Development and validation of a brief screening procedure for pediatric obsessive-compulsive disorder derived from the Spence Children's Anxiety Scale

Evidence-based assessment (EBA) is essential to accurate measurement of psychiatric disorders, including pediatric obsessive-compulsive disorder (OCD). There are, unfortunately, barriers to using these techniques in clinical settings, primary among which is the time entailed in instrument administration. The current study applied decision-tree statistics to parent and child forms of the OCD subscale contained within a commonly used pediatric anxiety assessment tool (the Spence Children's Anxiety Scale) with an emphasis on abbreviating the measure. The end product of this examination was a pair of algorithms derived from analysis of a sample containing both clinical cases who presented for treatment and community controls (n = 1094 parent/children dyads in total). These were noted to be either statistically significantly superior to or not different from the lengthier SCAS/P OCD subscales in terms of common metrics of clinical utility (i.e., sensitivity, specificity, positive predictive value, and negative predictive value) despite containing only 1 - 2 items each. The results demonstrate feasibility of data reduction strategies to improve clinical implementation of EBA, particularly decision-tree models. The fit of this instrument into a clinical setting is discussed, as are future extensions of these methods to diverse problem sets in the context of integrated healthcare services.