Settling the Score: Can CPT-3 Embedded Validity Indicators Distinguish Between Credible and Non-Credible Responders Referred for ADHD and/or SLD?

Anthony Robinson, Louisiana State University, Baton Rouge, USA.
Christopher Reed, Louisiana State University, Baton Rouge, USA.
Katrail Davis, Louisiana State University, Baton Rouge, USA.
Ross Divers, Louisiana State University, Baton Rouge, USA.
Luke Miller, Louisiana State University, Baton Rouge, USA.
Laszlo A. Erdodi, University of Windsor, Ontario, Canada.
Matthew Calamia, Louisiana State University, Baton Rouge, USA.

Abstract

OBJECTIVE: The purpose of the present study was to further investigate the clinical utility of individual and composite indicators within the CPT-3 as embedded validity indicators (EVIs) given the discrepant findings of previous investigations. METHODS: A total of 201 adults undergoing psychoeducational evaluation for ADHD and/or Specific Learning Disorder (SLD) were divided into credible ( = 159) and non-credible ( = 42) groups based on five criterion measures. RESULTS: Receiver operating characteristic curves (ROC) revealed that 5/9 individual indicators and 2/4 composite indicators met minimally acceptable classification accuracy of ≥0.70 (AUC = 0.43-0.78). Individual (0.16-0.45) and composite indicators (0.23-0.35) demonstrated low sensitivity when using cutoffs that maintained specificity ≥90%. CONCLUSION: Given the lack of stability across studies, further research is needed before recommending any specific cutoff be used in clinical practice with individuals seeking psychoeducational assessment.