Outcome of deliberate self-poisoning. An examination of risk factors for repetition.

Abstract

One of the most important outcomes following an episode of non-fatal deliberate self-poisoning is its repetition. In a prospective follow-up study the subjects were 992 people responsible for 1096 consecutive episodes of deliberate self-poisoning recorded at a teaching hospital accident and emergency department. Risk factors examined were socio-demographic variables, psychiatric and self-harm history, aspects of the self-poisoning episode, and appearance and behaviour at accident and emergency; the frequency of each was compared between those patients who repeated within one year (n = 116) and those who did not (n = 876). Those who repeated were more likely to have ingested more than one drug, to report a previous episode of self-poisoning, to be aged 25-54, and to have experienced previous psychiatric care or psychiatric admission. They were less likely to be in paid employment, or to have expressed a threat to another person or written a note. The best predictor--previous psychiatric contact--only had a positive predictive value of 21% (95% confidence interval 16-25%). Risk factors for repetition of self-poisoning should be kept up-to-date despite modest predictive power. More attention might be paid to clinical rather than socio-demographic aspects of self-harm.