Psychopathy, risk /need factors, and psychiatric symptoms in high -risk youth: Relationships between variables and their link to recidivism

Each year, substantial numbers of youth convicted of violent and property crimes are placed on probation and released into the community. Research that attempts to identify variables associated with an increased risk for recidivism has implications for the management of these offenders and the development of appropriate treatment programs. Based on the records of 150 male youth referred to an inpatient assessment unit for medical and psychological assessment, demographic, criminal, and social history variables were coded. The Psychopathy Checklist: Youth Version (PCL:YV), Brief Psychiatric Rating Scale for Children (BPRS-C), Youth Level of Service Case Management Inventory (YLS/CMI) and the Community Risk/Needs Assessment instrument (CRNA) also were coded. These youth had significant emotional, behavioural, and situational difficulties. Most met the criteria for Conduct Disorder and received ratings of "moderate" on BPRS-C factor "Behaviour Problems" and "mild" on the factor "Motor Agitation." Using a cut-off score of 26 on the PCL:YV, the base rate of psychopathy was 25% (see Grann, Tengstrom, Langstrom, & Stalenheim, 1998). The majority of youth were assessed as high risk and high need based on the YLS/CMI and CRNA total scores. BPRS-C ratings of uncooperativeness, hostility and inattention were positively correlated with PCL:YV total and factor 2 scores and need domains (YLS/CMI and CRNA need total scores), whereas ratings of blunted affect and emotional withdrawal were negatively correlated with PCL:YV Factor 1 scores. PCL:YV total scores also were positively correlated with five of the eight YLS/CMI subscales and the CRNA need total. As compared to youth with PCL:YV scores less than 26, youth with PCL:YV scores of 26 or greater had higher scores on the YLS/CMI and CRNA need domains. These youth also scored higher on BPRS-C scales that assess uncooperativeness, hostility and manipulativeness; they scored lower on scales that assess depressed mood, blunted affect and emotional withdrawal. Youth were followed for between one and three and one half years post-release; 109 (72.6%) youth were convicted during that period (22 (14.7%) violent offences, 87 (58%) non-violent offences). The most common categories of convictions were theft (43%) and non-compliance (33%); custody and custody with probation were the most common dispositions. Univariate analyses indicated that the YLS/CMI and CRNA need total scores, PCL:YV Factor 2 and total scores, and the BPRS-C factor score that reflects attention deficit disorder symptoms were significantly related to violent and/or nonviolent recidivism. Multivariate analyses indicated that the YLS/CMI and CRNA need total scores significantly predicted recidivism; neither the PCL:YV nor the BPRS-C significantly added to the prediction equation. Overall need scores, peer relations and family variables were particularly important and should be included in risk assessments for recidivism. Moreover, these variables have implications for preventing recidivism and should be targeted for treatment, as they are dynamic and potentially changeable. Finally, to more fully address the relative importance of specific need variables and identify potential protective factors, prospective studies with multiple assessment points should be conducted.