A Comprehensive Examination of Clinical Cutoff Scores for the Inventory of Callous-Unemotional Traits (ICU)

Abstract

Youth with elevated callous-unemotional (CU) traits represent a clinically important subgroup of youth who display particularly severe conduct problems and antisocial behavior and thusly impose great costs to themselves, other individuals, and society. The recent addition of the specifier for CU traits, “with Limited Prosocial Emotions (LPE),” to major classification systems has prompted the need for comprehensive and valid assessment tools that aid in the identification of these traits. One such tool is a multi-informant questionnaire, the Inventory of Callous-Unemotional Traits (ICU). However, a major limitation of this measure is a lack of well-validated cutoff scores. With this, the present study compared the clinical utility of various proposed cutoff methods and scores (i.e., empirically derived cutoffs using receiver operating characteristic (ROC), normative cutoffs, and four-item approximations of LPE criteria) in both a longitudinal sample of justice-involved adolescents (N = 1,216; Mage = 15.29, SD = 1.29) and a cross-sectional sample of community youth (N = 289; Mage = 11.47 years; SD = 2.26). A series of chi-square and independent-samples T-tests were conducted with the use of clinically important validators (e.g., juvenile delinquency, social rejection), and average effect sizes were computed. As a result, comparisons of clinical utility are made and a range of optimal cutoff scores, for each ICU informant version, are provided for the detection of elevated CU traits. The outcomes of this study have important practical implications as they provide researchers and clinicians with guidance on choosing cutoffs that may aid in determining elevations of CU traits.