Accident or suicide? Predictors of Coroners' decisions in suicide and accident verdicts.

OBJECTIVE: To examine the factors used by Coroners to distinguish between suicide and accidental death among young men in Merseyside and Cheshire.
DESIGN: Retrospective epidemiological survey of deaths due to external causes. Data sources included Coroner's Inquest, GP and hospital data. Logistic regression was carried out to determine the multiple effect of individual factors on defining Coroner's verdict.
SETTING: Merseyside and Cheshire, United Kingdom.
SUBJECTS: Males aged 15-39 years who died from unnatural causes during 1995 in Merseyside and Cheshire. MAIN OUTCOME MEASURE: Coroner's verdict.
RESULTS: An active mode of death was by far the strongest predictor of a suicide as opposed to an accident verdict. Other significant differentiating factors included expressed intent, behavioural change, deliberate self-harm and psychiatric contact.
CONCLUSION: The validity of using method of death as a predictor of intent is questionable. Evidence left by drug users who kill themselves may differ from that left by non-drug users and may need to be sought in less conventional ways. There may be a discrepancy between those factors deemed important by health professionals as indicators of suicide, such as deliberate self-harm, and those given most weight by the Coroner. It may be more pragmatic, in terms of public health policy development, to challenge the concept that self-destructive behaviour can be categorized as being either intentional or unintentional. There is some evidence suggesting that deaths due to suicide and accidents both result from elements of self-destructive behaviour and therefore, the practice of categorizing deaths as either suicides or accidents could be misleading.