found that, within one North American hospital referral region, violent incidents during the pandemic increased to 2.53/1000 visits compared to 1.13/1000 in the 3 months prior and 1.24/1000 in the year prior (11).

From a British perspective, the 2021 UK NHS Staff Survey revealed that within the previous 12 months, 14% of all NHS staff had experienced some sort of physical violence. A further 28% of the approximately 600,000 responses had experienced non-physical violence (12). Of frontline staff in hospitals with frequent patient contact, 19% were exposed to physical violence and 35% to non-physical violence. Results of the survey also indicated that only 42% of respondents reported an episode of violence. According to the survey, the groups most exposed to physical violence within the last 12 months were nursing and healthcare assistants (35.5%), followed by registered nurses and midwives (23.7%), and lastly medical and dental staff (11.4%). This was in contrast to harassment/bullying/verbal abuse where the groups most exposed were nurses and midwives (39.3%), followed by nursing axillary (38.0%), and finally medical and dental (34.2%). Two recent surveys conducted during the pandemic found that, over a one-month period, 30% of hospital doctors and 50% of General Practitioners had experienced verbal abuse, while the second survey found 35% of doctors had been subjected to verbal or physical abuse during the pandemic (13, 14).

The negative association of WPV with HCWs’ well-being has been extensively documented, particularly its role in depression, lifestyle changes, anxiety, PTSD and burnout (15–18). The 2022 GMC survey revealed that Emergency Medicine (EM) staff were at the highest risk of burnout, with 32% and 26% rates for trainees and trainers, respectively. Higher turnover of physicians has been reported in staff who have experienced WPV. This phenomenon is especially impactful in EM, which has a higher level of attrition than other specialities (19). Indeed, a review by McDermid et al. highlighted workplace aggression and violence as major themes in EM nurses (20). Moreover, a recent survey by the Royal College of Emergency Medicine reported that a large percentage of EM staff were considering altering working hours or leaving the field (21).

With this in mind, the aim of this study was to investigate and quantify trends in WPV within NHS EDs in the UK during the five-year period that included the COVID-19 pandemic. Since 2017 with the disbandment of NHS Protect, there has been no formalized national data collection on WPV for hospital trusts; therefore, this study serves to fill that knowledge gap, and provide justification for NHS trust-level policies to prevent WPV and support staff who fall victim to it.

We sent Freedom of Information requests to all 152 trusts within England, Scotland, Wales and Northern Ireland that included a Type 1 department defined by NHS England as “A major Accident and Emergency (A&E) department – A consultant led 24-h service with full resuscitation facilities and designated accommodation for the reception of accident and emergency patients” and/or a Type 2 department defined as “a consultant led single speciality A&E service (e.g., ophthalmology, dental) with designated accommodation for the reception of emergency patients” (22).

We requested specific information on WPV, including physical assaults resulting in injury and non-injury, verbal abuse, sexual harassment/abuse, and other abuse not falling into these categories. Trusts were given a minimum of 80 working days to respond to the request. Trusts were requested to provide data on patient, relative or visitor violence on HCWs. Where patient/visitor on patient/visitor, HCW on HCW or HCW on patient/visitor information was provided, these were excluded from the analysis.

We requested aggregate data if we could not obtain exact numbers based on confidentiality exemptions. Where, due to low numbers and confidentiality concerns, only ranges were provided, the most conservative figure was used. For example, a range of five to ten was recorded as five.

We gathered monthly ED attendance data from publicly and freely available governmental sources; NHS England, Public Health Scotland, StatsWales and Northern Ireland Department of Health websites (23–26). For trusts that merged during the data collection period, data from the constituent trusts were consolidated for the new trusts. A&E attendances were defined by all 4 sources as ‘unplanned new’ or ‘unplanned review’ attendances whether they were admitted or not (22, 24–26).

Ratios were calculated by incidents reported compared to total attendance figures, and we only used attendance figures for the trusts reporting for that time period.

Descriptive data were collected and analyzed using Microsoft Excel®.

A total of 133 (88%) trusts responded to our FOI requests. Six trusts could not provide data due to confidentiality or storage issues (4%). Therefore, data were available from 127 (84%).

This study presents the results of a large, multicentre investigation of WPV within EDs, at a regional and country level. To our knowledge, this is the first study to investigate the country-wide associations of WPV, the COVID-19 pandemic and the subsequent lockdown.

Of great concern is that our study has shown that WPV has persisted above pre-pandemic levels both in the total number of incidents and per attendance, a change that is statistically significant. While there were small detectable increases prior to COVID-19, there was a notable increase in violence at the start of the lockdown period and a further drastic and significant increase when the lockdown was eased.

Our results corroborate the results of an Israeli study noting HCWs’ perception of increasing WPV during the pandemic (30). Our data are also in keeping with the findings of the McGuire et al. study, which demonstrated an increase in WPV during COVID-19. Albeit, the increase in WPV described in our study throughout the COVID period was 50%, compared to 200% in the paper by McGuire et al. (8) A similar increase in WPV has been documented in primary care within the UK, with violent incidents doubling in 5 years (31).

Our study also demonstrates regional variation in WPV, both within the Home Nations as well as the English regions. For example, the rate of WPV in Scotland decreased over the COVID period, as did WPV in the Midlands and East of England regions. By contrast, WPV increased significantly in London (64.17/100,000 to 114.27 p < 0.05) and the North West of England (41.00/100,000 to 86.72/100,000 p < 0.05). Furthermore, physical violence increased significantly in the North West of England (13.01/100,000 to 35.65/100,000 p < 0.05). Although the factors contributing to this regional variation are beyond the scope of this study, it represents an interesting area for further research. Nonetheless, policymakers should be mindful of areas with an increased incidence of violence and prioritize these areas for the implementation of anti-violence policies and campaigns.

Additionally, we have identified certain regions where there is an increasing trend of physical violence. Specific policies designed to mitigate the effect of physical abuse, where strategies can differ from other types of abuse, should be targeted to these regions at risk (32).

As previously noted in the NHS staff survey, not all WPV incidents are reported. Indeed, 25–50% of all incidents go unreported (12). Therefore, the true incidence and, by extension, the impact of WPV is likely to be higher than that posited in this paper.

Academics have hypothesized several potential precipitants of WPV against HCWs. Previous studies identified the following as potential triggers: long wait times to see a clinician or receive medication, a shortage of staff, exposure to pain, high-stress levels and feelings of anger, frustration and unmet patient needs (33, 34). These proposed triggers provide an interesting comparison to our own study, which spanned the COVID-19 pandemic. NHS data revealed the height of the pandemic to be a period of low attendance and improved wait times within EDs (35). On that basis, therefore, one would expect violence levels to have fallen as the aforementioned precipitants were largely addressed, inadvertently or otherwise. This, however, was not the finding in our study. Therefore, explaining the increase in violence is not as simple as blaming the pandemic. By logical extension, a ‘return-to-normal’ cannot be expected to be the sole antidote.

More recently the NHS has been experiencing a retention crisis with healthcare workers of all disciplines leaving the field, including 34,000 nurses leaving their posts and 10,000 doctors giving up their license to practice in 2022 (36, 37). Worryingly in a survey of 4,500 junior doctors in the UK a third planned to work abroad in the next 12 months, and 40% would leave the NHS as soon as they found another job (38). Job dissatisfaction has recently manifested itself with the most widespread industrial action within the NHS. A recent meta-analysis of 6,400 nurses showed that WPV was associated with nursing turnover (39). Furthermore, recent studies concluded that perceived organizational support reduced the effect of WPV upon turnover (40, 41).

For a nationalized health service like the NHS, which already suffers from 133,000 vacancies (9.7% of the total workforce) in England alone, the impact of increasing violence against HCWs could be catastrophic toward this retention crisis (42). The negative psychological effects, such as stress and burnout, of WPV on HCWs have been well documented (15–18). With a steep rise in WPV, it is highly likely that the collective morale of the workforce has plummeted, and their well-being suffered. Within a cohort of ED nurses, there was shown to be a direct correlation between experiencing WPV and reduced job satisfaction and increased turnover (43).

Additionally, poor staff morale and widespread rota gaps could lead to worsening service provision capability, which can have significant detrimental effects on patient care and outcomes, as well as triggering further WPV; a vicious cycle ensues.

Previously, efforts to reduce the incidence of WPV and its effects on HCWs have been targeted toward security interventions, support groups and staff training, which have all been discussed as positive and useful interventions (44–46). In view of the findings of our study, these and other interventions to safeguard staff would benefit from increased resource allocation and implementation within the NHS.

In 2020 NHS England released the Violence Prevention and Reduction standard (47). Despite this, WPV remains concerningly high, and our results should provide policymakers with evidence that further efforts are needed to halt and reverse this worrying trend. In the current climate of widespread industrial action over working conditions within the NHS, WPV should be an area that is focused on nationally in light of this study’s findings. We further suggest that reducing WPV and increasing support for WPV victims would have an immediate effect on staff morale and well-being, and improve efforts at reducing staff turnover. This should be done expeditiously in view of the current NHS staffing crisis.

Due to a large number of trusts and incident reporting systems, there is the potential for a wide variation in the way incidents have been reported and recorded by each individual trust, which could impact results. Twenty trusts were unable to provide exact figures; however, as this was the case only if incident numbers were low, the figures are judged to have a small impact on the overall validity of the results and would lead to an under-represented number of WPV incidents. Further due to the anonymized nature of the data sets released via the FOI requests and publicly available attendance data, we did not receive any demographic data on any of the participants involved in the violence. Therefore, specific commentary on the representativeness of the data to the entire UK population is limited. Due to the small number of time points, particularly in regard to yearly data analysis, the potential for Type II statistical errors with no trend detected is elevated. Further data gathering covering more time points in the future could mitigate this potential.

Our study has demonstrated that there has been a dramatic rise of 50% in the number of WPV incidents recorded throughout 2017–2022 in the UK, with significant, drastic increases over the COVID lockdown period. WPV critically impacts upon staff morale and well-being, resulting in problems related to staff retention, recruitment and the efficient running of the department. Ultimately, patient outcomes are put at risk. There is some regional and geographical variance in rates of violence, the cause of which is uncertain. Further investigation is required.

We believe that the findings of this study highlight an emergency in WPV, and should serve as a precipitant for targeted reduction strategies. This paper provides evidence to encourage healthcare leaders to accelerate the formulation and implementation of policies to safeguard HCWs in a period of particularly low morale and well-being as shown by large-scale industrial action within the workforce. With the ongoing retention crisis within the NHS, a reduction in WPV will provide an immediate boost to an increasingly dejected and demoralized workforce.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

ND contributed to conceptualization, methodology, and data collection. TL contributed to software and formal statistical analysis. All authors contributed to manuscript draft and review. All authors have read and agreed to the published version of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.