Citation

Warren JI, South SC, Burnette ML, Rogers A, Friend R, Bale R, Van Patten I. Int. J. Law Psychiatry 2005; 28(3): 269-289.

Affiliation

Clinical Psychiatric Medicine, University of Virginia, United States. JIW@hscmail.mcc.virginia.edu

Copyright

(Copyright © 2005, Elsevier Publishing)

DOI

10.1016/j.ijlp.2003.09.012

PMID

15923037

Abstract

This study explores the performance of 132 female maximum-security inmates on the Psychopathy Checklist-Revised (PCL-R) and the HCR-20 (Historical, Clinical, and Risk Management Scheme) to examine the concordance between these two risk assessment instruments, and to assess their potential usefulness in determining level of risk for violent behavior and other forms of criminality. The two instruments demonstrated consistent and highly significant correlations across total scores, factor scores, and subscale scores. When the two instruments were entered into a multiple regression analysis to predict violent and non-violent crime, the HCR-20 did not add to the variance explained by the PCL-R. These results confirm earlier research that suggests that there is little or no difference between these two risk assessment instruments in their relationship to community or institutional violence. Further, Receiver Operating Characteristics (ROC) analyses show that both instruments demonstrated an inverse ability to predict convictions for murder, a close to chance ability to predict violent crime, but a shared ability to predict property and minor crime. Broadly, these results suggest that psychopathic women are involved in chronic patterns of non-violent criminality, while women charged and convicted of murder generally do not have elevated scores on the PCL-R or HCR-20. The relevance of these findings to rehabilitation and treatment is discussed.

Language: en