Dangerousness and mental health: the facts

Please give feedback on this information

Copyright note for Mind factsheets: Both individuals and organisations are welcome to print and photocopy any complete factsheet from the 'Information' section of Mind's website. Organisations are free to distribute them to service users and colleagues, but must ensure they always use the latest version of the factsheet, as available on the website, at the time of distribution.

Introduction
Key facts about violence
Definition of 'Dangerousness'
Background history
Violent behaviour and mental disorder
Homicide
Suicide
Personality disorders
Media Influence
Race and the media
Risk assessment
Relationships with victims
The influence of alcohol and drugs
Violence directed towards people with mental health problems  

Introduction

The aim of this factsheet is to explore the nature of dangerousness in all its forms. The most common perception of dangerousness is in the form of one person presenting a danger to others; but this is far from being the only form in which dangerousness presents itself.

More often than not, dangerousness presents in the form of people being a danger to themselves, whether this be through suicide or deliberate self-harm.

As well as exploring the differing forms of dangerousness, this factsheet looks at different perceptions of dangerousness, and how literature and the media can influence these perceptions.

It also seeks to explore the way in which public perceptions of dangerousness can impact upon different groups of people, particularly people with mental health problems and people from minority ethnic groups. The effects of such stigmatisation can be extremely negative, and in some cases has even led to people from these groups becoming the victims of violent crime themselves.

Finally, this factsheet will look at research into dangerousness, looking at the statistical evidence, potential links with mental health problems, relationships with victims, indicators of propensity towards violence, and at possible predictors of dangerousness.

Note on Language:
The terminology of psychiatric diagnosis which is used in this factsheet reflects the language of the sources referred to. The use of such language in no way implies Mind's unqualified acceptance of it.

Back to top

Key facts about violence

• Out of 1,564 cases of people convicted of homicide in England and Wales in the three years between April 1996 and April 1999, 164 (10 per cent) were found to have symptoms of mental health problems at the time of the offence. [1]
• In 2005, the total population in England and Wales was just over 53 million. It has been estimated that around 1 in 6 of the population will at any one time suffer from a significant mental health problem. [2]  This means that around 8.8 million people in England and Wales suffer from a significant mental health problem at any one time. Comparing this figure with the 50-60 cases of homicide a year involving people with mental health problems [3], the data does not support the sensationalised coverage about the danger people with mental health problems present to the community.
• The vast majority of violent crime and homicides are committed by people who do not have mental health problems. In fact, 95 per cent of homicides are committed by people who have not been diagnosed with a mental health problem. [4]
• Contrary to popular belief, the incidence of homicide by people diagnosed with mental health problems has stayed at a fairly constant level since the 1990s at between 50 and 60 a year. [5] 
• The fear of random, unprovoked attacks on strangers by people with mental health problems is unjustified. This has been highlighted by an American finding that patients with psychosis, living in the community, are 14 times more likely to be the victims of a violent crime than to be arrested for such a crime. [6]  
• According to the British Crime Survey, close to 50 per cent of all victims of violent crimes believed that their offender was under the influence of alcohol. Around 18 per cent of victims believed that the offender was under the influence of drugs. Around 30 per cent of victims believed that the offender attacked them because they were under the influence of drugs or alcohol. In contrast, only one per cent of victims believed the violent incident happened because the offender was suffering from a mental illness. [7]  
• People with mental health problems are more dangerous to themselves than they are to others. 90 per cent of suicides in the UK are committed by people in mental distress. [8] 

Back to top

Definition of "dangerousness"

Although it is an extremely emotive term, used widely in the mental health field and in the media, there is no consensus on the meaning of the term "dangerousness" [9]  Dangerousness has variously been described as:

"an unpredictable and untreatable tendency to inflict or risk serious, irreversible injury or destruction, or to induce others to do so" [10]

"a propensity to cause serious physical injury or lasting physical harm" [11]

Gunn states in his article 'Defining the Terms' that "the term dangerousness is made up of three elements - destructiveness, prediction and fear. The latter, fear, makes it at least partially subjective, therefore it can never be entirely objective." [12]  Prediction is also highly subjective.

Because of the element of subjectivity involved in the use of the term "dangerousness" and its element of fear, the term, and indeed the concept, can be misused or misapplied by those whose intention it is to over-estimate the risks involved in community care for people with mental health problems.

Back to top

Background history

In recent history, people with mental health problems were housed in large institutions, secluded from the rest of the community. However, since the early 1950s psychiatric hospitals have been slowly running down. With the advent of the 1980s large-scale closures of psychiatric hospitals, as part of the Government's policy of care in the community, the country has seen large numbers of former psychiatric patients discharged into the community.

The main intention of de-institutionalisation was to increase the freedom of people with mental health problems by promoting the principle of "least restrictive alternative". This "least restrictive" philosophy underpins the principles of care outlined in the Reed Report. [13]  This report states that care arrangements for people with mental health problems should: "have proper regard to the quality of care and the needs of individuals; as far as possible, in the community, rather than in institutional settings; under conditions of no greater security than is justified by the degree of danger; so as to maximise rehabilitation and the chances of sustaining an independent life; as close as possible to their own homes and families."

The main recommendation of the Reed report was that " …mentally disordered offenders should, wherever possible, receive care and treatment from health or social services rather than in the criminal justice system" and that "…we see community services as providing wherever possible for the majority of mentally disordered offenders".

Until 1981, inquiries had addressed scandals in hospitals on behalf of an angry public, almost invariably showing patients as helpless victims and staff as abusers. In July 1984, a shocking event had major repercussions for mental health care, and eventually, public attitudes. Sharon Campbell, a former inpatient, killed Isabel Schwartz, her former social worker, at Bexley Hospital. From then on, community care became associated with danger. [14] 

In 1992, there were a cluster of tragic cases involving people with mental health problems: Michael Buchanan, Christopher Clunis and Erhi Inweh. The publicity around these events has created a sense of danger - a sense that random, irrational acts of violence are on the increase. However, this is far from being the case.

In relation to homicides there is a common perception that all people with mental health problems are likely to behave in a violent way - this view being supported time after time by films, by novels and by the media. The facts demonstrate how exaggerated and unfair this view is. A confidential inquiry into homicides and suicides by mentally ill people shows that in one year, 1989-90, there were 235,100 admissions to inpatient psychiatric units in England of whom 16,890 were formally detained under the Mental Health Act 1983. Whilst in this same period, there were 525 homicides in England and Wales of which only a small proportion would have involved psychiatric contact. In the report Dr. William Boyd states: "If we then look at the small numbers dealt with we can gain some perspective of the size of the problem." [15]  It is also worth keeping in mind that many cases of violence in the community get reported three times - the event, the court case, and the inquiry report - thus greatly exaggerating the number of cases in the public's mind.
 
Back to top

Violent behaviour and mental disorder

Press coverage focusing on crimes of violence committed by people with a diagnosis of schizophrenia has led some commentators to seek to establish a link between psychosis and violence. Although there is research suggesting a modest link between psychosis and violence, it emphasises the fact that the majority of such crimes are associated with drug and alcohol abuse.

The public perception is that community care policy has failed and that there are now more people with mental health problems on the streets. Many believe that this means an increased risk of being harmed by somebody diagnosed with a mental health problem. A report by the Audit Commission [16]  points out that most people with schizophrenia live relatively normal lives in the community and the risk to the public has actually decreased since the community care reforms. The report cites evidence that the number of homicides by people with mental health problems has not increased, while the number committed by others has more than doubled.

A study based on Home Office figures shows that the majority of homicides are not linked to care in the community. [17]  Contrary to popular belief, the number of homicide convictions of people considered to be mentally disordered has fallen to half that reached in 1979 - before the rush to close old asylums. Compared with all killings, the number committed by people with mental health problems has fallen even faster. The proportion has dropped from almost half in the 1960s, to little more than one in ten today. Although homicide convictions have multiplied fivefold since the late 1950s to more than 500 annually, the number involving a mentally disordered offender has fallen to around 60.

According to the psychiatrists who carried out the study, the likelihood of someone being killed by somebody with a mental disorder is probably less than that of winning the National Lottery outright. Even then, victims are likely to be someone known to the killer, rather than a stranger. Although people today are at slightly increased risk of being killed by a stranger, according to the psychiatrists, that person is highly unlikely to have a mental disorder. [18]

The MacArthur Foundation's Community Violence Risk Study was set up to compare the rate of violence by former mental patients with the rate of violence by other members of the community. [19] The study found that:

• People with a major mental disorder diagnosis and without a substance abuse diagnosis are involved in significantly less community violence than people with a co-occurring substance abuse diagnosis.
• The prevalence of violence among people who have been discharged from a hospital and who do not have symptoms of substance abuse is about the same as the prevalence of violence among people living in their communities who have not spent time in psychiatric hospitals and who do not have symptoms of substance abuse.
• The prevalence of violence is higher among people (discharged psychiatric patients or non-patients) who have symptoms of substance abuse. People who have been discharged from a psychiatric hospital are more likely than other people living in their communities to have symptoms of substance abuse.
• The prevalence of violence among people discharged from a psychiatric hospital and who have symptoms of substance abuse is significantly higher than the prevalence of violence among other people living in their communities who have symptoms of substance abuse, for the first few months after discharge.
• Violence committed by people discharged from a hospital is very similar to violence committed by other people living in their communities in terms of the type of violence (hitting), the target of violence (family members), and the location (at home).

Back to top

Homicide

The National Confidential Inquiry into Homicides and Suicides by People with Mental Illness investigated 718 cases of people convicted of homicide notified to the Inquiry in the 18 months from April 1996.

The Safer Services report [20] presents information on 500 (70 per cent of the total sample) of these individuals based on psychiatric reports prepared for the courts.

• 14 per cent (71 individuals) were found to have had symptoms of mental distress at the time of the offence.
• 3.7 per cent (27 individuals) had symptoms of psychotic illness.
• 44 per cent of individuals were reported to have a history of mental disorder, however, the majority of these were alcohol or drug dependence and personality disorder rather than conditions usually regarded as severe mental illness.
• 17 per cent of cases investigated were found to have had contact with psychiatric services at some point in their lives.
• 9 per cent were found to have been in contact with mental health services in the year before the homicide.

Compared to those without mental health problems, those with symptoms of major mental illness at the time of the offence had a lower rate of previous conviction for violent offences, and alcohol and drugs were less likely to have contributed to the offence. They were less likely to have killed a stranger or acquaintance and more likely to have killed a family member - usually a spouse.

The Confidential inquiry found that 10 per cent of victims of homicide by people with mental health problems were strangers compared to 26 per cent of victims of people without mental distress. 82 per cent of victims of homicide by people with mental health problems were family members, compared with 25 per cent of victims of homicide by people without mental health problems.

Alcohol and drug use is another contributory factor to violence. In 28 per cent of homicides by people with mental health problems notified to the National Confidential Inquiry, alcohol or drug use was thought to have been a significant contributory factor.

Professor Louis Appleby, Professor of psychiatry and Director of the Inquiry said:
"The public fear of mental illness is partly based on the belief that the mentally ill are a major risk to strangers. Our report shows that most killings are in fact committed by people without mental illness. In homicides by people with mental illness, the victims are usually family members; both patients and their families deserve the full support of the mental health services."
 
Back to top

Suicide

In most Western countries close to 90 per cent of those who die by suicide have a mental disorder. In some Asian countries this proportion is much smaller, but there is still a strong link between mental illness and suicidal behaviour. [21] More than 60 per cent of those who die by suicide are thought experience a depressive illness at the time of the suicide. Bipolar disorder, alcohol and substance and schizophrenia are also linked to higher risk of suicide.

However, it is important to be aware that the vast majority of people with a mental disorder will never commit suicide. [22] 

A British study found that among people with no mental disorder, around 10 per cent had suicidal thoughts at some time in their life. [23] Around 2 per cent of people with no mental disorders had attempted suicide. The study showed that among people with mental disorders the rates were much higher. As expected, the researchers found that people who had experienced a depressive episode had high rates of suicidal thoughts. Around 52 per cent had a lifetime prevalence of suicide and 25 per cent had attempted suicide at some stage in their life. However, people with OCD had the highest rate of suicidal thoughts with a 64 per cent life-time prevalence.

The most common methods of suicide in men are hanging, strangulation and suffocation (44 per cent). Drug related poisoning accounts for 20 per cent of suicide in men, while 10 per cent use 'other poisoning' including motor vehicle exhaust gas. The three most common methods of suicide in women are drug related poisoning (46 per cent), hanging, strangulation and suffocation (27 per cent) and drowning (7 per cent). [24]

Back to top

Personality disorders

Epidemiological research conducted over the past twenty years [25] has shown that people with personality disorders (particularly those diagnosed with antisocial personality disorder) have a wide range of psychosocial problems which include:

• early unnatural death through higher rates of suicide and accidents
• high rates of associated mental illness and other problems - in particular, substance abuse, eating disorders, depression and anxiety
• worse outcomes for the treatment of mental and physical illness leading to high service utilisation and the "revolving door" phenomenon
• high rates of family disharmony and violence
• high rates of crime
• high rates of unemployment and homelessness.

The question of whether personality disorders are "treatable" or not has raised issues in relation to the Mental Health Act (1983). The Scoping Review of Literature on the Health and Care of Mentally Disordered Offenders states that "The 'treatability' criteria in the legislation mean that many of the most seriously disturbed people do not have their offending addressed but are simply incarcerated then released." [26] 

Back to top

Media influence

Public attitudes to, and knowledge about, mental illness are largely informed by the media. Coverage has created a climate of fear by associating mental illness with violence, and focusing on "unusual and tragic" conditions. [27]

Using some of the material from their research, the Glasgow Media Group explored the impact on the beliefs of an audience sample. Two-fifths of their general sample believed mental illness to be associated with violence, and they gave the media as their source. Their findings indicate the importance of working with the media to de-stigmatise mental health problems. Negative media coverage entrenches the stigmatisation of mental distress, and inhibits both the readiness of people with mental health problems to seek help, and their acceptance by local communities. [28] 

Back to top

Race and the media

Race and mental health, separately, are controversial issues. The two issues combined, contain extremely negative connotations in the public imagination. "Black and dangerous" is a popular media image which reinforces racist stereotypes of black people being "mad" and "bad".

Mind's survey, 'Raised voices' - on African-Caribbean and African user's views of mental health services in England and Wales - found that 26 per cent of respondents felt that the media bears some responsibility for some of the discrimination they experience. They felt that clear guidelines on the reporting of stories relating to black people are required. [29]  A further 17 per cent cited public fears about black people as a cause of discrimination.

Back to top

Risk assessment

In her article 'Prediction of Dangerousness' (1995), Dilys Jones states that: "dangerousness is difficult to define and measure objectively, but assessment of risk is a central function of the forensic psychiatrist. Any assessment of risk ideally takes both a clinical approach (based on information regarding an individual) and an actuarial approach (based on statistics and population studies)."

Mind has concerns about clinical risk assessments where the assessment is not rooted in clear evidence. There are clear human rights implications around the power which clinical judgements can have over individuals. Predictors of violence can be inaccurate and psychiatrists tend to overestimate violent behaviour. Previous behaviour is the best predictor of future behaviour. The degree of dangerousness changes with internal characteristics of the client (eg mental illness) and external (environmental) factors. [30]

Professor John Gunn writes that: "There are a number of studies which have shown that if a large population, whether or not they are considered to be mentally abnormal, is divided into those who have committed previous acts of violence and those who have not, the group with the previous history of violence will generate more of the future violence than the other group." [31]

Gunn goes on to say that: "whilst statistics can be extremely powerful in describing and perhaps predicting group activity, they cannot tell us which members of the group will do what." Therefore, such statistical information cannot be used as a predictor for individual circumstances.

The National Confidential Inquiry into Suicide and Homicide found that only half the Health Trusts in England and Wales provide staff with training on the assessment of suicide risk and risk of harm to others, and a minority had written policies on the communications of risk estimations. [32] 

Risk assessments can be influenced by conscious or unconscious racism, sexism, ageism or anti-gay or lesbian prejudice.

Black people, particularly young black men, are more likely than white people to be compulsorily admitted to psychiatric hospital, to be kept on locked wards and to receive higher doses of medication than white people. The use of sectioning and the involvement of the police are more likely in the case of young black men. This may be on a person's medical records and may influence professionals making subsequent risk assessments. [33]

A study conducted for the Campaign for Racial Equality and the Mental Health Act Commission found that 75 per cent of all professionals interviewed concurred with the view that black clients were more likely to be perceived as dangerous. Approved Social Workers thought that their own profession, in general, tended to take greater precautions when dealing with black clients. [34]

In a letter published in the Psychiatric Bulletin, Professor Mike Crawford suggests that he is not aware of any instruments that can accurately identify individuals at high risk of committing homicide. Further, he suggests that if we use the instruments that have been devised to detect violent incidents, we are likely to find that for every person identified correctly, 5,000 people might be identified as being at high risk of committing a homicide but will never do so. [35]

Back to top

Relationships with victims

There is a common misconception that homicides by people with mental health problems tend to be random, unprovoked attacks on complete strangers. Although this type of attack does occasionally happen, by far the majority of victims are family members, or otherwise acquainted with the aggressor.

The National Confidential Inquiry into Suicide and Homicide by People with Mental Illness highlights the fact that in only 10 per cent of cases of homicide by people with mental health problems was the victim a stranger, and in only 8 per cent was the victim an acquaintance. [36]  54 per cent of those killed by people with mental health problems were the spouses or partners of the perpetrator.

Back to top

The influence of alcohol and drugs

The National Confidential Inquiry into Suicide and Homicide by People with Mental Illness found that among perpetrators of homicide, the group diagnosed with mental health problems had a lower rate of alcohol or drug misuse compared with the group without any such diagnosis (46 per cent compared with 56 per cent). The Inquiry found that alcohol and drugs were less likely to have contributed to the offence in the case of people with mental health problems; 28 per cent, compared to 60 per cent in the group with no diagnosis.

The prevalence of violence in people who meet the criteria for a diagnosis of alcoholism is twelve times higher than that of the general population, and the prevalence of violence in people who meet the criteria for being diagnosed as abusing drugs is sixteen times that of the general population. [37] 

Back to top

Violence directed towards people with mental health problems

Far from being violent, dangerous offenders, most people suffering from mental distress are, in fact, much more likely to become victims of violence. This issue is highlighted in Mind's 'Not just sticks and stones' survey of the stigma, taboos and discrimination experienced by people with mental health problems. [38] The survey found that almost half (47 per cent) of people with mental health problems had been abused or harassed in public, and that 14 per cent had been physically assaulted. A quarter (25 per cent) of people felt they were at risk of attack inside their own homes, while over a quarter (26 per cent) of people were forced to move home because of harassment. When asked the question; "Who, or what, do you blame for this discrimination?" 66 per cent of people blamed a lack of public education, 60 per cent blamed media stories, 49 per cent blamed personal prejudice, 48 per cent blamed fears, and 22 per cent blamed politicians.

It is ironic that many women in mental distress do not feel safe in psychiatric hospitals and other mental health settings. In 1993, Mind's 'Stress on women' campaign highlighted the problems women can face in mixed hospital wards. Staff, visitors or other patients can be guilty of harassment and abuse. It may take the form of outright abuse through violence, or some more subtle form of coercion; for instance, use of threats or bribes to obtain sex, abuse of a position of trust by a professional or constant unwanted sexual remarks or looks. [39] 

Mind did another survey on the situation in psychiatric wards in 2002, and found that patients were still subject to sexual harassment and sexual abuse.

And yet again, in 2004, Mind's 'Ward watch' campaign found that many women still feel unsafe in psychiatric hospitals. A staggering 51 per cent of recent or current inpatients reported that they had been verbally or physically threatened during their stay in hospital. As many as 20 per cent said they had been physically assaulted. The Government has promised to phase out mixed sex wards, but Mind's survey shows that 23 per cent still report being on mixed sex wards while in psychiatric hospitals. [40]

Written by George Stewart, 1998. Updated by Inger Hatloy, September 2006.
 
[1] Department of Health 2001, Safety First, Report of the National Confidential Inquiry (NCI) Into Suicide And Homicide By People With Mental Illness, Department of Health.
[2] ONS 2000, Psychiatric morbidity among adults living in private households in Great Britain
[3] National Confidential Inquiry into Suicide and Homicide by People with Mental Illness - Annual figures August 2005 
[4] Kings College London, Institute of Psychiatry, Risk of violence to other people, Available http://www.mentalhealthcare.org.uk/content/?id=54 September 2006
[5] Dr Oscar Hill 2003, How much is violence associated with mental illness?, Royal College of Psychiatrists, Part of Anti-Stigma Campaign 'Changing Minds: Every Family in the Land', available: http://www.rcpsych.ac.uk/campaigns/changingminds/whatisstigma/violence.aspx
[6] Walsh E. et al 2003, 'Prevalence of violent victimisation in severe mental illness', The British Journal of Psychiatry 183: 233-238
[7] Coleman K., Hird C. and Povey D. 2006, 'Violent Crime Overview, Homicide and Gun Crime 2004/2005', Home Office Statistical Bulletin, available: http://www.homeoffice.gov.uk/rds/pdfs06/hosb0206.pdf
[8] Hall D. et al 1998, 'Thirteen-year follow-up of deliberate self-harm, using linked data', The British Journal of Psychiatry 172: 239-242
[9] Jones D. 1995, 'Prediction of Dangerousness', in Management of Violence and Aggression in Health Care, eds. Kidd, B. and Stark,C., Gaskell .
[10] Scott P.D.1977, 'Assessing Dangerousness in Criminals', British Journal of Psychiatry, 1977; 131:127-142
[11] Butler Report 1975, 'Report of the Committee on Mentally Abnormal Offenders', HMSO
[12] Gunn J. 1982, 'Defining the Terms', in Dangerousness: Psychiatric Assessment and Management, eds. Hamilton, J.R. and Freeman, H., Gaskell.
[13] Reed D.J. 1992, Review of Health and Social Services for Mentally Disordered Offenders and Others Requiring similar Services, Final Summary Report, Home Office.
[14] Muijen M. 1995, 'Scare in the community: Britain in moral panic', in Mental Health Matters 1996 ed. Heller et al, MacMillan Press Ltd.
[15] Boyd W, D., 1994, 'A Preliminary report on Homicide', Steering Committee of the Confidential Inquiry into Homicides and Suicides by Mentally Ill People.
[16] Reed J. 1997, 'Risk Assessment and Clinical Risk Management: The Lessons from Recent Enquiries', British Journal of Psychiatry, 170, supplement 32, pages 4-7. 
[17] Taylor P. J. and Gunn J., 1999, 'Homicides by people with mental illness: myth and reality'. British Journal of Psychiatry, 174, 9-14
[18] Taylor, P. J. and Gunn, J., 1999, 'Homicides by people with mental illness: myth and reality'. British Journal of Psychiatry, 174, 9-14
[19] MacArthur Violence Risk Assessment Study, Executive Summary, April 1999
[20] Appleby L., 1999, Safer Services: National inquiry into Suicide and Homicide by People with mental Illness, Department of Health.
[21] International Association for Suicide Prevention (IASP) 2006, Co-sponsored by the World Health Organisation (WHO), World Suicide Prevention Day, available: http://www.med.uio.no/iasp/english/wspd/2006/WSPD2006English.pdf#search=%22mental%20illness%20reducing%20suicide%20risk%22
[22] International Association for Suicide Prevention (IASP) 2006. 
[23] Meltzer H., et al 2002, Non-fatal suicidal behaviour among adults aged 16 to 74 in Great Britain, ONS, available: http://www.dh.gov.uk/assetRoot/04/06/07/66/04060766.pdf
[24] Brock A. and Griffiths 2003, 'Trends in suicide by method in England and Wales, 1979 to 2001', Health Statistics Quarterly 20, 2003, ONS, available: http://www.statistics.gov.uk/downloads/theme_health/HSQ20.pdf#search=%22Brock%20A%20and%20Griffiths%20C.%20Trends%20in%20suicide%20by%20method%22
[25] Moran P. 1999, Antisocial Personality Disorder: an epidemiological perspective, Gaskell.
[26] Hagell A. et al. 1999, Scoping Review of Literature on the Health and Care of Mentally Disordered Offenders, NHS Centre for Reviews and Dissemination Policy Research Bureau.
[27] Smith M. 1997, 'Role of the Popular Media in Mental Illness', The Lancet Vol 349
[28] Smith M., 'Role of the Popular Media in Mental Illness', The Lancet (1997) Vol 349
[29] Wilson M. and Francis J. 1997, Raised Voices, Mind.
[30] Jones D. 1995, 'Prediction of Dangerousness', in Management of Violence and Aggression in Health Care, ed. Kidd, B. and Stark, C., Gaskell.
[31] Gunn J. 1996, 'Let's Get serious About Dangerousness', Criminal Behaviour and Mental Health, 51-64 , Whurr Publishers Ltd.
[32] National Confidential Inquiry into Suicide and Homicide 1999, Dept. of Health.
[33] Willmot J. 1999, Risk assessment and Risk Management in Mental Health, Mind.
[34] Browne D. 1997, Black people and Sectioning, Little Rock Publishing.
[35] Crawford M. 2000, 'Homicide is impossible to predict', Psychiatric Bulletin 24:152
[36] National Confidential Inquiry into Suicide and Homicide 1999, Department of Health.
[37] Monahan J., et al 1993, 'Mental Disorder and Violence: Another Look', in Mental Disorder and Crime, ed. Hodgins, Sage.
[38] Read J. and Baker S. 1996, Not just sticks and stones, Mind.
[39] Darton K., Gorman J., Sayce L., Eve 1994, 'Eve fights back - the successes of Mind's stress on women campaign', Mind.
[40] Mind, 2004, 'Ward watch', Mind's campaign to improve hospital conditions for mental health patients, Mind.