kNOw workplace violence: developing programs for managing the risk of aggression in the health care setting

John A Forster, Mark T Petty, Colin Schleiger and Helen C Walters
Med J Aust 2005; 183 (7): 357-361. || doi: 10.5694/j.1326-5377.2005.tb07083.x
Published online: 3 October 2005


  • Strategies to prevent and manage violence and aggression in the health care setting have become a primary health and safety issue.

  • A series of vignettes are provided to highlight key elements in developing a program for preventing behavioural violence and aggression in a tertiary hospital.

  • Key components of the program include staff education and training, risk assessment and management practices, the use of patient contracts and policy development.

  • The program aims to integrate and balance occupational health and safety obligations to staff with the duty of care owed to patients.

In recent times, there has been increasing concern about the prevention of violence and aggression in health care settings. A number of studies have demonstrated that health care environments can be a violent workplace, with hospital staff, particularly nurses and doctors, at risk of verbal and physical assault.1-4

In 1999, the United Kingdom introduced a “zero tolerance” program with the single and simple message, “Violence and aggression against health care workers will not be tolerated”.5 In Australia there have been a few small case studies of occupational violence, but most information is drawn from research studies in the United Kingdom and United States.6 Australian state governments are beginning to formulate policy and framework guidelines on violence and aggression in the health care system. Among other things, these policies recognise the problem of violence towards health care workers and aim to help recruit and retain staff by ensuring a safe working environment.

Austin Health is an 800-bed tertiary teaching hospital providing health care services to the North East Metropolitan Region of Melbourne. In recent years, Austin Health has developed a comprehensive program for preventing violence and aggression. The program was devised in a progressive, step-by-step manner by a multidisciplinary working group that has met monthly since July 2000. The working group reviews incidents involving violence in the hospital, identifies and discusses related issues and published reports on the subject and develops appropriate protocols.

Here we present four fictional vignettes, based on a composite of real cases seen at Austin Health, highlighting some of the key stages in developing our program. The program focuses on behavioural interventions to support staff in actively preventing and dealing with aggression from patients, their relatives and other non-hospital employees in the workplace. Although staff and patients may also be exposed to violence from other staff members, this is addressed by a different staff code of practice, involving staff disciplinary and grievance procedures, that will not be discussed here.

Vignette 1: the angry old widower

Mr A was an 85-year-old widower living independently at home. Over a period of 3 months, he was admitted to a surgical ward on several occasions. He had peripheral vascular disease and required two below-knee amputations.

During his hospital stays, Mr A swore at staff and attacked them verbally. He also threw meal trays and water jugs at staff and, on several occasions, punched nursing staff and spat at his treating doctors. He was non-compliant with care routines and refused any psychiatric support. A neuropsychological assessment deemed him competent.

The impact on the nursing, medical and allied health staff was to create fear and anger, resulting in either avoidance or confrontation when providing care. Staff sick leave increased, and the multidisciplinary team became divided about how best to manage Mr A’s behaviour, as team members sought to reconcile patient care needs with staff safety. This resulted in frequent changes in treatment plans, none of which met with much success.

Ultimately, ward staff threatened to refuse to care for Mr A and requested that the Chief Medical Officer remove him from their unit.


Cases similar to this hypothetical one presented frequently for consideration early on in our working group’s history. They highlighted the lack of a clear set of policies and procedures in hospitals that recognised the implications of violence and the need to prevent it. Typically, staff “put up” with patient aggression as a “normal” way of working and as part of the job — until they reached breaking point.7 There were no proactive systematic and coordinated strategies to prevent further episodes of aggression and violence once a situation had been recognised. This was compounded by the absence of a clear message to patients that violence towards staff was unacceptable and would not be tolerated.2,8

The response seen in situations similar to Mr A’s also demonstrated a limited consideration of the patient and his or her experience of hospitalisation. Mr A found himself in the midst of a controlling system that held a fixed view of health care delivery. He was confined to bed, with limited opportunities to visit the outside world. Although this did not necessarily give rise to his aggression, his carers seemed to have little awareness of the anxiety and sense of powerlessness that led to his frustration.9

It was in response to situations such as these, and their effect on staff, that our working group was established to review how such situations could be better managed and/or prevented. Key outcomes of the group’s review were:

  • The development of a staff and patient awareness campaign to prevent aggression and violence. This took the form of a series of public posters promoting the message that violence is unaccept-able at Austin Health (Box 1) and a promotional “Say no to aggression” day that was led by the Chief Executive Officer and hospital executive.10

  • The development and implementation of an aggression management training program for clinical staff. This includes modules on managing verbal aggression and advanced skills in preventing and managing violence.11-13

Vignette 2: the man who couldn’t sleep

Mr B was an 88-year-old married man, living with and caring for his 89-year-old wife. He was admitted to a general medical ward after presenting to the emergency department with acute urinary retention. On admission, Mr B was described as being “disorientated and confused”. He was later reported as being “unable to sleep” and was found in another patient’s room. When a nurse tried to direct him back to his room, he became verbally aggressive and attempted to punch her. Eventually, Mr B responded to staff directions, returned to his bed and settled with intramuscular sedation.

After this episode, Mr B was found lying next to his bed with two superficial, self-inflicted abdominal knife wounds. When two staff members tried to attend to him, he struck one in the face and threw an over-bed table at the other. He then began to threaten these and other staff members with a pocket-knife. Security staff were required to help return him to bed. He was mechanically restrained before being given intramuscular sedation.

A subsequent mental state examination revealed marked fluctuations of consciousness from minute to minute — a diagnosis of delirium was made. He remained agitated and anxious, with gross disorientation in time and place.

Further assessment revealed that Mr B had experienced visual and auditory hallucinations of young boys playing boisterously near him; the boys were mildly threatening towards him. He described having thoughts of killing himself as his only means of escaping this torment. He had also perceived the staff as threatening to harm him, and had attacked them in order to defend himself.

Staff caring for Mr B felt “traumatised”. They required several debriefing sessions to support them before they felt able to return to work on the ward.


Cases similar to Mr B’s were relatively frequent in the early days of our working group. Management plans often demonstrated a lack of awareness and skill in assessing potential aggression in patients, particularly in those who experience delirium and may misinterpret the hospital environment and staff. This type of incident showed us that aggression risk assessment practices could be improved.

It was evident that attention was not directed to what happened before an episode of aggression, with a consequent tendency to be reactive to the first incident, rather than proactive. These cases highlighted the need for an aggression risk assessment approach to patients and a review of the hospital’s systematic response to potentially aggressive patients. Previous risk assessments for aggression had consisted of being aware of common predictors of violence, with the most accurate predictor being a history of violence.14,15 But, significantly, there had been little work at a local level on risk assessment for potential aggression and on linking this assessment to actions that could reduce the potential for high-risk situations.

Accordingly, we developed and implemented an aggression risk assessment and management tool. The key risk assessment identifiers used in this tool are shown in Box 2. The identifiers are based on objective information, the aim being to avoid the possibility of a staff member making a subjective assessment of patient aggression based on ignorance, unfamiliarity or negative assumptions.

A key feature of the tool is to “weight” identifiers as being low, medium or high risk and directly link them to actions for staff to take. Appropriate use of the tool allows staff to proactively develop fine-tuned strategies to prevent an aggressive incident from occurring.

A critical element in successfully implementing the tool has been the development of an alert system to communicate critical information about potential aggression to other staff members. Risk features are recorded in the patient’s clinical file and are included in hand-over information at each shift and during medical rounds and team meetings. The premise is that if staff feel safe it is more likely that the patient will be safe and receive optimal care.

Vignette 3: the visiting son

Mr C’s terminally ill father was an inpatient. During visits to his father, Mr C persistently carried on loud mobile telephone conversations in the ward, during which he would make derogatory, demeaning and abusive remarks about hospital staff, despite being requested not to do so while in the hospital environment. Mr C also intimidated staff by yelling and threatening them. His behaviour escalated to the point where he physically assaulted one of the nursing staff on the ward.

Mr C was served with a behaviour contract, which he refused to sign, stating that he would not do so on legal advice. Austin Health then served a “Not welcome” notice on Mr C — which essentially meant that he would be unable to visit his father in hospital. Mr C sought, unsuccessfully, to overturn the situation by approaching the Health Services Commission, the Minister for Health and the Shadow Minister for Health.

During the period when Mr C was “Not welcome”, he was given daily telephone updates on his father’s condition. When his father was dying, Austin Health arranged for Mr C to make visits in the presence of a security guard.

Mr C was subsequently convicted of assault.


Although less common than the potential threat of violence from patients, experiences similar to this case raised awareness that attacks on staff by visitors were an increasing occurrence.16 Visitors to the hospital are generally cooperative and positively contribute to the care and welfare of the patient they are visiting. However, a small minority have a negative effect on staff, other people, and possibly the patient being visited.7,13

Austin Health developed a policy and protocol for sanctions, such as “Not welcome” notices and contracts for acceptable visitor behaviour, in these situations.10 Typically, the process involves a series of warnings leading up to a “Not welcome” notice, which means the visitor is unable to enter Austin Health premises. A key message for staff was that violent behaviour towards them would not be tolerated and that sanctions would be applied. Staff were educated about appropriate standards of behaviour and appropriate use of warnings and “Not welcome” notices. They had regular liaison meetings with police, were informed about what records are required for court processes, and received briefing sessions on court procedure and giving evidence.

The notion of “Not welcome” notices and “Refuse to treat” options in health care systems is contentious in view of the sometimes conflicting obligations of having a duty of care to patients and providing a safe workplace for staff. (A “Refuse to treat” directive is the decision taken to discharge or preclude a patient from receiving care for non-life-threatening conditions if the patient displays ongoing, active and intentional violence towards staff.) The tension arising from imposing sanctions on patients and/or visitors provides a challenge for clinical and administrative teams. We have found that this can be avoided by providing a clear assessment of patient competency and a process for staff to follow.

The key message is that sanctions can be imposed on health care users6 who are deemed able to have responsibility for their actions. They should be given a warning notice that violent behaviour will not be tolerated. A “Refuse to treat” directive does not preclude patients from receiving care and treatment from the hospital, but implies that this will only be provided if their condition is life-threatening.

Vignette 4: the man with schizophrenia

Mr D, a long-term inpatient in a secure psychiatric facility, was in his mid-forties. He suffered from chronic paranoid schizophrenia and adult autism and was hard to understand because of muffled slurred speech, poverty of thought and an inability to adequately voice his needs and feelings. He often seemed frustrated by these circumstances.

At times Mr D was violent towards himself or others, often with little or no warning of imminent violent behaviour. A few years ago, he physically attacked a staff member, who did not return to clinical practice after the event.


Violent incidents in mental-health settings are not uncommon. However, in more recent years, deinstitutionalisation and mainstreaming of psychiatric services within acute-care hospital settings have meant that the staff resources and level of experience available to manage violence have been reduced. Further, as mental health treatment and care continue to move towards a community focus, patients needing inpatient treatment are sicker, with the result that violence towards health care practitioners is increasing.17,18

Although our programs offer aggression management and conflict resolution strategies for a variety of clinical situations, an organisational review conducted with mental health staff indicated that these programs did not adequately target special or specific situations. So a program was designed to meet the specific needs of mental health services at Austin Health. The program aimed to provide nursing staff with the skills and knowledge to confidently disengage from or manage violence within the workplace.19 A successful feature was a “train the trainer” component. This involved training “staff facilitators” to run the program autonomously in the workplace, ensuring ongoing skill development.


The vignettes presented here illustrate some of the kinds of situations in which clinical staff may be exposed to aggression and violence from patients and others. The key elements of the prevention program developed in response to these and similar situations are outlined in Box 3. We believe that having the support of hospital executive staff has enhanced the success of our program.

In trying to evaluate the success of the program, Austin Health found that scant data on aggression and violence in the workplace had been kept in a systematic way. Moreover, the validity and accuracy of the available data were questionable because of inconsistency in reporting and monitoring practices. This is consistent with published reports suggesting that 80% of incidents remain unrecorded.6 The problem is endemic to all health care settings, not just high-risk environments such as emergency departments or psychiatry units.2-4 Under-reporting by staff of aggression and violence may be largely due to a lack of clear policy and procedure as to what is acceptable behaviour and the absence of systematic mechanisms to deal with unacceptable behaviour.17

We used several proxy indicators to evaluate the impact of the anti-violence program (Box 4). While these figures indicate an improvement in the management of violence and aggression, they should be considered in a broader context. For example, the reduction in violence and resultant reduction in staff injury may also be attributed, to some degree, to a variety of support and educative initiatives that were being introduced concurrently with our program, such as clinical supervision and challenging behaviour workshops. It could also be argued that a recent increase in regular permanent staff and a reduction in casual staff use may have helped to reduce workplace violence.4 However, we believe our proxy indicators indicate that, with a consistent and comprehensive approach, workplace violence and aggression can be reduced.15,20

The lack of data to evaluate the success of the program also shows the importance of developing standard data collection tools and definitions for monitoring aggression and violence in the workplace and the need to encourage staff to report and document violent incidents.1,21 The challenge for all health systems will be to develop data collection methods that are consistent and allow meaningful comparisons.4,22


Since the inception of a coordinated program in 2000, Austin Health has progressively introduced strategies and policies that have improved its management of workplace violence and made it an industry leader in this field. Austin Health’s approach is holistic and incorporates the principle of integrating and balancing occupational health and safety obligations to staff with the duty of care owed to patients. We believe that exposure of health care staff to aggression and violence in the workplace can be reduced through systematic and coordinated strategies that include education and training,23 risk assessment and management practices, the use of patient/visitor contracts and policy development.

1 Poster informing patients that violence is unacceptable at Austin Health

2 Risk assessment identifiers

  • The person appears agitated or restless

  • There is resistance to suggested treatment

  • Aggression management has been required at time of transfer

  • The person has assaulted a health worker within the past 12 months

  • There is a known history of threatening or aggressive behaviour

  • The person has made a threat of aggression directed towards people or property

  • The person has friends or family members whose aggressive behaviour may place staff or others at risk

  • There is a known history of drug or alcohol misuse

  • A medical condition is present that may cause the person to misinterpret the environment or staff care activities (eg, confusion, disorientation, delirium, acute hallucinations, delusions)

3 Features that contribute to a successful institutional violence prevention program

  • The promotion of a philosophy that violence and aggression are unacceptable

  • A promotional campaign to convey to staff that the organisation values their wellbeing and safety in the workplace

  • A promotional campaign to convey to hospital patients and consumers that violence will not be tolerated and that sanctions will be applied

  • A response to incidents of aggression that considers both patient and staff safety

  • A risk management framework that includes a process for assessing potential risk of violence and developing subsequent strategies

  • Active involvement of senior clinicians and administrators in the incident response system

  • Debriefing and defusing mechanisms to support staff who have been exposed to aggression and violence in the workplace

  • Ongoing evaluation and development of programs to ensure the needs of staff, patients and the hospital continue to be considered

  • An educational program, accessible to all staff, that focuses on controlling the risk of violence and aggression

4 Indicators of violence in the workplace at Austin Health





% fall

Total number of workplace incidents involving injury




Total number of workplace injury incidents involving lost time




Number of assaults in mental health units




Days lost due to assaults in mental health units




Total number of violent and aggressive incidents per month




Number of violent and aggressive incidents in acute surgical unit*




Number of violent and aggressive incidents in acute medical unit*




* The aggression risk assessment tool was implemented and monitored as part of a project trial in these units.

  • John A Forster1
  • Mark T Petty2
  • Colin Schleiger3
  • Helen C Walters4

  • Austin Health, Heidelberg, VIC.


Competing interests:

None identified.

  • 1. Fry A, O’Riordan D, Turner M, Mills K. Survey of aggressive incidents experienced by community mental health staff. Int J Ment Health Nurs 2002; 11: 112-120.
  • 2. International Council of Nurses. Guidelines on coping with violence in the workplace. Geneva: ICN, 1999. Available at: (accessed Aug 2005).
  • 3. Jackson D, Clare J, Mannix J. Who would want to be a nurse? Violence in the workplace — a factor in recruitment and retention. J Nurs Manage 2002; 10: 13-20.
  • 4. Leather R. Workplace violence: scope, definition and global context. In: Cooper C, Swanson N, editors. Workplace violence in the health sector: state of the art. Geneva: International Labour Office, 2002: 3-18. Available at: (accessed Jul 2005).
  • 5. UK National Health Service. We don’t have to take this. NHS Zero Tolerance Zone resource pack. NHS, 2002. Available at: (accessed Mar 2005).
  • 6. Mayhew C, Chappell D. The occupational violence experience of some Australian health workers: an exploratory study. J Occup Health Safety 2003; 19(6): 3-43.
  • 7. Di Martino V. Workplace violence in the health sector. Country case studies: Brazil, Bulgaria, Lebanon, Portugal, South Africa, Thailand and an additional Australian study. Geneva: International Labour Office, 2002. Available at: (accessed Aug 2005).
  • 8. Harulow S. Ending the silence on violence. Aust Nurs J 2000; 7: 26-29.
  • 9. Whittington R, Wykes T. Aversive stimulation by staff and violence by psychiatric patients. Br J Clin Psychol 1996; 35: 11-20.
  • 10. NSW Health. Zero tolerance: response to violence in the NSW Health workplace. Policy and framework guidelines. Sydney: NSW Department of Health, 2003. Available at: (accessed Aug 2005).
  • 11. Australian Nursing Federation (ANF). Survey shows endemic nature of workplace violence. Aust Nurs J 2002; 10: 6.
  • 12. Lyneham J. Violence in New South Wales emergency departments. Aust J Adv Nurs 2000; 18(2): 8-17.
  • 13. O’Connell B, Young J, Brooks J, et al. Nurses’ perceptions of the nature and frequency of aggression in general ward settings and high dependency areas. J Clin Nurs 2000; 9: 602-610.
  • 14. Henry J, Ginn GO. Violence prevention in healthcare organizations within a total quality management framework. J Nurs Adm 2002; 32: 479-486.
  • 15. Keely BR. Recognition and prevention of hospital violence. Dimens Crit Care Nurs 2002; 21: 236-241.
  • 16. Nabb D. Visitors’ violence: the serious effects of aggression on nurses and others. Nurs Stand 2000; 14(23): 36-38.
  • 17. Atawneh FA, Zahid MA, Al-Sahlawi KS, et al. Violence against nurses in hospitals: prevalence and effects. Br J Nurs 2003, 12: 102-107.
  • 18. Scott H. Violence against nurses and NHS staff is on the increase. Br J Nurs 2003; 12: 396.
  • 19. Walters H, Kay R. Towards a compassionate control strategy. Aust Nurs J 2004; 12(4): 21-22.
  • 20. International Labour Office (ILO), International Council of Nurses (ICN), World Health Organization (WHO), Public Services International (PSI). Guidelines on workplace violence in the health sector. Comparison of major known national guidelines and strategies: United Kingdom, Australia, Sweden, USA (OSHA and California). Geneva: Joint Programme on Workplace Violence in the Health Sector, 2003. Available at: (accessed Aug 2005).
  • 21. Simon RI. Clinical assessment of dangerousness: empirical contributions [book review]. JAMA 2001; 286: 1768-1769.
  • 22. International Labour Office (ILO), International Council of Nurses (ICN), World Health Organization (WHO), Public Services International (PSI). Framework guidelines for addressing workplace violence in the health sector. Geneva: Joint Programme on Workplace Violence in the Health Sector, 2002. Available at: (accessed Aug 2005).
  • 23. Healy P. NHS urged to upgrade training against violence. Nurs Stand 2002; 17(9): 7.


remove_circle_outline Delete Author
add_circle_outline Add Author

Do you have any competing interests to declare? *

I/we agree to assign copyright to the Medical Journal of Australia and agree to the Conditions of publication *
I/we agree to the Terms of use of the Medical Journal of Australia *
Email me when people comment on this article

Online responses are no longer available. Please refer to our instructions for authors page for more information.