SAFETYLIT WEEKLY UPDATE

We compile citations and summaries of about 400 new articles every week.
RSS Feed

HELP: Tutorials | FAQ
CONTACT US: Contact info

Search Results

Journal Article

Citation

Randall JR, Colman I, Rowe BH. J. Affect. Disord. 2011; 134(1-3): 348-355.

Affiliation

Department of Public Health Sciences, School of Public Health, University of Alberta, Edmonton, Canada.

Copyright

(Copyright © 2011, Elsevier Publishing)

DOI

10.1016/j.jad.2011.05.032

PMID

21658779

Abstract

BACKGROUND: Assessment of self harm risk is both a common and difficult task in emergency room settings. Psychometric measures have been developed to help with this assessment but it is uncertain how well these measures perform and which are clinically useful for assessment in this setting. METHOD: Two reviewers independently assessed studies for relevance, inclusion, and study quality. Included studies classified mostly adult patients at risk for self-harm treated in an ED. The outcome variables selected were recurrence of self-harm/suicidal ideation or hospitalization. Only cohort study designs with follow-up were eligible. RESULTS: From 556 potentially relevant abstracts, 12 studies were identified for inclusion in the review. Overall, the risk of bias was considered moderate to low in this review. Of the included studies reporting future self harm as an outcome measure, only the scales that are part of the Manchester self harm project, the Implicit Associations Test and the Violence and Suicide Assessment Form were found to successfully predict self harm. The four studies that assessed admission as an outcome utilized eight different actuarial methods. Of the scales assessed, six were found to be significant predictors of admission. LIMITATIONS: It was not possible to perform a meta-analysis with the studies detected and it is uncertain whether publication bias or selection bias within the reviewed studies affected the results. CONCLUSION: Overall, while many methods used in the ED to assess suicidal and parasuicidal patients have strong psychometric properties, there is little clinical evidence supporting their use.


Language: en

NEW SEARCH


All SafetyLit records are available for automatic download to Zotero & Mendeley
Print