Self-harm in England: A tale of three cities

Research output: Contribution to journalArticle

  • External authors:
  • Keith Hawton
  • Helen Bergen
  • Deborah Casey
  • Sue Simkin
  • Ben Palmer
  • Jayne Cooper
  • Judith Horrocks
  • Allan House
  • Rachael Lilley
  • Rachael Noble
  • David Owens


Background: Self-harm is a major healthcare problem in the United Kingdom, but monitoring of hospital presentations has largely been done separately in single centres. Multicentre monitoring of self-harm has been established as a result of the National Suicide Prevention Strategy for England. Method: Data on self-harm presentations to general hospitals in Oxford (one hospital), Manchester (three hospitals) and Leeds (two hospitals), collected through monitoring systems in each centre, were analysed for the 18-month period March 2000 to August 2001. Results: The findings were based on 7344 persons who presented following 10,498 episodes of self-harm. Gender and age patterns were similar in the three centres, 57.0% of patients being female and two-thirds (62.9%) under 35 years of age. The largest numbers by age groups were 15-19 year-old females and 20-24 year-old males. The female to male ratio decreased with age. Rates of self-harm were higher in Manchester than Oxford or Leeds, in keeping with local suicide rates. The proportion of patients receiving a specialist psychosocial assessment varied between centres and was strongly associated with admission to the general hospital. Approximately 80% of self-harm involved self-poisoning. Overdoses of paracetamol, the most frequent method, were more common in younger age groups, antidepressants in middle age groups, and benzodiazepines and sedatives in older age groups. Alcohol was involved in more than half (54.9%) of assessed episodes. The most common time of presentation to hospital was between 10 pm and 2 am. Conclusions: Multicentre monitoring of self-harm in England has demonstrated similar overall patterns of self-harm in Oxford, Manchester and Leeds, with some differences reflecting local suicide rates. Diurnal variation in time of presentation to hospital and the need for assessment of non-admitted patients have implications for service provision. © Springer-Verlag 2007.

Bibliographical metadata

Original languageEnglish
Pages (from-to)513-521
Number of pages8
JournalSocial psychiatry and psychiatric epidemiology
Issue number7
Publication statusPublished - Jul 2007