The reduction of data collection of homicides committed by mental health patients is short-sighted and detrimental to the safety of patients and wider society, says Sarah Markham

Between 1993-2019 there were 1540 cases of people being killed by people with mental health conditions in the UK. 11% of people convicted of homicide in the UK between 2005 and 2015 were mental health patients, most diagnosed with schizophrenia (compared to a population rate of schizophrenia of around 1%).

The National Confidential Inquiry into Suicide and Safety in Mental Health (NCISH) is a unique project which collects data on all suicides in the UK. One part of the data which are collected is on suicides, homicides and sudden unexplained death of people who have been in contact with mental health services in the 12 months prior to the incident. The current database stands at almost 120,000 suicides in the general population including over 30,000 mental health patients. This dataset has allowed the NCISH to examine the antecedents of suicide and homicide by people under the recent care of specialist psychiatric services, identify factors in patient care which may have contributed to suicide or homicide, and recommend measures for clinical practice and policy designed to reduce the number of patient suicides and homicides. 

Sadly NCISH will no longer continue to receive funding from NHS England to carry out data collection, analysis, and research on patient homicide, although the funding for collecting data on suicide does remain. NCISH are now only commissioned to collect and report on the numbers of people who committed homicide within 12 months of contact with mental health services, and to present aggregate figures in their annual reports. Due to the recent contractual changes, the NCISH will no longer collect detailed information on the antecedents of homicide, the social characteristics of perpetrators, nor the clinical care they received. The programme will continue to report numbers and to maintain the database of cases as a resource that can be explored in future projects if required.  

In my view, it makes no sense to withdraw funds from the one body who is carefully analysing the data to shed light on the antecedent factors and causes behind these individual cases, and interpreting the data to see what can be done to reduce them. It seems especially sad when the body charged with doing this has had the value of its work recognised with recent awards. The data has also been cited in national policies and clinical guidance and regulation in the UK. 

My understanding is that the decision to cut the funding is due to wider budgetary pressures. However given the extremely high costs incurred by treating a patient in a high secure psychiatric hospital (around £250,000 – £300,000 p.a.), it seems on economic grounds alone, rather short-sighted not to fund research which can inform the prevention of homicide, especially as the length of stay for many patients in high secure settings is over five years. 

Mental health trusts are still failing to put the lessons learned from patient homicides into practice. My own local mental health trust Sussex Partnership NHS Foundation Trust has been recently criticised over a review of nine killings carried out by patients as well as another who died in its care. The review found the trust did not always learn from its mistakes and sometimes “severely underestimated” risks. Yet homicide by a mental health service patient always results in a Serious Untoward Incident Inquiry (SUII). The purpose of the SUII is for lessons to learned across the health service, especially the trust concerned to reduce the likelihood of similar tragic events.   

Julian Hendy runs the Hundred families charity, which supports families whose loved ones have been killed by someone with a serious mental illness. The charity works with the NHS and others to prevent similar avoidable tragedies in future. Late last year, Julian contacted me as I am a member of the independent advisory group for the mental health clinical outcome review programme. He asked me how and why the decision was taken for NCISH to cease all homicide prevention work?

Julian lost his own father to a patient homicide and both he and the members of the Hundred families charity are understandably deeply concerned about the recent discontinuation of funding for the NCISH work on patient homicide. There are on average around 120 victims of homicides by patients or people who were mentally ill at the time of the offence in the United Kingdom every year. Many of the victims are children. Many are elderly and vulnerable, and many are patients themselves. Although admittedly small numbers, homicides are catastrophic incidents which cause huge distress to families, neighbourhoods, perpetrators, staff and the general public. The ripples created by such events are profound and extensive across whole communities and can endure for years. These figures underestimate the true scale of the problem as they do not include those who kill themselves after killing others while mentally unwell.  

The cessation of all homicide research regardless of the extent to which this has been dictated by funding cuts, is a mistake and seems terribly short sighted. We need to understand better what causes them and how effective care and treatment can help avoid them. The failure to acknowledge and address perceived problems in the delivery of safe and effective care, undermines public confidence in the good work that is being done and the political will behind it. 

Julian is also concerned that the commissioning decision appears to have been taken without any notice, consultation, or discussion with stakeholders or families affected by patient homicide. In his view there is clearly much work still to be done in homicide prevention, and many areas would benefit from further research. For instance it appears that a substantial number of perpetrators involved in fatal knife crime in London have serious mental health problems and the reasons for their specific vulnerabilities, actions and involvement in gang culture are poorly understood. There has been an apparent increase in the numbers of elderly people with dementia or similar illnesses who have killed others, and the reasons behind this aren’t fully understood. “Hoodwinking” the failure to accurately self-report—has been highlighted in recent Welsh serious incident investigation research, is certainly a feature of some patient homicides, but has to date not appeared to have received academic attention.

More poignantly the needs of families bereaved by patient and other mental health homicides is poorly understood and researched, and may often be completely ignored by mental health services despite the extensive trauma they have endured. As Julian highlights “When I talk to bereaved families, and especially those who have lost children, it is extremely difficult to understand why state agencies would not want to make sure that services are effective and safe and that everything possible is being done to prevent such awful tragedies.”

As a mental health patient I know I would never want to harm anyone, least of all in the context of being ill. It strikes me as very short-sighted, not to say negligent for there not to be funding for research into patient homicide at a national level.

Sarah Markham is an academic mathematician and patient representative currently pursuing a second PhD in theoretical computer science. She is a member of the BMJ Patient Advisory Panel. Twitter: @DrSMarkham

Competing interests: None declared.