Changes in coroners' practice may be compromising quality of suicide statistics

Assessment of official suicide statistics found that between 1990 and 2005, the proportion of researcher-defined suicides given a verdict of suicide by the 12 coroners studied decreased by almost seven per cent, largely because of the increased use of misadventure/accident verdicts for deaths thought, on clinical review, to be suicides. Growth in the use of narrative verdicts by coroners may also have compromised assessment of small area differences in suicide rates. This is just one of the key findings from a report, published in the first issue of the new National Institute for Health Research (NIHR) journal Programme Grants for Applied Research, aimed at reducing premature mortality from suicide and non-fatal self-harm.

Suicide and self-harm are the most serious and devastating consequences of mental illness. More than 4,000 people take their lives every year in England and there are over 200,000 hospital presentations for self-harm annually.

Comprising findings from four linked studies led by the UK's leading experts in the field including University of Bristol academics, the report provides new evidence aimed at improving management of self-harm, reducing the incidence of suicide and providing reliable data to evaluate the impact of the National Suicide Prevention Strategy for England (2002).

Key findings from the studies include:

• The impact of UK legislation to reduce pack sizes of paracetamol was followed by a 43 per cent reduction in number of deaths and a 61 per cent reduction in registrations for liver transplantation over the following 11 years with the majority of sales outlets appearing mostly to be adhering to sales guidance.
• An initiative to withdraw the painkiller co-proxamol from use in the UK following concerns over its widespread use for suicidal poisoning was found to have resulted in approximately 600 fewer deaths than predicted between 2005 and 2010 based on previous trends, with no evidence of substitution by poisoning with other analgesics.
• An assessment of the relative toxicity in overdose of commonly used tricyclic antidepressants, found dosulepin and doxepin had the greatest toxicity. Citalopram was more toxic than other selective serotonin reuptake inhibitors.
• An audit of self-harm management found there was marked variation between hospitals in the management of self-harm; effects of this variation on patient outcomes were unclear, although psychosocial assessment may have been associated with reduced repetition. Levels of specialist assessment remained static between 2001–2 and 2010–11, but service quality appeared to improve.
• Assessment of official suicide statistics found that between 1990 and 2005, the proportion of researcher-defined suicides given a verdict of suicide by the 12 coroners studied decreased by almost seven per cent, largely because of the increased use of misadventure/accident verdicts for deaths thought, on clinical review, to be suicides. Use of narrative verdicts increased markedly. Coroners who gave more narrative verdicts also gave fewer suicide verdicts, and geographical variations in the use of narrative verdicts appeared to distort reliable assessment of small-area differences in suicide rates.