27.2 Examining the Psychometric Properties and Clinical Utility of the Toronto Obsessive-Compulsive Scale (TOCS) in a Pediatric Clinical Sample

Abstract

This session aims to examine the psychometric properties and establish clinical cutoffs for the parent-report version of the Toronto Obsessive-Compulsive Scale (TOCS), a 21-item screening tool that examines obsessive-compulsive traits. Participants had a primary diagnosis of OCD (n = 350; 50% female; ages 6-20 years), ADHD (n = 835; 25% female; ages 5-20 years), autism spectrum disorder (ASD; n = 805; 22% female; ages 2-21 years), or were typically developing controls (n = 398; 51% female; ages 3-21 years). Data were collected from 2 tertiary care clinics as well as from the Province of Ontario Neurodevelopmental Disorders (POND) Network. Confirmatory factor analyses, internal consistency, and convergent and divergent validity of the TOCS were examined in the OCD group. Receiver operating characteristic (ROC) analyses were used to establish a clinical cutoff by splitting the OCD and control groups into a discovery sample (166 OCD cases, 164 controls) and a validation sample (184 OCD cases, 234 controls). TOCS scores were compared across clinical groups using multivariate analysis of variance (MANOVA). The 6-factor structure of the TOCS was confirmed (χ2 [137] = 596.60; CFI = 0.90; RMSEA = 0.09; 90% CI, 0.09-0.10; SRMR = 0.06) with each factor demonstrating strong reliability (α range 0.71-0.92) and validity (eg, r = 0.40 with existing OCD measure). ROC analyses indicated excellent diagnostic discrimination of OCD cases from controls (area under the curve ≥ 0.95), with excellent sensitivity (77%-92%) and specificity (92%-98%) in both the discovery and validation samples. The OCD group scored significantly higher on all TOCS dimensions than the ADHD and ASD groups (all p < 0.001), with the exception of hoarding. The TOCS is a reliable and valid screening tool for pediatric OCD, with a strong ability to discriminate OCD cases. It is a multipurpose quantitative OCD measure that is useful not only for research (including genome-wide association studies), but also as a clinical screening tool.