The sad truth about the SADPERSONS Scale: An evaluation of its clinical utility in self-harm patients

Journal article

Background: The SADPERSONS Scale is commonly used as a screening tool for suicide risk in those who have self-harmed. It is also used to determine psychiatric treatment needs in those presenting to emergency departments. To date, there have been relatively few studies exploring the utility of SADPERSONS in this context. Objectives: To determine whether the SADPERSONS Scale accurately predicts psychiatric hospital admission, psychiatric aftercare and repetition of self-harm at presentation to the emergency department following self-harm. Methods: SADPERSONS scores were recorded for 126 consecutive admissions to a general hospital emergency department. Clinical management outcomes following assessment were recorded, including psychiatric hospital admission, community psychiatric aftercare and repetition of self-harm in the following 6 months. Results: Psychiatric hospital admission was required in five cases (4.0%) and community psychiatric aftercare in 70 (55.5%). 31 patients (24.6%) repeated self-harm. While the specificity of the SADPERSONS scores was greater than 90% for all outcomes, sensitivity for admission was only 2.0%, for community aftercare was 5.8% and for repetition of self-harm in the following 6 months was just 6.6%. Conclusions: For the purposes of suicide prevention, a low false negative rate is essential. SADPERSONS failed to identify the majority of those either requiring psychiatric admission or community psychiatric aftercare, or to predict repetition of self-harm. The scale should not be used to screen self-harm patients presenting to general hospitals. Greater emphasis should be placed on clinical assessment which takes account of the individual and dynamic nature of risk assessment.

Authors: Saunders, K; Brand, F; Lascelles, K; Hawton, K

• Publisher: BMJ Publishing Group
• Journal: Emergency Medicine Journal
• Volume: 31
• Issue: 10
• Pages: 796-798
• Publication date: 2014-10-01
• DOI: 10.1136/emermed-2013-202781
• EISSN: 1472-0213
• ISSN: 1472-0205
• Language: English