We aimed to find a valid cutoff score for the Screen for Child Anxiety Related Emotional Disorders, child (SCARED-C) and parent (SCARED-P) Spanish versions for detecting Anxiety Disorders (AD) in a non-clinical population. The predictive accuracy of the SCARED-C and SCARED-P was assessed using the Area Under the Curve (AUC) of ROC curves. In general, the predictive accuracy of the SCARED-C (full version, short version, and four factors) was good and better than that of the SCARED-P. To differentiate between children who meet the diagnostic criteria for any AD and children who do not, we propose cutoff scores of 25 and 17 for the SCARED-C and SCARED-P, respectively. The sensitivities are 75.9% and 62.8%, and the specificities are 68.5% and 69.5%. The SCARED-C factor that had the best predictive accuracy was Somatic panic followed by Separation Anxiety, Generalized Anxiety and Social Phobia. The SCARED-P factor with the best predictive accuracy was Separation Anxiety. The results support the use of SCARED-C as a screening test for Anxiety disorders while SCARED-P should only be used as complementary information.
Copyright © 2012 Elsevier Ltd. All rights reserved.