Violence in emergency departments: under-reported, unconstrained, and unconscionable

Two Australian studies found that over 60% of nurses had experienced violence in the workplace in the recent past (1–5 months),4,5 and, in one study, almost 30% had experienced violence on the day of the survey.5 However, the experience of violence at work seems to be a universal experience for ED nurses, with almost 90% experiencing physical intimidation or assault at some point in their career, and all experiencing verbal abuse. This is highlighted in a retrospective questionnaire completed by almost 12% of ED nurses across NSW.6 These rates are entirely consistent with international data.7-9 EDs are among the highest risk settings for violence in the health workplace (others being geriatric wards and alcohol and drug units).10

More concerning, perhaps, is the frequency of these episodes. In some EDs, violence is a daily occurrence, with nursing staff reporting several episodes of exposure to violence each week.4,5,11 The Australasian College for Emergency Medicine in its policy document Violence in emergency departments cites a rate of violent incidents of 3 per 1000 patient attendances at EDs, which typically equates to one every 1 or 2 days per department.12 The College believes “that in addition to acts of physical-contact violence, verbal abuse, threats, and aggressive behaviours are defined as acts of violence”.12

A review of security-based “code grey” responses (security staff response to unarmed threat or to violent behaviour) in the ED of a major metropolitan hospital showed a rate of 2.2 code calls per 1000 attendances for episodes of perceived threat to staff.13 The report provides a clear picture of the incidence of violent episodes and raises concern about the level of violence and risk experienced by staff. In as much as security staff were required to respond to these episodes, they were clearly not trivial.13

This concern is further strengthened by a reported need for restraint of patients at a rate of 3.3 episodes per 1000 attendances in Australasian EDs.14 This figure come from a survey of 116 Australasian EDs, which found that restraint was used for a range of patient conditions, in particular, violent or aggressive behaviour (not illness-related) (52%), psychosis (32%) and organic brain syndromes (10%).

Those with experience of working in EDs may be able to predict the occurrence of violence. Up to 50% of episodes are associated with alcohol or drugs, and the timing of violence is almost certainly related to social patterns of use of such substances, with violence occurring more commonly during the evening shift and at weekends.11,13 In addition, at these times, the ED (and hospital) generally is functioning with minimal staffing resources — not only lower staff numbers but also a higher proportion of junior staff on duty. Compared with more experienced senior colleagues, these staff members are usually less skilled in dealing with a challenging, aggressive patient or relative.

Studies have not shown a relationship between violence and an excessive length of ED waiting time. For example, in the study by Knott et al of 151 subjects, the median time from presentation to violent behaviour and a code grey call was 59 minutes. It is also likely that there are not strong links between violence and organic illness, but previous violent behaviour and, in some cases, mental illness may be contributing factors.13,14

It is likely that the social settings in which injury occurs, and the violent nature of these settings, also contribute to violence in the ED. Violent crime is increasing in our society.15 Increased societal violence will result in increased ED presentations for injury. People in such situations are often stimulated, angry, confused and tense, and may be in pain and worried about the severity of injury they or others have sustained. This volatile mixture of emotion and physiology results in a mental state characterised by anxiety, intolerance and a propensity to aggression and violence.

Retrospective surveys of staff and ED directors show that up to 70% of episodes of violence are not formally reported through hospital incident reporting or other systems.6,17 This trend is also seen in other health settings and internationally.7,10

The under-reporting of violence is not a new phenomenon. It is well known that sexual assault, domestic violence and workplace bullying are under-reported.16,18 The reporting of violent acts through hospital risk management systems and to the police (in cases of violence in which illness is not a mitigating factor) is important in engaging governance systems and activating appropriate legal responses, with the aim of decreasing the risk of further violent acts.

It would be useful to be able to understand violence in the ED using the paradigm applied in domestic violence or sexual assault settings; however, there are fundamental differences. Two features of ED violence set it apart — there is no overt dominance or power status attached to the aggressor, and the violence occurs in a setting of care where the victims are likely to excuse the behaviour on compassionate grounds.

In my opinion, the reasons for under-reporting in the ED setting are more likely to be related to organisational behaviour than to “complex victimology”. “System blockers” to reporting injury and adverse events abound in most hospitals — there are complex forms to complete, lack of time to complete reports, unclear policies or protocols, systems that are not convincingly confidential, negative feedback through lack of system response capacity, peer pressure, and the stigma of victimisation. In addition, episodes of violence are so frequent that staff members may become desensitised to its significance and start to see it as part of the job. Moreover, most violence in health care settings does not result in serious physical injury, and staff reporting clearly occurs more frequently when greater physical impact is involved.19

An understanding of the factors contributing to under-reporting of violence and strategies to address these factors will be important in turning around these behavioural tides.

• The Australasian College for Emergency Medicine’s policy document released in 2004 clearly states its position:12

  — hospitals need to work with EDs to assess the risk of violence and to proactively work to reduce that risk; and

  — EDs need structural protection and security response systems, including the presence of adequate security staff.

• The Australian Nursing Federation’s policy document entitled Zero tolerance to violence released in 2002 advocates strongly for a safe and secure work environment.20

The UK National Health Service has adopted a highly visible “zero tolerance” framework, which includes mandatory organisational and police reporting of violent acts.21

However, peak body policies do not necessarily change behaviour or have sufficient jurisdictional influence to effect organisational change. Surprisingly, more formal legal and regulatory frameworks (such as workers’ compensation systems and industrial legislative frameworks) experience similar frustrations.

Employers have a duty of care in relation to the health and safety of their employees at work. This duty cannot be delegated to anyone else and is strengthened by legislated occupational health and safety acts; eg, the Occupational Health and Safety Act 2004 (Vic), which sets out various duties for employers in relation to their employees, including in Section 21 (1): “An employer must, so far as is reasonably practicable, provide and maintain for employees of the employer a working environment that is safe and without risks to health.”22 Employers in breach of this Act may be at risk of prosecution or common law action.

Despite the clarity and formality of this governance requirement, the issue of workplace violence and the expected role of the employer in preventing or mitigating violence is yet to be effectively or comprehensively adopted.

Firstly, the presence of security staff in visible locations at the entrance to the ED may be an effective deterrent to violent behaviour. This is particularly so when combined with other structural elements of a security system, such as cameras, restricted access and metal detectors.23

Secondly, ED design issues also impact on security and safety.24 In the past, a fortification approach has been the preferred model, with glass barriers that prevent effective communication, and isolated waiting areas that limit contact between patients and staff. Such approaches are now regularly challenged in public areas with greater security risks than the EDs, such as banks, convenience stores and courts. In such settings, less obtrusive (but effective) barriers, better communication and a more comfortable environment all contribute to an improved experience for the user and thereby decrease the propensity to violence.25 This is particularly true in the case of violence associated with mental illness, where a more supportive and calm environment has therapeutic and behaviour modifying impacts.

Thirdly, an extremely effective tool which has been used in banks is now being introduced in some EDs. At the initial contact point, the presenting client can see themselves on a TV monitor interacting with the triage nurse at reception. This provides an instant message to each person that his or her behaviour is under observation, and implied messages about the expectation of appropriate behaviour. Less subtle messages are also being deliver-ed in waiting room posters, video presentations and patient information sheets. These give explicit information about expected behavioural standards.26

A range of systematised responses to violence are available. An important feature of these systems is an alert mechanism. Such alarm systems are now sophisticated and relatively inexpensive. A combination of fixed alert trigger points, and mobile individual duress alarms should be available in every ED. Similarly, a specialised staff team response must be available in every setting to deal with violent episodes. The presence of 24-hour trained security staff is a component of this response, which should also involve training of peers and management in techniques and strategies that minimise escalation of violent behaviour, and mitigate against its impacts. Follow-up systems for the debriefing or support of involved staff should also be linked to these systems.

The final component of current strategic approaches — the training of all staff in aggression management systems — is likely to be the most effective tool in the longer term.5,9 The aim is to reduce the incidence of aggression escalating into violence, and train staff in better management of violent episodes. Such systems provide staff with special skills and techniques to deal with these high-risk issues in a peer-supported and organisationally sanctioned program. In this way, they contribute strongly to a non-violent culture, and to the governance responsibilities of organisations to provide a safe and secure workplace.

A key factor in this type of intervention is a strong preventive orientation looking for high-risk indicators, and may extend to active screening systems (physical and behavioural). The active identification of clients at higher risk of violent behaviour (eg, those with a previous episode of violence or presenting as a result of interpersonal violence, those with disturbed behaviour due to substance abuse, and some mental health clients) raises the awareness of staff and encourages proactive behaviour management.

The searching of ED users and their belongings for weapons through either metal detectors or physical inspection is becoming more widespread and should not be considered invasive or inappropriate if applied selectively within a policy framework, and with clear communication to clients.