Mortality and suicide after non-fatal self-poisoning: 16-year outcome study.

Background Suicide reduction is government strategy in many countries. We need to quantify the connection between non-fatal self-poisoning and eventual suicide. Aims To determine mortality after an episode of self-poisoning and to identify predictors of death by any cause or by suicide. Method A retrospective single-group cohort study was undertaken with 976 consecutive patients attending a large accident and emergency unit in 1985–1986 after non-fatal self-poisoning. Information about deaths was determined from the Office for National Statistics. Results Of the original patients, 94% were traced 16 years later; 17% had died, 3.5% by probable suicide. Subsequent suicide was related to numerous factors evident at the time of the episode of self-poisoning but, when examined for their independent effects, only the severity of the self-poisoning episode and relevant previous history seemed important. Conclusions Patients attending a general hospital after self-poisoning all require good basic assessment and care responsive to their needs. Attempts to reduce the huge excess of suicide subsequentto self-harm are not likely to achieve much if they are based on the identification of subgroups through‘risk assessment’.