Injury prevention/original research
A Clinical Tool for Assessing Risk After Self-Harm

https://doi.org/10.1016/j.annemergmed.2006.07.944Get rights and content

Study objective

Our aim is to develop a risk-stratification model for use by emergency department (ED) clinical staff in the assessment of patients attending with self-harm.

Methods

Participants were patients who attended 5 EDs in Manchester and Salford, England, after self-harm between September 1, 1997, and February 28, 2001. Social, demographic, and clinical information was collected for each patient at each attendance. With data from the Manchester and Salford Self-Harm Project, a clinical decision rule was derived by using recursive partitioning to discriminate between patients at higher and lower risk of repetition or subsequent suicide occurring within 6 months. Data from 3 EDs were used for the derivation set. The model was validated with data from the remaining 2 EDs.

Results

Data for 9,086 patients who presented with self-harm were collected during this study period, including 17% that reattended within 6 months and 22 patients who died by suicide within 6 months. A 4-question rule, with a sensitivity of 94% (92.1-95.0% [95% confidence interval]) and specificity of 25% (24.2-26.5% [95% confidence interval]), was derived to identify patients at higher risk of repetition or suicide.

Conclusion

Application of this simple, highly sensitive rule may facilitate assessment in the ED and help to focus psychiatric resources on patients at higher risk.

Introduction

The mean annual rate of people presenting to general hospitals in the United Kingdom after self-harm has increased in recent years, and the latest estimate is 170,000 per annum.1 In the United States, the number of presentations to emergency departments (EDs) after self-harm has been reported as more than 400,000 annually, with one third resulting in hospital admission.2 This number puts considerable pressure on medical and psychiatric services in hospitals.3 A number of guidelines have been published in the United Kingdom4, 5 on the management of self-harm. These have recommended that all patients treated in the ED after an episode of self-harm be given a psychosocial assessment before discharge by a member of staff trained for the task. There is no requirement for the assessment to be carried out by a mental health specialist. Even so, most hospitals fell short of these standards when surveyed.6 Many EDs lack sufficient psychiatric support to provide specialist psychosocial assessments on all patients.

Patients who present with self-harm are at high risk of further self-harm and suicide.7, 8 A quarter of patients who commit suicide have attended a general hospital because of self-harm in the preceding 12 months.3 The majority (up to 75%) of repeated episodes occur within 6 months of the index episode,9 and suicide risk is also highest during this period.10 A reduction in suicide is a key goal of mental health policy in England11 and was highlighted in the United States by the Surgeon General’s Call to Action to Prevent Suicide.12

The prediction of suicidal behavior is difficult because of the low specificity of identified risk factors.13 The initial assessment of self-harm patients is the responsibility of medical staff in the ED in most hospitals. Risk-stratification models have been used in ED settings in recent years to improve clinical care.14 These models prioritize sensitivity over specificity (ie, they identify most of the index cases but also pick up many nonindex cases). They enable clinicians to make informed choices about the appropriate treatment of patients. Methodologic standards for the development of clinical decision rules in emergency medicine have now been established to optimize the quality of novel clinical decision rules and to assist the emergency clinician in assessment(s).15 However, a recent US review concluded that there was limited evidence for the accuracy of screening tools in identifying suicide risk.16 A previous report by our group showed that the predictive value of global clinical assessments for self-harm was poor and could not be used as a basis for providing intervention.17

The aim was to develop a risk-stratification model for use by ED clinical staff in the assessment of patients attending with self-harm. Our goal was to derive a short (3- to 5-element) decision rule, first to correctly identify 95% or more of repetitions of self-harm and all suicides within 6 months and second to provide the best possible specificity. This rule was required to be simple and reliable enough to be used by ED clinical staff to help them discriminate between patients at higher and lower risk of repetition or subsequent suicide.

Section snippets

Study Design and Setting

The data were collected prospectively by using information from a citywide monitoring of self-harm known as the Manchester and Salford Self-Harm project.18 This project included 5 EDs in the neighboring cities of Manchester and Salford, which together serve a catchment population of 608,922 individuals. As part of routine clinical practice in all of the participating hospitals, after a presentation with self-harm, a standard psychosocial assessment form was completed by the emergency clinician;

Results

There were 11,819 episodes of self-harm treated in the study departments during the 3 ½-year study period (September 1, 1997, to February 28, 2001); 2,743 (22%) episodes were not included in the analysis because the patients had either refused treatment or left before assessment. Thus, 9,086 patient episodes were available for analysis. Of these, 6,933 episodes were in the derivation set and 2,153 episodes in the validation set, which is summarized in the recruitment diagram shown in Figure 1.

Limitations

This study has limitations. The accuracy of the model depends on the completeness of the database from which it was derived. Although we collected detailed psychosocial data on about 80% of possible self-harm episodes, patients using cutting as a method of harm may have been underrepresented in our study. The Manchester and Salford Self-Harm data set is based on self-harm episodes that are treated in EDs. Our findings may not be applicable to patients who do not wait for treatment or community

Discussion

We have derived the Manchester Self-Harm Rule that correctly identifies some 94% of patients who repeated self-harm in the next 6 months. Furthermore, there were no suicides missed by our rule within the 6-month follow-up period. The validation of the rule, in a set of patients treated in 2 EDs to those patients included in the derivation data set, produced similar results, suggesting that our rule is robust. The predictive value of the rule is superior to the predictive value of global

References (31)

  • K. Hawton et al.

    Suicide and other causes of death, following attempted suicide

    Br J Psychiatry

    (1988)
  • S. Gilbody et al.

    The early repetition of deliberate self-harm

    J R Coll Physicians Lond

    (1997)
  • J. Cooper et al.

    Suicide following deliberate self-harm: a 4 year cohort study

    Am J Psychiatry

    (2005)
  • National Suicide Prevention Strategy for England

    (2002)
  • The Surgeon General’s Call to Action to Prevent Suicide

    (1999)
  • Cited by (0)

    Supervising editor: Debra E. Houry, MD, MPH

    Author contributions: JC and KM-J designed the original study, with help from EG and LA. JC collected the data, and JC, NK, and JD analyzed the data. JC, NK, JD, and KM-J contributed to writing the article, with help from EG and LA. All authors have seen and approved the final version. JC and LA secured funding for the project. JC takes responsibility for the paper as a whole.

    Funding and support: The project was funded by Manchester Health Authority, South Manchester University Hospitals NHS Trust, Central Manchester and Manchester Childrens’ University Hospitals NHS Trust, Pennine Acute NHS Trust and the Mental Health Services of Salford NHS Trust.

    View full text