Preventing Violence against Healthcare Workers in Hospital Settings: A Systematic Review of Nonpharmacological Interventions

,


Background
Te World Health Organisation (WHO) defnes violence as "the intentional use of physical force or power, threatened or actual, against oneself, another person, or against a group or community, that either results in or has a high likelihood of resulting in injury, death, psychological harm, maldevelopment or deprivation" [1].WHO reports that violence in healthcare settings represents about a quarter of violence in all workplaces and that violence against healthcare workers is a worldwide problem [2].More than 50% of those responding to this seven-country study reported at least one incident of physical or nonphysical violence in the preceding year.A more recent systemic review indicated that the 12-month prevalence of any form of physical or nonphysical violence against healthcare workers was 61% globally [3].
Healthcare workers experience patient-or visitorperpetrated violence at some stage in their careers [4], resulting in minor to major physical injury, death, psychological trauma, mental health concerns, and/or emotional burdens such as anger, sadness, fear, and disgust [5][6][7].Likewise, the healthcare environment poses a risk of exposure to violence like frontline police personnel, with the healthcare sector accounting for more than half of all violent incidents reported in workplaces [8,9].Workplace violence in healthcare predominantly occurs in "high risk" areas including emergency departments, psychiatric units, intensive care units, obstetrics and gynaecology, and acute medical/surgical units [10], yet the epidemiology of violent patient behaviours varies greatly.Contributing factors associated with heightened aggression include the haemodynamic instability of patients, their impaired neurological status, employee workloads, lack of experience, and burnout and environmental demands in hospital contexts [4,[10][11][12][13].
With such a range of patient, professional and organisational factors arguably inherent or inevitable in healthcare contexts, it is important to better understand and prevent violence by reporting and responding to it.Nonetheless, actual rates of violence may be higher than evidence suggests, as many studies indicate an underreporting phenomenon due to numerous factors including confusion over what constitutes violence [8], time constraints preventing disclosure [8,14,15], fear of repercussions [5], and the belief that violence is merely "part of the job" [16].
Recently, serious incidences of violence in healthcare have been covered broadly in the media, and in some instances, fatalities have been reported.For example, an Australian surgeon was fatally punched by a patient [17], an American mental health patient was charged with the manslaughter of a nurse [18], in China, a doctor was stabbed to death by a patient's family member [19], and in South Africa, an anaesthetist was fatally shot in an apparent revenge attack following the death of a child during surgery [20].Likewise, serious assaults are also common.An Australian nurse was knocked unconscious and required cardiopulmonary resuscitation following a patient attack [21]; in the United States, a nurse had their stethoscope wrapped around their neck and was forced to defend herself [22], and England's chief nurse called for an end to physical and verbal assaults on nurses as they worked through the COVID-19 pandemic [23].
Physical injuries such as bruising, cuts, and abrasions are the most reported injuries from violence in hospitals [5], although the psychological burden on healthcare workers following an assault includes fear, anger, anxiety, guilt, shame, self-blame, reduced job satisfaction, increased desire to change employment, and a reduction in health-related quality of life [6,7].Organisationally, violence is also associated with an increased risk of absenteeism, lowered morale, and decreased productivity amongst healthcare workers [24][25][26].Te economic cost of violence in healthcare is signifcant with treatment, leave expenses, staf burnout, compensation, increased turnover, and security, litigation, and property repair expenses burdening the sector [8,9,11,27].Despite its impacts, few studies have explored the effcacy of interventions to prevent violence.Whilst some studies have explored the role of pharmacological strategies like sedating or restraining violent patients [5,8,15,28], drug-based therapeutic solutions [5,15,28] are not a leading solution to workplace violence in healthcare and can lead to serious consequences for patients [5,15,29].Currently, no systematic reviews have been conducted to measure the efectiveness of nonpharmacological interventions to prevent patient-perpetrated violence against healthcare professionals within hospital settings.Key objectives of this review include identifying interventions efective to prevent violence and evaluating the extent of evidence to support specifc preventive approaches to violence.We considered any nonpharmacological intervention as those which included but were not limited to education and training, human resourcing interventions such as stafng, security interventions, environmental modifcation, policies and procedures, and behaviour contracts.
1.1.Aim.Te aim of this systematic review is to analyse nonpharmacological interventions for the prevention and management of violence towards healthcare workers by patients and visitors.

Design. Recommendations from the Preferred
Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) 2015 statements have been used to develop the methods for undertaking this systematic review [30].Te systematic review protocol has been registered in the International Prospective Register of Systematic Reviews (PROSPERO) CRD42017065700.

Inclusion and Exclusion
Criteria.Any interventional study that focused on preventing violence in a population of adult hospital patients or visitors of 18 years of age and older was eligible for inclusion (see Table 1).Interventions included were those designed with the stated objective of preventing or reducing the incidence of violence/aggression perpetrated by patients or visitors in the hospital.Studies were not restricted based on design; however, studies were required to have a comparison group of some kind.Outcomes were focused on the efectiveness of the intervention on deescalation, detection of the risk of violence, frequency of violence, injury, and the use of restraint/seclusion.Tis review only focused on reported acts of violence or aggression committed by patients or visitors within the hospital setting.
1.2.3.Search Strategy.Scopus, PubMed, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Psy-cINFO, and the Cochrane Library were searched.Te reference list of included articles was also read to search for additional potential studies for review.Te search strategy is described in Table 2.All imported data from the search strategies were stored in EndNote with duplicates removed.Initially, 5,558 citations were retrieved, and of these, 1,883 were removed via EndNote as duplications.Te remaining 3,675 were then screened for eligibility independently by two authors (NM, NR).A total of 160 full-text articles were retrieved for detailed scrutiny to further assess eligibility.Twelve articles met all inclusion criteria and were selected for review.Table 3 provides a summary of those full-text studies reviewed for inclusion and those excluded, and Figure 1 provides an outline of the screening process.

Data Extraction.
Included studies were categorized into groups based on key interventions.Additional information extracted included the type and size of the study, participant groups, outcomes, and the risk of bias score calculated.Data were independently extracted and checked for concordance between reviewers (NR, NM) using a data extraction tool with a minimum of 95% concordance for extraction accuracy to ensure the completion of data.
1.2.5.Quality Assessment.In systematic reviews, especially those involving health services research, the appraisal of methodological quality and quality of reporting is crucial.Traditionally, separate tools have been used for quantitative and qualitative research, owing to the distinct nature of these methodologies which poses challenges in the cohesion of evidence synthesis.Te Quality Assessment for Diverse Studies (QuADS) tool addresses this gap by providing a standardized, empirically grounded tool suitable for a variety of study designs, including mixed-methods research [36].Te QuADS tool was used as it demonstrates strong reliability and ease of use for application to multi-or mixedmethods health services research reviews [36].Te quality assessment was performed independently by two reviewers (NM and NR) who followed the six-step procedure outlined in the QuADS criteria with a kappa value of 0.91 indicating excellent agreement with criterion-based quality assessment performed for each paper against the 13 diferent QuADS criteria independently (see Table 4).Any discrepancies were resolved through collaborative discussion.
1.2.6.Data Synthesis.Tere was marked methodological heterogeneity in the studies, and as a result, narrative synthesis was chosen as the most appropriate method to analyse and explain fndings.Narrative synthesis involved a modifed version of the framework developed to investigate fndings and to explore relationships within and between the data [37].Tis process involved tabulating study data under headings that included participants, intervention, comparisons, outcomes, and themes.Tabulation allowed for consideration of themes related to key interventions, the development of textual descriptions, and the exploration of relationships within and between studies for an overall assessment of the strength of evidence [37].
Regardless of the variations between intervention types, population groups, and design methods, each study reported on the incidence of physical and/or verbal violence either as a combined or singular phenomenon.Five outcomes of interest were reported across the twelve studies for inclusion as follows: "Frequency of Violence," "Capability to De-escalate," "Screening for Risk of Violence," "Injury," and "Need for Restraint/Seclusion."

Interventions for Decreasing the Frequency of Violence.
Overall, ten of the twelve studies noted a decrease in the frequency of violence postintervention whilst two reported no statistically signifcant diference [11,29].Of the ten, one also noted a decrease in the incidence of patient-to-patient violence and the use of seclusion/restraints but no change in the rate of patient-to-staf violence [32].Reductions in the incidence of violence ranged from 23% to 91.6% [10,27]

Interventions for Screening for Risk of Violence.
Tree studies utilised screening assessment tools to aid in the identifcation and perception of potentially violent patients [8,9,34].One of these studies reported an overall reasonable efectiveness in identifying those most at risk of committing violence with moderate sensitivity (71%) and high specifcity (94%) [9].Te second study noted decreased frequency and recurrence of violent incidents [8], whilst the third noted a reduction in the perceived risk of violence from 46 to 35% [34].Characteristics of interventions screening for the risk of violence included the Alert Assessment Form protocol, and the use of specifc characteristic criteria to recognise those patients most at risk [8,9,34].OR TITLE-ABS (de AND escalation) OR TITLE-ABS (reducing)) AND (TITLE-ABS (hospital) OR TITLE-ABS (ward) OR TITLE-ABS (unit)) AND (LIMIT-TO (DOCTYPE, "ar") OR LIMIT-TO (DOCTYPE, "re") OR LIMIT-TO (DOCTYPE, "ip")) AND (LIMIT-TO (LANGUAGE, "English")) AND (LIMIT-TO (DOCTYPE, "ar") OR LIMIT-TO (DOCTYPE, "re") OR LIMIT-TO (DOCTYPE, "ip")) AND (LIMIT-TO (EXACTKEYWORD, "Human")) AND (EXCLUDE (SUBJAREA, "BIOC") OR EXCLUDE (SUBJAREA, "DENT")) PubMed

Interventions to Reduce
Violence-Related Injury.Two studies reported a decrease in the frequency of injury as a direct result of violence.One noted the rate of incidents decreased by 91.6% with staf visits to the medical centre reducing by 42.2% [27].Te second study had minimal impact on the occurrence of self-defence-related injuries in participants with pre-and postintervention rates of 37.5% and 35.7% [35].Tese study interventions were characterised by the creation of an instrument allowing employees to better report threats and acts of violence which directly impeded on patient care.

Interventions to Reduce the Need for Restraint/Seclusion.
Only one study highlighted a reduction in the use of seclusions and restraints [32].Te characteristics of the intervention diminished the need for restraining and secluding patients by examining the efcacy of the Mandt System.

Discussion
Importantly, following a systematic review of interventions for preventing violence in hospitalised patients, we identifed ten of the twelve included studies reported a decrease in the frequency of violence following the introduction of an intervention.However, with the current incidence of violence of concern in the literature, we note this review identifed only 12 studies of predominantly low quality.Nonetheless, the studies reported on a diverse range of efective interventions for improving health professionals' capabilities to deescalate violence in hospitals; screening for risk of violence; reducing the frequency of violence; preventing or reducing injuries from violence; and lowering the need for restraint/seclusion.Deescalation techniques predominantly adopt a "prevention is better than a cure" approach to violence management.Education-based training strategies are commonly utilised and readily adopted into practice by clinical staf [38].Tese interventions report an enhanced knowledge of self-defence, break-away techniques, and confdence in aggression management tactics [39][40][41].Regardless of their successes, these studies were of low quality with reporting methods being inconsistent.In noting this, it only reiterates that despite the best eforts to teach, train, and educate healthcare workers, current deescalation techniques are inadequately established.
Screening for potentially violent patients aids in the facilitation of early intervention, prevention, and management [13].Questionnaires, observational checklists, and risk identifcation screening tools are frequently used to assist in the recognition of high-risk patients [42][43][44].Such interventions have commendable success rates; however, they continue to fail when used as a stand-alone measure, and a general lack of consensus remains regarding follow-up   Journal of Nursing Management instructions [42].Terefore, it can be concluded that without the formulation of supplemental instructions, screening tools alone will prove insufcient.Multiple interventions have been formulated to assist in the decrease of patient-perpetrated aggression.Much of the literature highlights that targeted, clinically applicable, and efective solutions are needed to assist in the identifcation, prevention, and reduction of workplace violence [45,46].Despite this, current methods remain substandard irrespective of their optimistic impact [47].Until such a time that workplace-associated violence is adequately addressed, it can be rationalised that patients will continue to burden the healthcare sector with acts of aggression.
When evaluating the net beneft of establishing interventions for workplace violence management and prevention against leaving practice unchanged, the answer is obvious.Te fnancial strain it places on the hospital sector is immense [48].Likewise, psychological, physical, emotional, and social repercussions to staf leave them under signifcant pressure and unable to fulfl their role-required responsibilities [49,50].Immediate action through the establishment of quality interventions is fundamental to bringing positive change to the healthcare sector.By doing so, staf injury rates have the potential to decrease, proving benefcial to the workforce.
Te use of seclusion and restraint of patients, physical or chemical, remains highly controversial and problematic [51].Although it can be argued that these measures can improve patient and staf safety [52], many have serious negative repercussions and create ethical challenges in the delivery of care.Increased stress and agitation, loss of independence, restricted movement, haemodynamic instabilities, and deprivation of dignity have all been reported as adverse outcomes following restraints and seclusion [53].
Many studies have proposed the use of education-based interventions to reduce the need for such "invasive" measures.Seclusion and restraint (chemical and/or physical) of patients remain, both legally and ethically [51].Whilst the direct and indirect use of these measures can be somewhat justifed by maintaining safety [52], they can pose serious negative consequences.Increased agitation and stress, loss of independence and ability to move, the development and exacerbation of pressure area injuries, haemodynamic instability, and deprivation of dignity are all reported adverse outcomes from the use of restraint and seclusion [53].It has, therefore, been proposed that through education and awareness, the use of such measures can be reduced whilst maintaining a safe environment for all [53].In doing so, the perceived deprivation of basic human rights is lowered, and it becomes less distressing for patients, their families, and staf and facilitates a more positive environment for all.
Across all studies, we found that studies were of low quality and that no consistent outcome measures and reporting or defnitions of violence were used.Studies of various designs and interventions suggest that further work is needed in designing high-quality studies to robustly evaluate interventions that aim to prevent violence in hospitals.Further research is, therefore, needed to develop high-quality interventions using recognised methodologies and test-developed interventions using rigorous designs.
Tis review is the frst of its kind in collating and synthesising the available evidence on the important topic of violence prevention.Conclusively, the deterrence of patientperpetrated violence benefts the entirety of the healthcare system.Unnecessary expenses are reduced, workforce retention improves, the safety of staf, patients, and visitors is maintained, and the overall functionality of the hospital is enhanced.Despite some promising interventions presented in the literature, the quality of the reporting of such remains inconsistent as per our quality assessment.Moreover, the literature is scant in the context of frequent violence in healthcare settings.Multiple evidence gaps were noted which require urgent investigation to ensure the formulation of evidence-based policies and the implementation of effective interventions into practice.

Implications for Nursing Management
Tis systematic review highlights the need for nurse managers to prioritize the development and implementation of evidence-based strategies to mitigate violence in healthcare settings.Nursing leaders are encouraged to actively engage in the development and enforcement of robust violence prevention protocols, focusing on a multifaceted approach that includes education-based interventions, efective deescalation techniques, and rigorous screening tools for potential violence.Emphasis should be placed on continuous training and support for clinical staf, fostering a culture of safety and preparedness against patient-perpetrated aggression.Furthermore, nursing management must advocate for high-quality research to refne existing interventions and explore novel strategies.By doing so, not only can the incidence of violence be reduced but also the physical and psychological well-being of healthcare professionals can be safeguarded, ultimately enhancing patient care and overall hospital functionality.Te role of nursing management is pivotal in bridging the gap between current practices and the ideal state of violence-free healthcare environments, ensuring the safety and dignity of both staf and patients alike.

Table 1 :
Inclusion and exclusion criteria.

Table 2 :
Preliminary search strategy for systematic review.

Table 3 :
Table of included studies, n �

Table 4 :
Quality Assessment for Diverse Studies (QuADS).