Combination of self-harm methods and fatal and non-fatal repetition: a cohort study

Background: Assessment and aftercare for people who self-harm needs to be related to an understanding of risks of adverse outcomes. We aimed to determine whether self-harm by a combination of methods and its early repetition are associated with adverse outcomes – especially non-fatal repetition and suicide. Method: 10 829 consecutive general hospital attendances due to self-harm in one large English city were monitored, through scrutiny of Emergency Department attendances, over three years and followed up to determine the incidence of non-fatal repetition. Subsequent deaths, by any cause and by suicide, were determined from national statistical records. Results: 6155 patients accounted for the 10 829 episodes: 72% by self-poisoning, 21% self-injury, and 746 episodes (7%) due to a combination of methods. After a combined-methods index episode, non-fatal repetition (P = 0.001) and suicide (P=0.002) occurred sooner and more frequently than it did among those who had self-poisoned. Further hospital attendance due to self-harm within a month was associated with a 3.7-fold (95% CI 2.1–6.4) risk of subsequent suicide. Limitations: The data exclude self-harm episodes that do not result in a hospital attendance. Index episodes in the study are not generally life-time first episodes so follow-up data are based on an arbitrary start-point. Both of these limitations are common to all studies of this kind. Conclusions: At psychosocial assessment and the making of aftercare arrangements, combined methods of self-harm or another recent episode should be considered ‘red-flag’ indicators for attention to care.