Usefulness of Aggressive Behaviour Risk Assessment Tool for prospectively identifying violent patients in medical and surgical units

Abstract

Aim: The aim of this study is to evaluate the usefulness of the Aggressive Behaviour Risk Assessment Tool for prospectively identifying violent patients in medical-surgical units.

Background: Although patient violence against nurses is a serious occupational hazard, there is a lack of simple screening tools with acceptable sensitivity and specificity for identifying potentially violent patients in medical-surgical units.

Methods: A prospective cohort study involving patients admitted to six medical-surgical units at an acute care hospital was conducted from August 2009 to December 2009. Primary nurses completed the 17-item checklist within 24 hours of admission. A second identical checklist was completed by another nurse to assess the inter-rater reliability. Following a violent event or just prior to discharge, the violent event outcome section was completed to collect information about violent event, if any. A multivariate logistic regression model with backward elimination was used to select a set of parsimonious items that best predict violent behaviours.

Results: Fifty-six patients out of 2063 (2·7%) had one or more violent events. A parsimonious set of ten items were selected for the tool. Receiver Operating Characteristics analysis of Aggressive Behaviour Risk Assessment Tool showed that the area under the curve was 0·82 (95% Confidence Interval, 0·75-0·90). The sensitivity and specificity at the cut-off score of 1 were 70·9% and 89·3%, respectively. The Cohen's Kappa for inter-rater reliability was 0·647.

Conclusion: The Aggressive Risk Assessment Tool is a simple, easy-to-use assessment tool with acceptable inter-rater reliability, sensitivity and specificity that may be useful for prospectively identifying violent patients in medical-surgical units.