Predicting Seclusion in a Medium Secure Forensic Inpatient Setting

Abstract

The current study investigated the predictive validity of several popular risk-related assessment instruments with respect to seclusion. The Hare Psychopathy Checklist-Revised (PCL-R; Hare, 2003), Version 2 of the HCR-20 (HCR-20; Webster, Douglas, Eaves, & Hart, 1997), the Violence Risk Appraisal Guide (VRAG; Harris, Rice, & Quinsey, 1998), and the Level of Service Inventory-Revised (LSI-R; Andrews & Bonta, 1995) were coded from institutional files for a sample of 130 patients from a medium-secure forensic inpatient unit. Seclusion was indexed in terms of total number of seclusions during a period of two years and total time spent in seclusion. ROC analyses indicated that all instruments had small to moderate and significant predictive validity with respect to frequency of seclusion, but were less strongly predictive of duration of seclusion. Overall, Factor 2 of the PCL-R was the best predictor of seclusion. Implications for practice and the necessity for future and local research are discussed.

Keywords:

• inpatient violence
• risk assessment
• forensic
• seclusion

Acknowledgments

The authors wish to express their gratitude to Dr. John F. Newbauer and Dr. Mark H. Stone of the Adler School of Professional Psychology for their many helpful suggestions as members of Dr. Reimann's dissertation committee. We also express our appreciation to Dr. Marc Levi, Staff Psychologist, Forensic Program and the Clinical Information Services Staff at the Ontario Shores Centre for Mental Health Sciences (formerly Whitby Mental Health Centre) for their help in accessing the data necessary for this study. We also wish to thank Dr. Wagdy Loza's two anonymous reviewers for their helpful comments on an earlier draft of this article.