Violence in mental health care: the experiences of mental health nurses and psychiatrists

Violence in mental health care:the experiences of mental health nursesand psychiatrists

Peter Nolan PhD RGN RMN

Professor of Mental Health Nursing, School of Health Sciences,

University of Birmingham

Janie Dallender MSc

Research Associate, School of Health Sciences,

University of Birmingham, Birmingham, England

Joaquim Soares PhD

Senior Researcher, Centre for the Development of Health Sciences, Stockholm

Sarah Thomsen MPH

Research Associate, National Institute for Psychosocial Factors and Health,

Stockholm

and Bengt Arnetz MD PhD

Professor of Psychosocial Medicine, National Institute

for Psychosocial Factors and Health, Stockholm, Sweden

Accepted for publication 8 January 1999

NOLANNOLAN PP., DALLENDERDALLENDER JJ., SOARESSOARES JJ., THOMSENTHOMSEN SS. && ARNETZARNETZ BB. (1999)(1999) Journal of

Advanced Nursing 30(4), 934±941

Violence in mental health care: the experiences of mental health nurses

and psychiatrists

Violence against mental health service personnel is a serious workplace

problem and one that appears to be increasing. This study aimed to ascertain the

extent and nature of violence against mental health nurses and psychiatrists,

and to identify what support, if any, they received following exposure to

violence. Mental health staff working within ®ve West Midlands Trusts in the

United Kingdom were surveyed using a postal questionnaire to investigate the

extent and nature of violence they encountered in their daily work. There was

an overall response rate of 47%, which included a response rate for psychia-

trists of 60% (n� 74) and for mental health nurses of 45% (n� 301). Though

both groups experienced violence at work, nurses were found: to have been

exposed to violence signi®cantly more during their career; to have been a victim

of violence within the previous 12 months of the survey; and to have suffered a

violent incident involving physical contact. Whilst a higher proportion of

nurses than psychiatrists received some support following a violent incident, a

large proportion of both groups did not receive any, although most felt in need

of it. The implications of this study for training and management are discussed.

Correspondence: Peter Nolan, School of Health Sciences,

The Medical School, Edgbaston, Birmingham B15 2TT, England.

Journal of Advanced Nursing, 1999, 30(4), 934±941 Experience throughout the nursing career

934 Ó 1999 Blackwell Science Ltd

Keywords: management, mental health nurses, psychiatrists, support,

training, violence, workplace

INTRODUCTION

The possible relationship between violence and mental

illness has long been the subject of interest to researchers,

although many con¯icting conclusions have been reached

(Wessely & Taylor 1991). However, in the mind of the

general public, and in the experiences of some health

professionals, this relationship remains a strong one (Scull

1979). Most mental health professionals encounter various

forms of violence at some time during their working lives.

In the United Kingdom (UK) there has been a tendency

either to ignore such incidents or accept them as an

inevitable part of mental health care. Glennister (1997)

remarks that there is an unquestioned assumption in the

literature that caring always lies at the heart of the mental

health services and consequently issues such as social

control, coercion and violence, are rarely acknowledged or

discussed.

The former General Nursing Council for England and

Wales issued a directive in 1974 stating that all mental

health nurses should be instructed in the care and

management of violent patients but gave no guidance as

to how this could be achieved (General Nursing Council

1974). Violence at work was also addressed by the Health &

Safety At Work Act (1974), in which it was stated that

employers had a statutory duty to render the work

environment safe for employees, especially those who

were likely to encounter aggression and con¯ict.

Little by way of quanti®cation of violence in health care

settings was undertaken until the Health and Safety

Commission (1987) published a report of a survey of

health service employees which asked them to specify the

extent to which they had been involved in violent

episodes. The ®ndings provided, for the ®rst time, an

estimation of the extent of violence encountered by staff in

health services. The survey stated that less than 1% of

those surveyed had suffered major injury requiring

medical assistance, but 11% had suffered minor injuries

requiring ®rst aid, 5% had been threatened with a weapon

and more than 18% had been threatened verbally. Such

was the impact of the report that the Confederation of

Health Service Employees (COHSE) considered suing

health authorities that failed to take appropriate action

to protect the well-being of their employees (Mason 1991).

Commenting on the prevalence of violence in health care,

Evans (1991) stated that it is unacceptable that a vast

health network should offer employees working under

considerable pressure less protection than is provided in

any other industry.

LITERATURE REVIEW

Despite much effort to educate all those working in health

care settings in the prevention and management of violent

incidents, it is now recognized that violence poses an

increasing problem for mental health service employees

(Hansen 1996). Estimating the exact prevalence of

violence remains elusive due to failure to agree on

operational de®nitions (Love & Hunter 1996), various

methodological problems encountered when comparing

mixed concepts and constructs (Wessely & Taylor 1991),

and the wide variation in the way violent incidents are

reported and recorded (Shah et al. 1991). The term

`violence' has been applied to behaviours ranging from

mild verbal abuse to grievous bodily harm, so it is not

surprising that studies report great variation in the

prevalence of violence depending on the settings studied

(Dublin & Lion 1992). Because of this, caution needs to be

exercised when reviewing prevalence rates, as it is highly

likely that published ®gures underestimate the extent of

the problem (Love & Hunter 1996).

Nonetheless, the number and nature of injuries

sustained by nurses as a result of violence has been

estimated to be higher than the number of injuries

reported in what are considered to be high risk industries

such as mining, forestry and heavy construction (Love &

Hunter 1996). Furthermore, there is overwhelming

evidence that nurses are more likely to be physically

assaulted, threatened and verbally abused than any other

health professional group (Whittington et al. 1996).

Weiser et al. (1994) estimated that approximately 10%

of psychiatric patients are violent towards staff. The same

study also reported that 50% of psychiatric staff have been

physically assaulted at some time during their careers.

Ruben et al. (1980) and Madden et al. (1976) concluded

that approximately 50% of psychiatrists had been assaulted

during the course of their work and, in a multinational

survey, Poster (1996) found that 75% of mental health

nurses had been physically assaulted at least once in their

careers. Whittington & Wykes (1994) found that 65% of

nurses in their sample had been violently assaulted by

patients, and Arnetz et al. (1996) concluded that 30% of

Swedish nurses experienced violence at work.

Precursors of violence

Violent behaviour is observed more frequently in some

patients than others, for example in younger patients

(James et al. 1990), those who have been violent prior to

Experience throughout the nursing career Violence in mental health care

Ó 1999 Blackwell Science Ltd, Journal of Advanced Nursing, 30(4), 934±941 935

admission (Conn & Lion 1983), and those with severe

forms of mental illness (Pearson et al. 1986). Other

precursors of violence include confusional states, non-

compliance with medication (Whittington et al. 1996),

short hospital stays in overcrowded wards (Edwards & Reid

1983), poor staff±patient ratios (Yates 1981), increased

rates of admission (Adler et al. 1983), and especially

clients with criminal histories and personality disorders

(Hansen 1996). Martin (1984) reviewed the reports of all

psychiatric hospital inquiries in the UK published during

the 1960s and 1970s and found that amongst the conditions

likely to lead to violent incidents was the practice of

entrusting large numbers of seriously mentally ill patients

to the care of a few poorly trained staff for prolonged

periods in a therapeutically impoverished environment.

Attempting to reduce the levels of violence in mental

health care settings is problematic for, as Cahn (1995) and

Wessely & Taylor (1991) have observed, violence and

mental illness seem to coexist in some patients.

There is some evidence to support the claim that victims

of violence have certain predisposing characteristics

(Arnetz et al. 19961 ), although this has been challenged by

Whittington & Wykes (1994). Support for this claim is based

on the ®nding that approximately 5% of staff are involved

in 20% of reported violent incidents (Hodgkinson et al.

19852 ). Poster (1996) found a signi®cant difference between

those who reported previous assaults and those who did

not, the former being more likely to expect an assault than

the latter. The adoption of authoritarian attitudes, failing to

involve medical staff, poor communication (Brailsford &

Stevenson 1973), demoralization and incompetence (Depp

1983) have been found to be signi®cantly associated with

staff who reported higher rates of assault.

Younger staff, those with limited experience of health

care and especially those who work permanently on night

duty, have been found to be particularly vulnerable to

violence in the workplace (Whittington et al. 1996), as are

those who express dissatisfaction with their work and

those who suffer from work-related musculo-skeletal

injuries (Arnetz et al. 1996). The study by Arnetz et al. also

found signi®cant correlations between violence and the

age of staff, their gender, smoking habits, coffee consump-

tion at work, and their use of alcohol to relax after work.

These factors explained 17% of the variance amongst staff

who reported being victims of violence and 13% of the

variance among those who had been threatened, although

Whittington & Wykes (1994) found the only signi®cant

differences between staff who had been assaulted and those

who had not were age and grade. They also found that staff

who had been assaulted more than once were usually

assaulted by the same patient, indicating problematic

relationships rather than `dif®cult' patients as being an

important factor. Community psychiatric nurses have been

found to be more likely to encounter violent situations than

other nurses (Morning 1994) and Love & Hunt (1996)

concluded that male mental health nurses with managerial

responsibilities were particularly at risk.

Prevention

Managing violence has become a priority for mental health

nurses, although very little attention has been given to a

critical analysis of methods and outcomes. Hansen (1996)

proposed that by identifying hazards and other environ-

mental factors, attacks on and injuries to staff could be

reduced or eliminated. An occupational health perspec-

tive underscores the need for proactive monitoring and

heightens incentives for prevention through the introduc-

tion of external regulation (Love & Hunter 1996). It has

been claimed that training staff in skills to cope with

violence can signi®cantly reduce the incidence of injury

(Infantino & Musingo 1985).

However, it has also been shown that nurses who had

attended a course in managing violence were twice as

likely to be assaulted as those who had not (Fischer 1988).

Attempting to explain this phenomenon, Whittington

et al. (1996) observed that staff who had attended training

sessions on managing violence were more likely to

precipitate violence than non-attendees. Merely focusing

on the techniques of restraint and control in such courses

is insuf®cient. Instead, emphasizing the importance of

therapeutic relationships, client-centredness, negotiation

and collaboration should be central in training sessions

(Harris & Morrison 1995).

DEFINITION OF VIOLENCE

Berkowitz (1989) pointed out that `violence' is a term used

to refer to a variety of different actions, while Buss (1961)

stated that the term subsumed a large number of responses

that varied in topography, energy expenditure and conse-

quences. The operational de®nition of violence accepted

within the context of this study was: `displaying aggres-

sive behaviour, including spitting, scratching, deploying

physical force, or using an object as a weapon, either to

threaten or physically assault'.

AIMS OF THE STUDY

This study aimed:

1 To ascertain the extent and nature of violence against

mental health employees.

2 To identify the support received by staff following

exposure to violence.

P. Nolan et al.

936 Ó 1999 Blackwell Science Ltd, Journal of Advanced Nursing, 30(4), 934±941

METHOD

Setting and subjects

Between November 1996 and March 1997, all quali®ed

and trainee psychiatrists and ®rst level mental health

nurses currently working with mental health clients

from ®ve National Health Service Trusts in the West

Midlands region of England were sent a postal ques-

tionnaire. The sample was drawn from the staf®ng lists

held by the Personnel and Human Resource Managers of

the Trusts. Overall, 47% of mental health employees

returned the questionnaire. The response rate for

psychiatrists was 60% (n� 74) and for nurses, 45%

(n� 301) (Table 1).

The questionnaire

The questionnaire contained 20 items which sought to

elicit the number, type and severity of violent incidents or

threats of violence against staff, and whether support was

offered after such incidents. Items in the questionnaire

had previously been validated by Arnetz et al. (1996).

Procedure

Psychiatrists, trainee psychiatrists and ®rst level mental

health nurses were sent a questionnaire and covering

letter to their workplace via their Personnel or Human

Resource Manager. The letter stated that subjects'

responses would be completely con®dential and that the

West Midlands Trusts would receive a report summariz-

ing the main ®ndings of the study. Stamped addressed

envelopes were enclosed for return of the questionnaire to

the researchers. A reminder was sent to all subjects 4

weeks after the initial mailing of the questionnaire, again

through their Personnel Managers.

Ethical permission was sought at the beginning of the

study from senior managers in the Trusts. Assurances

were given that once the data had been analysed, the

questionnaires would be destroyed. The researchers had

no direct contact with the respondents and were unknown

to them prior to and during the study.

Statistical analyses

The data were stored on the SPSS for Windows program

(version 6á1) and analysed using Pearson's chi-square

statistical test (v2). This is a test frequently used when

examining whether there is a difference between the

responses of two independent groups of respondents, in

this case nurses and psychiatrists, when those responses

are in a categorical form (i.e. yes/no or male/female). All

tests were two-tailed and statistical signi®cance was

assumed at P < 0á05.

RESULTS

Characteristics of the victims of violence

Signi®cant differences were found between employees

who had been victims of violence (or threats of violence)

at work during the course of the last year and those who

had not. Three-quarters of victims were aged 39 years or

under compared to half of non-victims (v2(d.f. 4)� 13á6,

P < 0á01). Most employees in the sample had worked in

mental health care for over 10 years. Yet, a signi®cantly

higher proportion of victims (21%) had only between 6

and 10 years experience compared to non-victims (12%)

(v2(3)� 8á4, P < 0á05). Ten per-cent of victims reported

being satis®ed with their work compared to 21% of

non-victims (v2(3)� 11á3, P < 0á05). Furthermore, 70% of

victims felt they had little control over their work

compared with 55% of non-victims (v2(3)� 14á9,

P < 0á01).

Victims of violence received information regarding their

duties more regularly than non-victims; 8% of victims

said the information was unclear compared to 18% of

non-victims (v2(4)� 11á6, P < 0á05). Victims (25%) had

more feedback from their line manager when they had

done a poor job compared to non-victims (10%; v2(4)�18á6, P < 0á001).

Exposure to violence

Table 2 shows how exposure to violence or the threat of

violence had been experienced by a higher percentage of

senior psychiatric registrars and hospital-based mental

health nurses than other groups working in mental health

care. However, due to the low number of senior psychi-

atric registrars, signi®cance testing was not applied.

Figure 1 shows that just over half of the psychiatrists

reported that they had been exposed to violence during

Table 1 Number of respondents from each professional group

Professional category

Number of

employees

replying

Percentage

of ®nal

sample

Consultant psychiatrists 46 12

Senior psychiatric

registrars

4 1

Trainee psychiatrists 24 6

Community psychiatric

nurses

95 26

Hospital-based nurses 201 54

Unknown category 5 1

Total 375 100



the course of the last year compared to three-quarters of

the nurses.

Nurses were signi®cantly more likely to have been

exposed to violence either `several times' or `very often'

(v2(4)� 21á0, P < 0á001). This pattern was also found for

exposure to violence throughout subjects' careers,

whereby nurses had been exposed to violence signi®cant-

ly more often than psychiatrists (v2(3)� 21á7, P < 0á05).

Figure 2 shows how nurses were signi®cantly more

likely to have been spat at (v2(1)� 19á5, P < 0á001),

scratched or pinched (v2(1)� 20á4, P < 0á001), slapped,

punched or kicked (v2(1)� 22á9, P < 0á001), physically

threatened (v2(1)� 5á3, P < 0á05), or hit by a weapon than

psychiatrists (v2(1)� 5á6, P < 0á05).

Perpetrators of violence

Respondents who had been victims of violence during the

last year were asked to identify their aggressor as `patient',

`patient's family', `colleagues' or òther'. Ninety-six per

cent of nurses and 95% of psychiatrists said their patients

had been aggressive towards them. A signi®cantly higher

proportion of psychiatrists (33%) than nurses (19%)

reported aggression from patients' relatives (v2(1)� 4á0,

P < 0á05). A higher proportion of psychiatrists (11%)

than nurses (6%) reported enduring aggressive behaviour

from their colleagues, although this difference was not

signi®cant.

Severity of violent act

Eighty-six per cent of psychiatrists compared to 50% of

nurses said that they had sustained no injuries as a result

of violence against them. Forty-four per cent of nurses

sustained minor injuries, whilst 6% reported serious

injuries compared to 11% of psychiatrists who sustained

minor injuries and 3% of psychiatrists who sustained

major injuries (v2 (2)� 16á8, P < 0á01).

Sources of support after exposure to violence

Half the nurses and 27% of psychiatrists who had been

the victims of violence during the course of the last year

received support after the incident. Thirty-two per cent

of psychiatrists felt no need for support compared to

17% of nurses (v2 (2)� 8á6, P < 0á05). Table 3 shows that

colleagues were the main source of support for both

groups.

DISCUSSION

The study shows how the frequency of violent incidents

involving mental health employees varies between these

professional groups. Indeed, even within our sample of

nurses, there was a large variation in the exposure to

violence (or threat of violence). A far greater proportion of

hospital-based nurses reported an incidence of violence

within the last 12 months compared with their commu-

nity-based counterparts. Hence exposure to violence (or

threat of violence) against hospital-based nurses in our

study was found to be higher than in previous studies,

such as Poster (1996) and Whittington & Wykes (1994).

Exposure to violence for community-based nurses was

found to be much less, although not as low as that

reported by Arnetz et al. (1996). With the reduction of

Table 2 Number and percentage of subjects exposed to violence

(or threat of violence) during their career by profession

Professional

category

Number of subjects

experiencing violence

% of category

sample

Consultant psychiatrist 20 44

Senior psychiatric

registrar

4 100

Trainee psychiatrist 11 50

Hospital-based mental

health nurse

163 81

Community-based

mental health nurse

47 50

Missing data 7

Figure 1 Act of violence or threat of violence against mental

health nurses and psychiatrists during the course of the last year:

j, nurses; h, psychiatrists.

P. Nolan et al.


inpatient beds, it is conceivable that only those patients

who are acutely ill are hospitalized, thus creating a highly

volatile atmosphere in admission wards. A combination of

highly volatile patients and inexperienced staff may

account for some of the factors involved in causing this

situation.

The incidence of psychiatrists' exposure to violence was

similar to that found in earlier studies (Madden et al.

1976, Ruben et al. 1980). Our study shows the exposure to

be slightly higher for trainee psychiatrists, thus supporting

the inexperience hypothesis, although the number of

trainee psychiatrists was too low to generalize this ®nding.

Moreover, psychiatrists tended to report being exposed to

violence only once in their careers, whereas nurses

reported having been exposed to violence (or threats of

violence) on a more frequent basis.

As in the studies by Arnetz et al. (1996) and Whittington

et al. (1996), our study showed that exposure to violence

was greatest amongst employees who were under 39 years

of age. However, whilst other studies have shown shorter

mental health care experience to be signi®cantly correlated

to workplace violence (Whittington et al. 1996), this study

Figure 2 Type of violence

experienced by mental health

nurses and psychiatrists: j,

nurses; h, psychiatrists.

Table 3 Sources of support

after an act of violence

occurring within the last 12

months Sources of support

No. (%) psychiatrists

receiving support

No. (%) nurses

receiving support

Total (%)

receiving

support

Line manager 1 (1%) 59 (45%) 60 (46%)

Colleagues 7 (5%) 105 (80%) 112 (85%)

Family 4 (3%) 49 (37%) 53 (40%)

Someone outside workplace other

than family

3 (2%) 23 (18%) 26 (20%)

Total 15 236 251

Note. Some respondents indicated that they had received support from more than one source; hence

percentage ®gures do not add up to 100.



found that the highest association was for those who had

worked between 6 and 10 years in mental health care.

Job dissatisfaction was also found to be signi®cantly

correlated to workplace violence, again supporting earlier

research (Whittington et al. 1996). Brailsford & Stevenson

(1973) concluded that under-involvement and poor

communication between staff and patients were signi®-

cantly associated with assault and our study found that

employees who perceived they had less control over their

working situation were also more likely to be victims of

violence. Yet, in this study it was found that those who

had received most information regarding what was

expected of them were more likely to be victims. The

nature of this information was often described by victims

as `negative feedback' on their work performance. This

®nding suggests that it is the speci®c nature of the

communication between line managers and staff rather

than lack of communication that appears to be linked to

incidents of violence. It is unclear whether an incident of

violence causes communication with managers to be

perceived negatively by the victims, or whether an initial

lack of clear and positive guidance predisposes employees

to potentially threatening situations. Our research tends to

indicate the latter.

Most violent attacks against both nurses and psychia-

trists were carried out by their patients. Nurses appear to

be particularly vulnerable. This may simply re¯ect the fact

that nurses spend more time with patients than other

professional groups working in mental health care and are

often involved with them in a more intimate, interactive

manner, especially in acute phases of their illness. There

are therefore more opportunities for patients to vent their

aggression on nurses and, equally, the nature of the

nurse±patient relationship may provide more triggers for

aggressive behaviour.

Psychiatrists in this study reported less exposure to

violent incidents than nurses. This may be due to the

distance they maintain from patients and the tendency

for junior staff to be more intimately involved in care

provision. Yet, psychiatrists are more vulnerable to

aggressive incidents initiated by relatives of patients

and this may re¯ect the fact that relatives see the

consultant as the person with overall responsibility for

the welfare of the patient. Further research into violence

perpetrated by patients' relatives and friends would be

instructive.

CONCLUSIONS

This study makes clear that violence in the workplace is

still a key issue for mental health professionals and merits

the continued attention of researchers, educators and

managers in order to identify the causes of violent inci-

dents, minimize the risk of them occurring and protect both

staff and patients from injury and criminal proceedings.

It is recommended that training to deal with violence in

the workplace be speci®cally targeted at members of staff

who are most likely to be at risk of threats of violence and

actual physical aggression. In our study these were

young, hospital-based nurses with less than 10 years

experience. However, as other studies have shown

community rather than hospital-based nurses as being

more àt risk', we recommend that each employing

authority or Trust identify those most àt risk' within

their own district.

Our study also shows that training should include

management of aggression from patients' relatives and

from fellow colleagues; training should not simply focus

on violent patients but should consider violent and

aggressive acts in general. It is important to note that the

provision of training may be accompanied by an increase

in reported incidences of violence in the workplace, as

seen in the studies by Fischer (1988) and Whittington

et al. (1996). The reasons for this need closer attention

than they have hitherto received.

Training will doubtless prove insuf®cient unless the

underlying causes which make certain members of staff

vulnerable to aggression at the hands of their patients are

explored. The maintenance of high morale in the team,

coupled with the promotion of individuals' self-esteem,

play a signi®cant part in protecting staff from violence and

are linked to effective communication between junior and

senior members of staff. Feedback should not solely focus

on shortcomings in the person's work, but should also

identify, acknowledge and build on the individual's

strengths.

Finally, whilst not every individual who has experi-

enced a violent incident would wish to receive support

from managers or colleagues, this option should be

available for those who seek it and those who may have

no outside sources to which they can turn. Having good

quality support available can be perceived by staff as a

hallmark of the regard the organization has for them.

Moreover, this study showed that support was needed

even when the violent incident did not involve actual

bodily harm. Most psychiatrists felt they needed support

after exposure to threatening as well as aggressive

behaviour. The criteria for eligibility for support should

therefore be broad enough to encompass staff exposed to

non-physical forms of violence.

The 47% response rate from the sampling frame limits

the generalizability of the study. As there were only four

senior psychiatric registrars in the sample, comparisons

between speci®c clinical health professionals could not be

made reliably. It should also be noted that the study was

con®ned to a particular region, and its ®ndings may not be

relevant to other regions. Nonetheless, it is reasonable to

assume that the ®ndings and recommendations discussed

here have implications for other mental health workplace

settings.

P. Nolan et al.


References

Adler W., Kreeger C. & Ziegler P. (1983) Patient violence in a

private psychiatric warden. In Assaults Within Psychiatric

Facilities (Lion J. & Reid W. eds), Grune & Stratton, New York,

131±142.

Arnetz J., Arnetz B. & Petterson I. (1996) Violence in the nursing

profession: occupational and lifestyle risk factors in Swedish

nurses. Work and Stress 10(2), 119±127.

Berkowitz L. (1989) Frustration-aggression hypothesis. Examina-

tion and Reformulation Psychological Bulletin 106, 59±73.

Brailsford D. & Stevenson J. (1973) Factors related to violent and

unpredictable behaviour in psychiatric hospitals. Nursing

Times 18, 9±11.

Buss A.H. (1961) The Psychology of Aggression. John Wiley &

Sons, London.

Cahn T. (1995) Violence Ð an element of the psychiatric

profession. Schweizer Archives Fur Neurologie und Psychiatrie

146(3), 109±114.

Conn L. & Lion J. (1983) Assaults in a university hospital. In

Assaults Within Psychiatric Facilities (Lion J. & Reid W. eds),

Grune & Stratton, New York, pp. 61±69.

Depp F. (1983) Assaults in public mental hospitals. In Assaults

Within Psychiatric Facilities (Lion J. & Reid W. eds), Grune &

Stratton, New York, pp. 21±43.

Dublin W. & Lion J. (1992) Clinician Safety. Report of the APA

Task Force, Report 33. American Psychiatric Association,

Washington, District of Columbia.

Edwards J. & Reid W. (1983) Violence in psychiatric facilities in

Europe and the United States. In Assaults Within Psychiatric

Facilities (Lion J. & Reid W. eds), Grune & Stratton, New York,

pp. 131±142.

Evans R. (1991) It changed my life. Sunday Times, 23 June, p. 53.

Fischer N. (1988) The fear of assault. Bethlem Maudsley Gazette

35, 38±40.

General Nursing Council (1974) Syllabus of Subjects for Exami-

nation for the Certi®cate of Mental Nursing. GNC Publications,

London.

Glennister D. (1997) Coercion, control and mental health nursing.

In The Mental Health Nurse (Tilley S. ed.), Blackwell Science,

Oxford, pp. 43±57.3

Hansen B. (1996) Workplace violence in the hospital psychiatric

setting. An occupational health perspective. AAOHN Journal

44(12), 575±580.

Harris D. & Morrison E. (1995) Managing violence without

coercion. Archives of Psychiatric Nursing 9(4), 203±210.

Health and Safety Commission (1987) Violence to Staff in the

Health Service. Her Majesty's Stationery Of®ce, London.

Hodgkinson P., McIvor L. & Phillips M. (1985) Patient assaults on

staff in a psychiatric hospital: a two-year retrospective study.

Medicine, Science and Law 25, 288±294.

Infantino J. & Musingo S. (1985) Assaults and injuries among staff

with and without training in aggression control techniques.

Hospital Community Psychiatry 36(12), 1312±1314.

James D., Fineberg N., Shah A. & Priest R. (1990) Increase in

violence on an acute psychiatric ward: a study of associated

factors. British Journal of Psychiatry 156, 846±852.

Love C. & Hunter M. (1996) Violence in public sector psychiatric

hospitals. Benchmarking nursing staff injury rates. Journal of

Psychosocial Nursing and Mental Health Services 34(5), 30±34.

Madden D., Lion J. & Penna M. (1976) Assault on psychiatrists by

patients. American Journal of Psychiatry 133, 422±425.

Martin J.P. (1984) Hospitals in Trouble. Basil Blackwell, Oxford.

Mason P. (1991) Violent trends. Nursing Times 87(21), 16±17.

Morning D. (1994) Mental health. Coping with violence: work-

shops for CPNs. Nursing Standard 8(23), 2±8.

Pearson M., Wilmot E. & Padi M. (1986) A study of violent

behaviour among inpatients in a psychiatric hospital. British

Journal of Psychiatry 149, 232±235.

Poster E.C. (1996) A multinational study of psychiatric nursing

staff's beliefs and concerns about work safety and patient

assault. Archives of Psychiatric Nursing 10(6), 365±373.

Ruben I., Wolkon G. & Yamamoto J. (1980) Physical attacks on

psychiatric residents by patients. Journal of Nervous Mental

Disorders 168, 243±245.

Scull A.T. (1979) Museums of Madness. Penguin Books, London.

Shah A., Fineberg D. & James D. (1991) Violence among psychi-

atric inpatients. Acta Psychiatrica Scandinavica 84, 305±309.

Weiser M., Levkowitch Y., Shalom S. & Neuman M. (1994)

Emotional reactions of psychiatric staff to violent patients.

Harefuah 126(11), 642±645, 691.

Wessely S. & Taylor P.J. (1991) Madness and crime: criminology

versus psychiatry. Criminal Behaviour and Mental Health

1, 193±228.

Whittington R. & Wykes T. (1994) Violence in psychiatric

hospitals: are certain staff prone to being assaulted? Journal of

Advanced Nursing 19, 219±225.

Whittington R., Shuttleworth S. & Hill L. (1996) Violence to staff

in a general hospital setting. Journal of Advanced Nursing

24(2), 326±333.

Yates J. (1981) Staff±patient ratios and hospital enquiries. Nursing

Times 26, 2143±2145.



Documents

Violence in mental health care: the experiences of mental health nurses and psychiatrists