Are homicides by people diagnosed with mental illness preventable?

Around 5-10 per cent of homicides per year are committed by psychiatric patients. And forensic psychiatrist Tony Maden says should his recommendations – including community treatment orders – in a report he submitted to the government be implemented, the homicide rate by psychiatric patients could be cut by 10 per cent in five years. Is Maden being realistic? Adam James investigates


It’s now 14 years ago that Christopher Clunis stabbed to death a a stranger, Jonathan Zito, at a London tube station.
A report into the attack revealed a catalogue of errors in Clunis’s care.

The 23 year old diagnosed with schizophrenia had a history of violence and medication non-compliance. He had been seen by 43 different psychiatrists in five years and had crossed from one side of the London to the other on four occasions, passing through three out of the four former regional health authorities.

The killing by Clunis – and the subsequent public outcry – saw successive governments subsequently invest years of time and money into both improving the care for such dangerous patients, and finding means to provide professionals with more powers to compulsory detain and treat them. Other homicides – particularly the murder by Michael Stone of Lin Russell and her daughter Megan 10 years ago – intensified the Labour government’s intention.

A new mental health bill, first brought before parliament in 2002, outlined government plans. But a united front of mental health and law groups criticised the bill for being draconian and a threat to civil liberties. It was abandoned in March this year in favour of modifications to existing mental health law.

This new “streamlined” bill has yet to be published. But Mental Health Minister Rosie Winterton has announced new initiatives to shake-up how mental health professionals assess the violence risk of patients. These initiatives are based on the publication of a Department of Health-commissioned review of 25 homicides by people diagnosed with a severe mental illness, carried out by Tony Maden, professor of forensic psychiatry at Imperial College London.

The new measures include developing a “national evidence framework” for assessing and managing risk and to develop guidance on information sharing between health, social care and criminal justice agencies about high-risk patients. The government also wants to review the “cumbersome” care programme approach (CPA) used by professionals to coordinate and review care of patients. Strengthening the role of the patient’s care coordinator (usually a nurse or social worker), is something the government also has in its sights. But, perhaps most significantly, Winterton confirmed plans to press ahead with community treatment orders (CTOs) whereby patients previously detained in hospital will, if necessary, be required to be treated (i.e. take their medication) when living in the community.

Around 5-10 per cent (equivalent to 55-60 out of a 600-700 total) of homicides are committed per year by psychiatric patients. And Maden said that should all his recommendations – including CTOs – from his report be implemented, it would be “reasonable” target for the homicide rate to be cut by 10 per cent in five years. This equates to 5-6 fewer homicides by psychiatric patients per year, and an overall reduction in homicides of just 0.7-1.2 per cent.

Critics say, however, that Maden is being way over optimistic. Tom Fahy, professor of forensic mental health at the Institute of Psychiatry, accepts Maden’s recommendations could reduce violence. But adds they are “very unlikely” to reduce homicide rates at all by people diagnosed with a mental illness. “Every homicide is a disaster and tragedy,” says Fahy. “I support the notion of improving risk assessment. But at the same time one needs to be realistic about what that might deliver….Focussing on homicide is a mistake. They are freak events”

Questions have also been raised about the validity of implementing a costly national policy change on the basis of Maden’s analysis of 25 cases over a 10 year period. “It’s going to create a lot of burden on clinicians – and all from an analysis of just 25 homicides,” says consultant psychiatrist Sashi Sashidharan, co-director of the National Centre for Research in Ethnicity and Mental Health at the University of Warwick.

But it is the spectre of CTOs that campaigners feel most uneasy about. Andy Bell, chair of the Mental Health Alliance, a group of 78 professional, human rights and service user organisations, fears CTOs in England and Wales could end up being implemented purely as a form of social control without any health benefits or to simply increase hospital beds in a struggling NHS. Nevertheless, CTOs have the potential to benefit some revolving door patients, he says. “We have to make sure there are safeguards and clear conditions – such as how quickly you get off CTOs,” says Bell. “We are also concerned that they might be used, like an anti social behaviour order, to regulate behaviour, such as not going to a public house.”

Maden is keen to stress that “99.9%” of patients would be unaffected by CTOs. But, such orders, he says, would provide professionals with desperately-needed additional powers to ensure medication compliance on high-risk ’revolving door’ patients with a history of violence and non compliance, without having to detain them in hospital. “The lack of such powers discredits the whole enterprise of risk management,” says Maden.

Meanwhile, attention has turned to Scotland where CTOs [in Scotland they are referred to as Community Compulsory Treatment Orders] have been in place since October last year. A total of 144 CTOs have been issued in Scotland between October 2005 and March 2006, according to initial findings by the King’s Fund. It is too early to say whether CTOs have helped reduced homicides. But, interestingly, Simon Lawton-Smith, a senior fellow with The King’s Fund, reports patients in Scotland themselves consider CTOs to be “fairer”. This is because, says Lawton-Smith, CTOs can lead to patient care being better resourced (e.g. provision of ‘talking therapy’) and a tribunal system allows patients more of a say. “In Scotland CTOs are not being implemented on the basis of kicking down someone’s door and injecting them on the kitchen table,” comments Lawton-Smith. “No-one has reported any horror stories.”

Nevertheless, Lawton-Smith adds that professionals in Scotland are unhappy about the added administrative burden, particularly that associated with the tribunal process. “A lot of people in Scotland feel that [implementing CTOs] is putting more pressure on them,” says Lawton-Smith.

However, the CTO legal framework in Scotland is vastly different to what is planned for England and Wales, points out Dr Tony Zigmond, vice-president of the Royal College of Psychiatrists. In Scotland, a patient must at least be judged to have a treatable mental disorder (a “treatability test”) and lack decision-making capacity before being subjected to either hospital compulsory powers or a CTO. The government, to the fury of most mental health groups, has backed off from inscribing both these conditions into law for England and Wales. “If you are going to give people more powers to deprive someone of their civil liberties you have to be very clear when drawing up any legal framework,” says Zigmond.

Moreover, while Zigmond supports CTOs “in principle”, he says worldwide evidence demonstrates they are ineffective. “The Cochrane Review of CTOs stated that 85 CTOs are needed to prevent one [hospital] admission, and 238 to prevent one arrest,” he says. This leads Zigmond to conclude that CTOs are “probably pointless”. Maden admits to not having read the Cochrane Review, published this year. But he argues CTOs would, in effect, be an extension of restrictions orders which already allow high-risk patients to live in the community under conditions. “These are well-recognised as being effective with high-risk patients” says Maden

Some fear, however, that on the long term CTOs will achieve the exact opposite of what they are intended for – alienating patients further from mental health services. Sashidharan says: “I’ve never seen evidence that CTOs work. We need to look at this issue [of ensuring compliance] from the other side – with an emphasis on engaging patients more, and giving services resources to do this, such as improving the staff/patient ratio and looking at what needs the patient has, other than just medication. Successful psychiatry, wherever you look, is about the level of engagement with patients.”

Maden, nevertheless, is a vigorous defender of his review and clinical intentions. He is particularly scathing against those who claim his recommendations will not make any significant difference to the overall homicide rate, and would be extremely costly.

“I believe homicides by mentally ill people are preventable,” he asserts. “The argument that the overall homicide rate will not be significantly reduced is perhaps valid for those working in public health. But it is a terrible point for psychiatrists to make. Doctors working with cancer do not assume that they can not do anything about cancer. They presume that if they work hard they will get somewhere. I am not concerned about the overall homicide rate I am concerned about the homicide rate by the mentally ill.”


Some Maden recommendations:
* All mental health teams have “access” to structured clinical assessments” of violence risk
* Community treatment orders (CTOs)
* Early intervention when behaviour of patient with history of violence deteriorates. Care Programme Approach meetings should set clear, operational criteria for intervention
* Violence risk assessment should be undertaken early in a patient’s contact with services.
* Concern about violence risk should be shared openly with patients and carers when possible

A Tragic Statistic – one homicide reviewed by Tony Maden

Background: 25 year old brought up partly in UK, partly in Nigeria. Fights at school and expelled. Involved in criminal activity. Stabbed a fellow college student. Four year sentence for wounding.

History: February 1996 given 4.5 year sentence for robbery. Approached a Jaguar car, forced driver out, punched him and stole keys. During sentence suffered psychotic breakdown and transferred to medium secure unit in January 1997. Returned to prison in August 1997 but transferred in April 1999 to another medium secure unit. Committed assault in September 1999, throwing hot water in face of patient who had alleged to staff he was using cannabis. Transferred to locked ward. Fondled a female nurse.

On November 24, 1999, convicted of actual bodily harm for assault and sentenced to be detained under S37. Transferred to open ward in January 2000. Went absent without leave. Discharged to mother’s home in June 2000. Placed on supervision register. Last contact with services on March 14, 2001. Friendly but non-compliant with medication. Arguments with mother. Not seen again until after he killed his mother on May 14, 2001.

Maden’s comment: History as violent criminal before developing mental illness. Severe personality disorder. Never compelled to comply with medication in community. Forensic services involved but never took long-term responsibility and he was returned to the community through general services.

Strong case in favour of forensic services managing his care. Structured assessment would have made risks explicit. Community treatment order would have been useful, but missed opportunity to recommend restriction order following assault in the independent medium secure unit. Case raises many issues, including follow-up of patients transferred back to prison after treatment in a medium secure unit.

* A shortened version of this article appeared in Mental Health Today

Leave a Reply