A predictive model of criminality in civil psychiatric populations

Purpose
The purpose of this paper is to develop a predictive model of criminal risk in civil psychiatric populations, by determining the relative impacts of psychopathy, drug use, impulsivity and intelligence on levels of criminality.

Design/methodology/approach
The sample consisted of 871 civil psychiatric patients, selected from the MacArthur Violence Risk Assessment Study, who had been diagnosed with a mental illness or personality disorder, and hospitalised less than 21 days. Each participant was administered the Hare Psychopathy Checklist Screening Version (PCL:SV), Barratt Impulsiveness Scale (BIS-11) and the Wechsler Adult Intelligence Scale (WAIS-R). In addition, information on background demographics, drug use and criminality was obtained via a self-report questionnaire.

Findings
Pearson correlations identified significant positive relationships between past arrests, psychopathy, impulsivity and drug use. Intelligence was negatively related to past arrests. Multiple regressions identified a significant main effect for Factor 2 psychopathy on past arrests when controlling for all covariates, but not for Factor 1 psychopathy, intelligence or impulsivity. Drug use and gender had small univariate effects.

Research limitations/implications
It is suggested that future research investigates the influence of specific mental disorders on different types of offending.

Originality/value
By investigating predictors of criminal behaviour in civil psychiatric patients, the present study makes valuable contributions to the research literature, enhancing our theoretical understanding of the relationships between psychopathy and criminality/recidivism. It also has notable implications in applied practice, for example in the development and refinement of risk assessment methods.