"Near-fatal" deliberate self-harm: characteristics, prevention and implications for the prevention of suicide

J Affect Disord. 2004 Apr;79(1-3):263-8. doi: 10.1016/S0165-0327(02)00391-9.


Background: There have been few studies of the most serious cases of deliberate self-harm (DSH). These cases represent an important clinical problem. They may also be suitable as a "proxy" for suicide in research studies.

Methods: We developed a definition of "near-fatal" deliberate self-harm (NFDSH) and identified all cases attending accident and emergency departments in an urban area during an 18-month period. We compared them to less severe DSH on social and clinical characteristics, and examined their initial clinical management. We compared their age and gender profile with that of completed suicides. NFDSH cases who agreed to be interviewed were asked about their experiences of seeking help and their views on prevention of suicide.

Results: 158 cases were identified, 8% of all DSH. The most common method of self-harm was self-poisoning. The most common psychiatric diagnosis was depression. Compared to less severe DSH, near-fatal cases had higher rates of several social and clinical risk factors for suicide and showed evidence of greater suicidal intent. Forty-one percent were allowed home following assessment. Most admissions lasted 2 days or less. The age and gender profile was closer to that of suicide than in DSH generally. Few subjects agreed to be interviewed; those who did described poor previous experiences of services.

Limitations: We did not assess each case of DSH individually but applied a workable definition of NFDSH. Firm conclusions cannot be drawn from the qualitative part of the study because of the low rate of agreement to interview.

Conclusions: NFDSH is an important clinical phenomenon associated with indicators of high suicide risk. This high risk is frequently not reflected in clinical management. Further studies of NFDSH could contribute to what is known about suicide prevention but the reluctance of NFDSH subjects to be interviewed may be a difficulty for future research.

Publication types

  • Multicenter Study

MeSH terms

  • Adult
  • Age Factors
  • Depression
  • Emergency Service, Hospital / statistics & numerical data
  • Female
  • Humans
  • Male
  • Middle Aged
  • Poisoning
  • Prospective Studies
  • Risk Factors
  • Self-Injurious Behavior / prevention & control*
  • Self-Injurious Behavior / psychology*
  • Severity of Illness Index
  • Sex Factors
  • Suicide / psychology*
  • Suicide Prevention*
  • Suicide, Attempted / prevention & control*
  • Suicide, Attempted / psychology*