Violence is currently prevalent in every sphere of social life. Nowadays, health care personnel are facing more harsh behaviours than ever before, here in Jordan. The rising rate of violence in health care settings has become a major problem for nurses. Nurses are at considerable risk of occupational (work-related) violence. Working primarily in psychiatric departments resulted in an increased risk for both physical assault and non-physical violence (Nachreiner, et al., 2007). Psychiatric health care providers have high rates of work place violence victimization, but yet little is known about the strategies used by them and their facilities to manage, reduce, and prevent violence (Peek-Asa, et al, 2009). Their presence in stressful situations such as incidents (violent incidents), suicide attempts, waiting to visit a doctor, or transfer of patients to another ward or another hospital exposes them to more abuse or harsh behaviour from patients, families, relatives and friends than other hospital staff (Kwak et al., 2006).

The motivation of this paper stemmed from a recently news in the media reporting the increased incidents of violence and aggression faced by nurses in Jordanian hospitals. The media news prompted the author to reflect on current knowledge and understanding of these events in both in Jordan and around the world to make recommendations for managing reducing, and prevention of these events in the future. Recommendations for future research in this area were addressed also. Recommendations for future research will enable nurses to deepen their understanding of violence and aggression in psychiatric settings & which in turn will lead to improved strategies, policy and practice leading to increased safety for nurses and patients. This paper was intended to be a commentary paper on the phenomena of violence in psychiatric settings; however, to comment on this phenomenon an extensive literature review was conducted and will be presented also. The paper design compared the “violence” with the “crime”. The perpetrator of this crime is the psychiatric or mentally ill patient, while the victim is the psychiatric/mental health nurse. The scene where the crime occurred is the psychiatric setting.

The Aim

This paper aims to provide a general understanding of the whole picture of violence against nurses in psychiatric settings. In order to achieve that, this paper addressed the following topics: (1) Recent epidemiology of episodes of violence in psychiatric words, (2) Defining violence and related concepts, types, and forms, (3) The perpetrator, (4) The victim, (5) Prevention of violent incidents, (6) Assessment of violence, (7) Management of violent episodes.

Methods

The following databases were searched: EBSCO host service databases (Academic Search complete, Cumulative Index to Nursing and Allied Health Literature (CINAHL Plus), MEDLINE, Psychology and Behaviours Sciences Collection). These databases were searched for English language papers published between 1 January 2006 and 1 April 2011 using the key words “violen*” (violence or violent) and “in-patient” or “psychiatric words” or “psychiatric settings”.

Limiters were used in each database to include and exclude certain studies. The search was limited to full text articles, available references articles , articles published between 1 January 2006 and 1 April 2011 in scholarly (peer reviewed) journals. Special limiters for Academic Search Complete were periodical publications, English language articles, and articles with PDF full text. Special limiters for CINAHL Plus were articles with available abstract, English language articles, research articles only, articles that considered humans only as research subjects, articles with at least one nurse author, studies conducted in inpatient settings only, and articles with PDF full text. Special limiters for MEDLINE were: articles with abstract available, English language articles, articles that considered humans only as research subjects, articles published in nursing Journals only. Only PDF full text articles were searched in psychology and Behavioural Sciences Collection.

After completing search, 197 studies resulted. Most of them were included in this paper. However, some were not included because they did not respond to the objects of this paper. Some of studies in references lists of the resultant articles were also reviewed and included for epidemiological purposes even they are older than five years. Some of them were also used for critically reviewing the updated studies (≥5 years).

Definitions

Violence in the workplace can take various forms ranging from abusive language, threats, physical assaults, and even homicide (Wassell, 2009). There are many different definitions of violence. This section will define and differentiate between violence forms and forms. The world Health Organization (WHO) define 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, maldevlopment, or deprivation” (WHO, 2005, p.5). Work-related violence is any activity or event occurred in the work environment involve the international use of physical or emotional abuse against an employee, resulting in negative physical and emotional consequences (Nachreiner, et al., 2007). A less restrictive definition was the definition of Baron and Neuman; they define workplace violence as direct attacks which occur in the workplace itself or within an organization (Baron & Neumann, 1998). Physical assault is hitting, slapping, kicking, pushing, grabbing, sexually assaulted, or any type of physical contact aimed to injury or harm (Nachreiner et al.). A threat occurred when someone used words, gestures, or actions for freighting another one without attempting harm or injury (Nachreiner et al.). Sexual harassment occurred when one is a subject for any type of unwanted sexual behaviour (words or actions). (Nachreiner et al.). Verbal abuse is calling another person -must be associated with the name- with unfavourable words for the purpose of hurting & emotionally injuring. Jenkins (1996) believes that even threat of physical violence is considered physical violence (Jenkins, 1996). WHO determined three types of violent acts: physical, sexual, and psychological (WHO, 2005, p.6). Violence and aggression are two interrelated concepts extensively studied in nursing literature. Although, they are not the same, nursing literature widely used them interchangeably. In this paper, violence and aggression will be used interchangeably.

Epidemiology

The risk of being subjected to violence among health staff is 16 times higher than in other occupational groups in the service sector (Kingma, 2001). There is an escalating alarming trend of all forms and types of violence towards nurses in health care settings (Whelan, 2008). Nurses are at the highest rates of nonfatal workplace assault and violent victimization in all health care settings (Lanza, Zeiaa, & Rierdan, 2006). There is a considerable difference in the prevalence and incidence of episodes of violence in mental health settings, depending on the countries in which the studies were carried out. A survey of 4.826 nurses conducted by the American Nurses Association, 17% reported that they have been physically assaulted, and 57% reported that they had been abused in the last year (Peek-Asa, et al., 2009). Also, only 20% reported that they felt safe in their current work environments. Psychiatric nurses are the highest subjects of violent victimization rates of all types of nurses (Islam, Edla, Mujuru, Doyle, & Ducatman, 2003). In an analysis of the results of the Assaulted Staff Action Programme (ASAP) that persists for 15 years; 1.123 mental health nurses (69.58%) were victims of violence by patients. They were the subjects of physical (85.32%), sexual (1.18%), nonverbal intimidation (1.67%), and (6.01%) assaults. 46.34% of injures were soft tissue bruises, 10.16% were head and back injuries, 5.76% were bone/tendon/ligament injures, 12.39% were open wounds, scratches, or spitting incidents, 1.8% were abdominal wounds, and 18.65% were psychological fright. 36.69% were mild injures, 31.52% were moderate, and 14.13% serious and intense (Flannery, Farley, Rego, & Walker, 2007). A survey in psychiatric institutions in Switzerland reported that 70% percent of nurses reported being physically attacked at least once in their career (Needham, et al., 2004). A multiregional study of nursing staff members from acute psychiatric settings showed that 76% of the participants reported that they were assaulted at least once (Peek-Asa, et al., 2009). In a study in Iran, verbal abuse was experienced by 87.4% of nurses during a 6-month period, and physical violence by 27.6% during the same period of time (Shogi, Sanjari, Shirazi, Heidari, Salemi, & Mirzabeigi, 2008). The most frequent and most severe forms of verbal abuse reported were judging and criticizing, accusing and blaming, and abusive anger (Kisa, 2008).

In one Arabic study, a national cross sectional survey was conducted in Kuwait to document the prevalence and determinants of violence against nurses in healthcare facilities. 48% of nurses experienced verbal violence; and 7% only experienced physical harm over the previous six months (Adib, Al-Shatti, Kamal, El-Gerges, & Al-Raqem, 2002). Another Arabic study in Bahrain, Hamadeh and colleagues found the average assault rate on nurses is 4.4%. (Hamadeh, Al Alaiwat, Al Ansari, 2003). No similar studies were conducted in Jordan. However, this high results may be an emergent indicator of the escalating trend in Jordan and other Arabic countries because of their similar Arabic culture to Kuwait and Bahrain. Epidemiological studies were recommended to conduct to determine the incidence and prevalence of this phenomenon in Jordanian health care settings and especially in mental ones.

Despite the high prevalence of violence acts toward nurses, only 20% of violent incidents are reported by nurses. This is due to staff being accustomed to violence; peer pressure not to report; differential reporting based on gender of the victim, fear of blame; excessive paper work; and incomplete or invalid information on reports completed by persons not witness to the event (Crilly, Chaboyer, & Creedy, 2004). This problem should also be addressed in future studies to investigate its causes and to solve it.

The Victim

Nurses are usually the subjects of violence victimization. However, Other mental health care professionals such as physicians and physiotherapists are also at a considerable of violent acts (Stubbs & Dickens, 2009). This paper addressed nurses only as subjects of victimization from psychiatric and mentally ill patients.

Psychiatric patients assaults on nurses victims are a worldwide occupational problem. There is remarkable consistency in victim characteristics over time (Flannery, Juliano, Cronin, & Walker, 2006). Men nurses were exposed to more abuse than women (Shagi, et al., 2008). However, there is an inconsistent finding in the literature identifying whether males or females are more violence-prone (Camerino, Estryn-Behar, Conway, Der Heijden, & Hasselhorn, 2008). The risk of experiencing abuse was higher in nurses with more job experience or who worked more hours (Shagi et al.). On the other hand, a longitudinal cohort study showed that younger nurses with less job experiences are at increased risk violence (Camerino et al., 2008). However, exposure to violence was not significantly associated with age, gender, duration of employment in nurses working in child and adolescent psychiatric units (Dean, Gibbon, McDermott, Davidson, & Scott, 2010). Having a lower job title (air or practical nurse), being in closer contact with patients, having special personality traits, using drug or alcohol, reporting extreme fatigue , may lead to higher risks for aggression and harassment at the workplace (Cooper & Swanson, 2002). Nachreiner et al, agree with Copper & Swanson on that LPNs an increased risk for both physical assault and non-physical violence compared to RNs (Nachreiner et al, 2007). Violent incidents are often related to the low awareness of nurses about the adequate therapeutic communication skills in dealing with patients (Cooper & Swanson, 2002).

Perceptions & attitudes of nurses on patients’ violent incidents in psychiatric settings are extensively studied in the literature. Psychiatric nurses attitudes are different across countries (Jansen, Middel, Dassen, & Reijneveld, 2006). According to some nurses violence is perceived as dysfunctional/undesirable. Whereas in others, violence is perceived as a functional comprehensible phenomenon (Abderhalden et al., 2002). 97% of participants believed that dealing with aggressive behaviour was a part of work in mental health inpatient unit (Dean, Gibbon, McDermott, Davidson, &Scott, 2010). In the same study, 69.7% of participants believed that the current level of physical aggression in the ward was unacceptable, whereas only 12% report that it was acceptable, and the others reported feelings of uncertainty (Dean et al, 2010). They rationalize that by recognize that staff with more positive attitudes exhibited lower state anxiety. There are negative attitudes of nursing students’ to violent incidents, and these attitudes are deteriorated over time (Bowers, Alexander, Simpson, Ryan, & Carr-Walker, 2007). The perception of aggression scale (POAS) is a newly developed attitude inventory assessing nurses’ attitude toward aggressive patients (Palmstierna & Barredal, 2006).

Consequences of violent incidents on nurses were also extensively investigated in the literature. Responses to violence encompassed three major categories relating to physical emotional and professional impact (Dean, et al., 2010). They found that physical injuries divided to: direct injuries from the violent incidents, injuries while implementing restrictive interventions, and physical symptoms such as headache and muscle tension (Dean, et al.). Major physical injuries were on the head, the trunk, the upper and lower extremities (Langsrud, Linakker, & Morken, 2007). Ongoing mental fatigue, stress, shock, helplessness, anger, vulnerability, feelings of being emotionally drained, anxious, impaired sleep and concentration were all emotional responses of nurses as a result of being violent (Dean, et al.). Nurses also respond with the following emotions and behaviours: frustration, despair, hopelessness, substance abuse, absenteeism, retaliation and the development of “I do not care” attitude (Bimenyimana, Poggenpoel, Myburgh, & Niekerk, 2009). The results of verbal abuse or violence by patients, often result in a severe psychological impact in nurses (Inoue, Tsukano, Muraoka, Kaneko, and Okamara, 2006). The most common emotional reaction to violence was anger, followed by shame, humiliation and frustration (Kisa, 2008).

The Perpetrator

Violence is common among individuals entering mental & substance abuse words. Episodes of violence on psychiatric wards have been extensively studied, with one of the main aims being to identify who is more likely to be violent during hospitalization. However, it is difficult to determine that, because violence is a complex behaviour links with a variety of biological, psychopathological, and social factors.

15-years study concludes that older male patients with schizophrenia and younger personality-disordered patients are high-risk assailants (Flannery, Juliano, Cronin & Walker, 2006). Antisocial personality disorder poses a great risk for violence among women than men (Yang & Coid, 2007). Anxiety disorders and any personality disorder are more severe among violent women; alcohol dependence and hazardous drinking are more severe among violent men (Yang & Coid). In a recent literature, Cornaggia and colleagues found that the psychiatric diagnosis most frequently associated with aggressive behaviour is paranoid schizophrenia. As patients with paranoid schizophrenia retain sufficient ability to plan and commit acts of violence related to their delusions (Cornaggia, Beghi, Pavone, & Barale, 2011). Higher levels of hostility-suspiciousness predict the worsening of the pattern of violence (Amore et al., 2008). Persecutory delusions appear to increase risk of violence in some patients; co-occurrence of persecutory delusions and emotional distress may increase the risk of violence (Bjorkly, 2006). Lower social class of origin, offending behaviour in the parental generation loss of the father, a new partnership of the remaining parent, growing up in blended families promoted the development of offending behaviour in general (Stompe, et al., 2006). Past history of violence toward others, & substance abuse disorders are risk factors for future violence (Flannery, Julliano, Cronin, & Walker). Past history remains the most consistent and stable predictor of future violence (Amore et al., 2008). Dual diagnosed patients with substance abuse disorders and bipolar sample have more violent acts (Grunebaum, et al., 2006). Black and minority ethnic are more often perceived as potentially violent to others (Vinkers, Vries, Van Baars, & Mulder, 2009). Internalizing symptoms and affective reactivity contributed to aggression severity more than impulsivity and demographics (Kolko, Baumann, Bukstein, & Brown, 2007). Even the month of birth of patients is considered a risk factor of violent behaviour (Cailhol, et al., 2009).

Repeatedly violent patients had a higher length of residency, a higher number of previous violent behaviours (Grassi, et al., 2006). A past history of head injury with loss of consciousness was more frequent among persistently physically aggressive patients (Amore, et al., 2007). Assessment

Many studies also discussed high risk children in the literature. Aggression appears associated with a wide variety of commonly psychiatric disorders in children (Connor & McLaughlin, 2006). Children of bipolar parents are at high risk of hostility, aggression, violence (Farchione, et al., 2007). Adolescent conduct disorder patients are more likelihood to be violent (Ilomak, Viilo, Hakko, Marttunen, Makkikyro, & Rasanen, 2006). Children with learning disabilities who had a comorbid psychiatry diagnoses reported a significantly higher amount of peer victimization than children without a cormobid psychiatric disease (Baumeister, Sterch, & Geffken, 2007).

Many other results showed results opposed to what known. Foley and colleagues found that violence at presentation with first-episode psychosis is not associated with duration of untreated psychosis (Foley, Browne, Clarke, Kinsella, Larkin, & O`Callagham, 2007). No substantial evidence support the relation between insight and violence risk (Bjorkly, 2006).

The Scene

The scene of violence victimization against the psychiatric nurses by psychiatric and mentally ill patients is the psychiatric settings. The structure of the setting can provoke the manifestation of violence (Steffgen, 2008). Also, environmental design have been demonstrated to deter violence (Wassell, 2009). Inadequate staffing levels and lack of opportunities for clients to participate in therapy may provoke violence behaviour (Sturrock, 2010). The role of uncertainty concerning job stability represent a casual factor Also, lack of job security may cause violent behaviour The absence of social support and co-workers increases the risk of nurses in this setting to physical and verbal violence (Steffgen). The occurrence of workplace violence may cause damage to both the individual and the institutions. Organizations may face increased absenteeism, sick leave, property damage, decreased performance and productivity, security costs, litigation, worker’s compensation, and increasing turnover rates (Jackson, Clare, & Mannix, 2002).

Prevention

Previous nursing literature suggested a number of strategies that can be considered by nurses to prevent violence. There is limited research on effective interventions to prevent patient violence (Kling, Yassi, Smailes, Lovato, & Koehoorn, 2010). However, failing to accept and implement preventive measures in psychiatric settings has an impact to reduce violence in these settings (Wassell, 2009).

Improved reporting may be of big benefit of reducing physical violence (Nolan & Citrome, 2007). This may be occur by early recognition and intervention of potentially occurring violent incidents in the future.

After conducting the Violence Prevention Community Meeting (VPCM), a significant decrease in patients violence were found across day, evening and night shift for pre-treatment vs. treatment and pre-treatment vs. post treatment comparisons. VPCM is a semi-structured protocol for the purpose of violence prevention (Lanza, Rierdan, Forester, & Zeiss, 2009).

Early recognition has strong practical implications for psychiatric nurses by helping them to assist patients with the detection of early warning signs. Early recognition is pay special attention to the early social and interpersonal factors that may deteriorate the patient behaviour to violent one (Fluttert, Meijel, Webster, Nijman, Bartels, & Grypdonck, 2008).

Steffgen identified many preventative measures of workplace violence such as: measures concerning the physical environment, measures concerning the management of the organizations and the behaviours of the members in the organizations, policies, counselling and training measures (Steffgen, 2008).

A 6-module program have been shown to be effective in reducing and preventing violent incidents in a 6-months evaluation period. The 6 modules were about violence risk assessment, theoretical models of violence, assertive training, ethical & legal issues of violence management (Anderson, 2006).

Dubin et al identified six gold recommendations to prevent violence incidents in psychiatric emergencies. Firstly, all newly admitted patients should be assessed for risk of violence; those who have risk factors should be continually assessed. Second, nurses should avoid evaluating and/or treating patients at risk for violence alone or in an isolated office. Third, nurses must remember that patient’s violence is a response to feelings of helplessness, passivity, and perceived or actual humiliation; therefore nurses should avoid becoming verbally or physically towards them. Fourth, nurses are supposed to use non-coercive methods such as de-escalation to prevent escalation of patients’ aggression. Fifth, limit setting should always offer the patient two options with one option being the preferred option. Sixth, an armed patient should not be threatened and the clinician should respond in a non-threatening manner offering help and understanding. Finally, evaluation of environment safety should occur periodically and changes should be implemented that will enhance safety (Dubin, Julius, Novitsky, & William, 2009).

Assessment

The first step in mental nursing process and one of the most important duties in psychiatric settings is assessment. Psychiatric nurses are faced with a great number of situations in which risk assessment are needed. Risk assessment is a process concerned with a variety of issues —risk for what, when, where, and to whom-not just the were “Prediction” of future violence (Haggard-Grann, 2007). Predicting violence has been compared to forecasting the weather. Like a good weather forecaster, the nurse does not state with certainty that an event will occur. Instead, he/she estimates the likelihood that a future event will occur. Like weather forecasting, predictions of future violence will not always be correct (Scott & Resnick, 2006). Three major types of violence risk assessment are extensively reviewed in the literature. The three types are: (1) Clinical violence risk assessment, (2) Structural risk assessment tools, (3) Functional assessment.

The number of risk assessment instruments has increased in the recent years (Haggard-Grann). Risk assessment tools should include situational aspects, behavioural patterns, and predicted events or stressors (Haggard-Grann). The first step when determining which instrument to use for a specific risk assessment is to determine the purpose and context for which the instrument is needed (Haggard-Grann). Decision should be made regarding whether the assessment is for the first time (to separate the highly risk patients from others) or for continuous ongoing assessment (Haggard-Grann). If adopted in clinical practice with a professional way, these instruments will indeed aid in the assessment and early recognition of violent incidents. However, they are inescapable part in the clinical practice in psychiatric settings. (Haggard-Grann). Awareness of the limits and abilities of such instruments is required. Lurigio and Harris underscored the importance of performing more accurate assessment tools that can for example determine the upcoming type of violence, or the likelihood of weapons use (Lurigio & Harris, 2009).

A risk assessment tool mainly contains two types of factors: static and dynamic. Dynamic factors are of a great importance in a decision context whereas static factors are at less importance. Dynamic factors should be assessed regularly in a structured time schedule (ex. every 1 hr). (Haggard-Grann). Static variables are based on intrapersonal factors (ex., personal & biological characteristics) that served as risks factors for a patient to be potentially violent in the future. (Haggad-Grann).

Many violent risk assessment tools were developed for the aim to assess the violent incidents in psychiatric settings.

Sexual Aggression scale is an effective assessment tool to record systematically the occurrence of sexually aggressive behaviours for patients who reside in psychiatric hospitals. (Jones, Sheitman, Hazelrigg, Camel, Williams, & Paesler, 2007). It is a brief scale consists of 4 sub scales with a brief description of them.

The Alert System is a system includes a risk assessment form used by nursing staff to assess patients upon admission to the psychiatric setting in order to identify these at an increased risk of violence (Kling, et al ., 2010). If identified as at risk for violence, a flag is placed on the patient’s chart and wristband to contain staff of a patient’s potential for violence (Kling et al). The warning is intended to allow workers to take precautions to prevent violent incidents in flagged patients. These precautions may include: wearing a personal alarm, being near a security personals, not having sharp objects in the patient’s room, and not entering the patient’s room alone (Kling et al). Study results indicate that the Alert System is effective in identifying potentially violent patients. However, the ultimate goal of implementing the Alert System is to reduce the risk of violent incidents (Kling et al).

Risk for in-patient violence in acute psychiatric intensive unit can be a high degree be predicted by nurses using the Broset violence checklist (Bjorkdahl, Olsson, & Palmstierna, 2006). The BVC is a method to predict risk for violence from patients within the coming 24 hrs in acute psychiatric inpatient settings (Bjorkdahl, Olsson, & Palmstierna). BVC is used to assess the patient three times daily: in the morning, noon, and night shifts (Bjorkdahl, Olsson, & Palmstierna). The BVC assess absence or presence of six behaviours: confusion, boisterousness, irritability, verbally, threatening, physically threatening and attacking object (Bjorkdahl, Olsson, & Palmstierna).

The HCR-20 is a structured professional checklist designed for the assessment of risk future violence in patients with violent history / or a major mental disorder or personality disorders. (De Vogel & De Ruter, 2006). The HCR-20 consists of 20 items, divided into three subscales: historical scale, clinical scale, and risk management scale. The predictive validity of the HCR-20 was good (De Vogel & De Ruiter).

The Forensic Early Warning Signs of Aggression Inventory (FESA) was developed to assist nurses and patients in identifying and monitoring early warning signs of aggression in forensic patients (Fluttert, Meijel, Leeuwen, Bjorkly, Nijman, & Grypdonck, 2011).

The Maudsley Violence Questionnaire contains 56-items measure a number of cognition (including: beliefs, rules, distortions and attributions) that are related to violence (Warnock-Parkes, Gudjonsson, & Walker, 2007).

The Psychopathy Checklist (PCL) is a clinical rating scale designed to measure psychopathic attributes in mentally ill patients, Patients who score higher have higher rates of violent recidivism (Scott & Resnick, 2006). The PCL uses a semi-structured interview, case-history information, and specific criteria to rate each of 20 items on a three- point scale (0, 1, 2). (Scott & Resnick). Total scores (ranging from 0 to 40) reflect an estimate of the degree to which the patient matches psychopathy (Scott & Resnick).

The Violence Risk Appraisal Guide (VRAG) is a risk assessment instrument of 12 items. It is probably the most well-known assessment instrument aimed to assess dangerousness in high-risk mentally ill patients. It is used to appraise the violence risk in psychiatric and other health settings (Scott & Resnick, 2006). It is constructed by taking variables known to predict violent behaviour among men with mental disorders who have records of previous violent behaviour then summarizing the variables into one scheme (Haggard-Grann, 2007).

Interactive Classification Tree is a recent tool for assessing the violence risk of patients discharged from psychiatric facilities (Scott & Resnick, 2006). This tool utilizes a sequence of questions related to risk factors for potential violence (Scott & Resnick). According to the answers, another related question is posed, until the pt is classified into a category of high or low risk of future violence (Scott & Resnick).

Structured risk assessment tools have inherent limitation when used alone. Criticisms of instruments include the following: they provide only approximations of risks; their use is not generalizable beyond the studied populations: they are rigid, and they fail to inform violence prevention & risk management (Scott & Resnick, 2006).

Functional assessment approaches seek to clarify the factors responsible for the development, expression and maintenance of problem behaviours. This is achieved through assessment of the behaviour of interest, the individual’s predisposing characteristics, and the antecedent events, considered important for the initiation of the behaviour, and the consequences of the behaviour, which maintain and direct its developmental course (Daffern, Howells, & Ogloff, 2007). They identify 9 common functions of violent behaviour in psychiatric settings in the literature: demand avoidance, to force compliance, to express anger, to reduce tension, to obtain tangibles, social distance reduction (attention seeking), to enhance status or social approval, compliance with instruction, to observe suffering (Daffern, Howells, & Ogloff). Functional assessment have many implications for the prediction and prevention of inpatient violence and for the treatment of violent patients. The distinction of functional assessment approaches and structured assessment tools is that the first emphasize the correct classification of the form of a particular behaviour and the other one emphasize the purpose of the behaviour (Daffern, Howells, & Ogloff).

The clinical risk assessment method is the oldest method of violence risk assessment. It is the classical method of expecting, predicting, and assessing of risk. This means that the nurse gathers the information that he or she believes to be useful and on the basis of that information makes a judgment of the risk (Haggad-Grann, 2007). Unfortunately, this method cannot predict future violence with high accuracy. The accuracy of a