• David Gunnell, Olive Bennewith, Tim J Peters, Allan House, and Keith Hawton.
• Department of Social Medicine, University of Bristol, Canynge Hall, Whiteladies Road, Bristol BS8 2PR, UK. d.j.gunnell@bristol.ac.uk
• J Public Health (Oxf). 2005 Mar 1;27(1):67-73.

BackgroundPrevious research into the epidemiology and management of self-harm has been largely based in centres with a special interest in this behaviour or focused on hospital admissions only. There are no national data on the characteristics and management of people presenting to hospital following self-harm.MethodsData were collected from 8-week service audits carried out in a stratified random sample of 31 general hospitals in England.Results4033 episodes of self-harm resulting in presentation to Accident and Emergency Departments were identified. Overdose alone accounted for 79 per cent of episodes, 80 per cent of presentations were outside normal office hours (9 am - 5 pm, Monday to Friday) and the peak period of attendance was from 8 pm to 2 am. In only 56 per cent of episodes was a specialist psychosocial assessment conducted prior to discharge and less than half (46 per cent) led to admission to a hospital bed. Psychiatric admission occurred in 10 per cent. Episodes involving older subjects (>45 years) and those using methods other than laceration or overdose were the most likely to lead to assessment and admission.ConclusionsNon-fatal self-harm is one of the strongest predictors of suicide, yet nearly half of all hospital attendances in England following self-harm do not lead to a specialist assessment. Patterns of service provision should take account of the observation that most self-harm attendances occur outside normal working hours and those at greatest risk of repetition are the least likely to receive assessments.

33 keywords...

hide…