You don't need to be signed in to read BMJ Group Blogs, but you can register here to receive updates about other BMJ Group products and services via our Group site.

risk

Do we need yet another homicide inquiry?

17 May, 09 | by Steven Reid, Evidence-Based Mental Health

Last week NHS Yorkshire and Humber released the findings of an external investigation into the care and treatment of Benjamin Holiday. You can find the report here. In January 2005 Holiday, a young man with schizophrenia, killed a pregnant woman, Tina Stevenson, stabbing her in the street close to her home. He was under the care of a community mental health team and living at home with his mother. Having no recollection of the attack he pleaded guilty to manslaughter with diminished responsibility and has been detained indefinitely in a special hospital.

Department of Health guidelines (shouldn’t that be directives?) call for a series of reviews when a homicide has been committed by a person who has been in contact with mental health services:
1) A fast-track (72 hours) investigation to identify any necessary immediate action
2) An internal mental health trust investigation (within 3 months)
3) And finally, an independent investigation commissioned by the Strategic Health Authority.

The reasons for an external investigation are obvious: lessons must be learned to prevent such events happening again. But do they work? They certainly provide a revealing insight into the day to day care of someone living with a severe mental illness, affording a degree of transparency for relatives and carers that is rarely available elsewhere. And outcomes? There have been at least 400 homicide inquiries in the last 15 years yet the rates of homicide by people with mental disorder remains unchanged: 30-50 from 800 homicides annually, with fewer than 10 committed by people with a psychotic illness (the remainder largely related to personality disorder and drug misuse). The recommendations have become familiar too: improved risk assessment, better communication and better record-keeping. Another report isn’t needed to tell us that. The Holiday Inquiry illustrates their inefficiency. Tina Stevenson was killed in 2005; the inquiry reports four years later in which time things have changed. We have a new Mental Health Act, community treatment orders, crisis resolution teams and I’m sure that like elsewhere, services in Hull have been completely reorganised. The authors acknowledge that much of what they say was reported in 2006 by the mental health trust’s internal review.

Much has been made of bringing an end to the ‘blame culture’ in the NHS which encourages staff to cover up errors for fear of retribution. The new focus is on systematic factors that allow such errors to be made. It is human nature, however, to search for a scapegoat and in this inquiry: “…if we are allowed to indulge in one piece of viewing this investigation through a ‘retrospective-scope’ we would conclude this was a missed opportunity to possibly break into the cycle”. What they are referring to here is the decision by a social worker not to complete an application to detain Holiday under Section 3 of the Mental Health Act two months before the homicide. At the time however, he was agreeing to an informal admission, and following his discharge one month later he was seen by a nurse on two occasions and given his depot antipsychotic injection. It is just as plausible that by detaining him at that time he may have become less cooperative with treatment and disengaged completely – a classic example of hindsight bias.

Of course these points have been made before. In 1999 the DoH’s Safer Services report recommended that homicide inquiries be scrapped but they keep on coming. Clearly they fulfil a social need. The Department of Health guidelines state that mental health services must ‘be seen to investigate’ any serious incident and though not explicit, the allaying of the publics’ anxiety about the mentally ill is key. We used to have asylums to contain the mad, but now we have community care – an idea that now attracts as much vilification as political correctness. The fear of mental illness is as potent as ever and homicide inquiries serve to remind us that whilst they may live among us, they remain alien and dangerous – as illustrated by headlines such as this: Crazed Killers – trust’s failings exposed.

Antiepileptic Alert: scaremongering or suicide prevention?

8 Jan, 09 | by Steven Reid, Evidence-Based Mental Health

The debate has been rumbling on for a year, but last month the US Food and Drug Administration (FDA) announced that manufacturers of antiepileptic drugs will have to add warnings to package inserts stating that these drugs increase the risk of suicide. They have also issued a public health advisory requiring that “health care professionals should notify patients, their families and caregivers of the potential for an increase in the risk of suicidal thoughts or behaviors so that patients may be closely observed”. These alerts arrive following a review of 199 clinical trials that showed a doubling of risk of suicidality (not completed suicide) in people using antiepileptics. So warning patients makes sense doesn’t it? But then why are so many neurologists, psychiatrists and notably the American Epilepsy Society unhappy with the decision?

It boils down to the perception of risk and the old relative vs. absolute question. The trials included over 40000 patients and found that the absolute risk of suicidal thinking or behavior was 0.43% in the group taking antiepileptics versus 0.24% in the group on placebo. That equates to one additional case of suicidality (again, not completed suicide) for every 500 patients taking the active drug. Do antiepileptics seem quite so scary now?

The FDA says that the warnings are needed to promote monitoring for a significant, even if small, risk. Given that epilepsy itself is associated with a threefold increased risk of completed suicide watching for signs of depression should be routine. As well as epilepsy antiepileptics are also used in bipolar disorder and with chronic pain – conditions with their own significant risk of suicide. But a regulator urging increased monitoring doesn’t seem to affect clinical practice. Take the case of antidepressant prescribing in adolescents (see an earlier post). When trials indicated an increased risk of suicidality with SSRIs the FDA slapped on a black box warning. The adolescent suicide rate went up (possibly coincidental) and prescribing rates plummeted (probably not coincidental). That’s the concern with this latest alert: people will be put off taking antiepileptic drugs, and unlike antidepressants there is no alternative treatment. Providing information is important but so is context and I think this is a decidedly dubious decision by the FDA.

EBMH blog homepage

HeadtoHead

A quarterly digest of the most important clinical research of relevance to clinicians in mental health.Visit site

Latest from Evidence-Based Mental Health

Latest from Evidence Based Mental Health