PCL-R and CAPP-IRS Show Incremental Validity in Schizophrenia

The PCL-R may be a better tool for predicting risk and life-time dimensions than CAPP-IRS.

The Psychopathy Checklist-Revised (PCL-R) and Comprehensive Assessment of Psychopathic Personality-Institutional Rating Scale (CAPP-IRS) have different relative strengths in use with patients with schizophrenia disorders, with the PCL-R being better for assessing risk and the CAPP-IRS having better associations with clinical features, according to a study recently published in Psychiatry Research.

The study researchers included 72 male forensic patients with schizophreniform disorders and at least 6 months of treatment in this study, meant to assess the relative reliability and validity of each scale. The Shapiro-Wilk’s normality test was completed along with descriptive statistics computations, then the reliability of both the CAPP-IRS and PCL-R tests were analyzed as follows. The study researchers chose 15 random protocols from CAPP-IRS and blindly coded them twice to extract Intra-Class Coefficients (ICC1).

These coefficients were interpreted with the following scores: >.50 as moderate, >.70 as good, and >.90 as excellent. Associations among the PCL-R and CAPP-IRS were investigated with partial and bivariate correlations. Cohen’s q was used to compare relative correlations between the two. The PCL-R’s and CAPP-IRS’s predictive values were analyzed using stepwise linear regression. The Positive And Negative Symptoms in Schizophrenia (PANSS) scale was first used to score and report symptoms of schizophrenia, followed by either of the psychopathy scales.

With the exception of emotional and cognitive domains, the CAPP-IRS scales all demonstrated an acceptable level of internal reliability, with attachment ICC1 =.52, dominance ICC1 =.78, behavior ICC1 =.85, self ICC1 =.88, and emotional ICC1 =.67. Cognitive domain ICC1 was not acceptable, at .29. There was moderate correlation between the PCL-R scales and the CAPP-IRS domains of dominance, behavior, and self, suggesting an overlap in validity. There was no association between the PCL-R scales and CAPP-IRS domains of cognition, emotion, and attachment, meaning CAPP-IRS may have value in assessing psychopathy factors not measurable by PCL-R. 

Related Articles

Both scales showed a moderate correlation with PANSS, though there was some overlap between dimensions of psychopathy and symptoms of schizophrenia. This overlap was more evident in the CAPP-IRS and symptoms of schizophrenia than in the PCL-R and symptoms of schizophrenia. There was an association between PCL-R and historical risk factors, but no such association existed with the CAPP-IRS. 

The researchers conclude that “the PCL-R and the CAPP-IRS converge in many aspects, but both have unique incremental validity. The PCL-R is closer to risk assessment tools, whereas the CAPP-IRS shows a closer and more differentiated associations with clinical features.” 


De Page L, Mercenier S, Titeca P. Assessing psychopathy in forensic schizophrenia spectrum disorders: validating the Comprehensive Assessment of the Psychopathic Personality-Institutional Rating Scale (CAPP-IRS)Psychiatry Res. 2018; 265:303-308.