A-228 A Story Told in Four Parts: Embedded Validity Indicators’ Effectiveness in Predicting Medical Symptom Validity Test Validity Classification

Abstract

Abstract Objective: Prior cross-validation studies have demonstrated the utility of embedded Performance Validity Tests (PVT) as reliable indicators of performance validity, though studies of the incremental predictive power from potential combinations of embedded PVTs remain sparse. The goal of the study was to compare the relative utility of eight embedded PVTs to predict pass/fail on the Medical Symptom Validity Test (MSVT). Methods: The study comprised 225 mixed neuropsychiatric patients referred for outpatient neuropsychological evaluation. The sample was 56% female and racially diverse (40% Black, 35% White, 17% Hispanic, 5% Asian, and 2% other ethnicities). The patients were administered the following eight embedded PVTs: Rey Auditory Verbal Learning Test Forced Choice (RAVLT-FC), RAVLT Effort Score, Digit Span Age Corrected Scaled Score (DS-ACSS), Stroop Color and Word Test Word Reading T-Score, Brief Visual Memory Test-Revised Recognition Discrimination (BVMT-RD), Reliable Digit Span, Letter Fluency T-Score, and Trail Making Test Part A T-Score. The patients were classified into valid/invalid groups based on standard MSVT cut scores. Results: A forward entry logistic regression model was attained in three steps (p<.05); Step 1: RAVLT-FC (Classification Accuracy=83.2%; Nagelkerke R2 =.27), Step 2: RAVLT-FC + DS-ACSS (Classification Accuracy=84.2%; Nagelkerke R2 =.36); Step 3: RAVLT-FC + DS-ACSS + Stroop Word-T (Classification Accuracy=85.1%; Nagelkerke R2 =.39). Conclusion: The RAVLT-FC + DS-ACSS + Stroop Word-T combination provides robust predictive power. The study provided a novel approach to selecting embedded PVTs that best stand in place of the MSVT when time constraints limit the use of freestanding PVTs.