Study: Assessment of risk in patients who commit suicide and homicide is often poor

The assessment of risk in patients who go on to die by suicide or commit homicide is often poor, a new study has found.

A report by The University of Manchester's National Confidential Inquiry (NCI) into Suicide and Homicide by People with Mental Illness out today has raised concerns about the way that risk is assessed and led to criticism of the tick-box approach to clinical practice.

Researchers looked at a sample of 81 cases where risk had been judged to be low, but seven days or less after the assessment the patient died by suicide or committed homicide, to retrospectively address the quality of the assessments. They found that in about a third of the cases (36% of the patient suicides and 41% or the patient homicides) the overall quality of risk assessment was unsatisfactory. In the majority of cases, however, risk assessment and management were satisfactory, even though they were followed within seven days by a fatal incident.

Professor Louis Appleby, commissioned by the Healthcare Quality Improvement Partnership (HQIP) on behalf of the NHS England, NHSSPS Northern Ireland, the Scottish Government, the Welsh Government and the Channel Islands, said: "The results suggest there is a need for risk management to be individually tailored, or personalised, to each patient rather than following a "tick-box" approach.

"In the majority of the 81 sample cases studied risk assessment and management were satisfactory, but in a significant minority of cases we studied we found clinical risk assessment and management may not have been adequate. This is an uncomfortable conclusion for clinicians but one that should lead to improvements.

"In a small but significant number of cases, even when risk was recognised, appropriate management did not follow, for example patients were granted unescorted leave within a few hours of detention for acute psychosis. This could be the consequence of a "tick-box" approach to risk assessment, something that has been widely criticised by clinicians." 

Professor Appleby, from the University's Centre for Behaviour and Mental Health who is also part of Manchester Mental Health and Social Care Trust, said the risk factors associated with suicide and serious violence in mental illness were well documented. Previous studies show one in four people who die by suicide have a history of recent contact with mental health services and one in 10 who commit homicide.

The researchers developed a framework for evaluating the quality of risk assessment and management in mental health patients based on existing best practice guidelines which included taking into account the patient's history, mental state and current circumstances, an overall judgment of the risk factors, a management plan, communication of the management plan and overall quality of assessment.

Unsatisfactory assessments before a homicide were often associated with a diagnosis of personality disorder or alcohol misuse.

The report authors hope that services will now use the framework and key principles set out in the report to examine their own processes, in particular that of individual risk assessment formulation followed by personalised risk management.  Further studies should focus on assessment of patients with personality disorder, look at more cases using the framework, Professor Appleby added.