Mental illness in people who kill strangers: longitudinal study and national clinical survey

This careful study has important implications for reducing both
homicides of strangers and homicides of known victims too. The system of
public enquiries set up after the report on the care of Christopher Clunis
for all homicides by people with mental illness, may inappropriately
increase public alarm and stigma about mental illness.

To reduce stigma it is not enough correctly to argue that suicide is
a far greater problem than homicide in mental illness. What perhaps
captures the public imagination with headlines after each public homicide
enquiry is the mysterious and occasionally unpredictable nature of
psychosis. Much remains mysterious but progress is being made and the
public should be reassured that serious efforts are under way.

Clearly a serious untoward investigation needs to take place in every
case of homicide, but if I read this study correctly the implications are
that it will be far more productive nationally to target the individual
psychological and social needs of the patient and their family than to
implement systems of greater public control which would appear to treat
patients as a public health hazard lacking minds of their own.

In the population studied over 3 years there would appear to have
been only 8 people with a diagnosis of schizophrenia who killed a stranger
who had not attended mental health services in the previous week. By
comparison there were 64 people with a diagnosis of schizophrenia who
killed someone known to them.

In terms of implementing public health measures to reduce the overall
figure it would be more productive to ensure detailed audit to check that
the individual psychological needs of all these people with a diagnosis of
schizophrenia, are met routinely. Especially compelling is the argument
that screening for family interventions, long known to be important in the
recovery from psychosis, must be routine, since this study shows victims
known to the perpetrator remain almost 80% of all victims.

If the figures for affective disorder include affective psychoses,
this gives even more weight to apply such good policies in health care to
the wider set of people with psychosis overall, since in any case
schizophrenia is recognised to be a heterogeneous diagnosis.

Concerning the other diagnostic categories, again one of the clinical
implications would appear to be to offer more individual psychological
treatment. Irritability and anger are symptoms that stretch across
diagnostic categories as they are non-specific.

They may therefore be overlooked by traditional mental health
services based on diagnosis. There would appear to be good arguments for
screening all patients with mental illness for significant unresolved
problems of this nature and offering them a course in anger management,
which in my clinical experience is often heard as a dignified and
constructive proposals even by patients who do not have insight into other
aspects of their problems.

It is welcome the study is funded by the National Institute for
Clinical Evidence. It provides compelling evidence that its guidelines on
good practice in the individual treatment of schizophrenia, probably
extended to all psychotic syndromes, should now be backed with a serious
national survey to identify how well they are being implemented, and what
are the obstacles.

Dr Chris Burford
Consultant Psychiatrist

Competing interests:
Committee Member of UK chapter of ISPS, the International Society for the Psychological Treatments of the Schizophrenias and other psychoses. www.isps.org