Detection of noncredible cognitive performance with Wechsler Memory Scale-IV measures in mild traumatic brain injury litigants.

To investigate the operating characteristics of selective measures on the Wechsler Memory Scale-IV (WMS-IV) to predict noncredible neurocognitive dysfunction in a sample of mild traumatic brain injury (mTBI) litigants. Participants included 110 adults who underwent a comprehensive neuropsychological examination. Criterion groups were formed based upon their performance on stand-alone measures of cognitive performance validity testing (PVT). Participants failing two stand-alone PVTs exhibited significantly lower scores across all WMS-IV dependent variables of interest compared to participants who passed both PVTs. Participants who failed one PVT were excluded. Bivariate logistic regression revealed that all six dependent variables were significant predictors of PVT status. The best prediction model consisted of three WMS-IV variables including Logical Memory Delayed Recall (LM2), Logical Memory Recognition (LMR), and Visual Reproduction Recognition (VRR). This model demonstrated an accuracy of 90.2%, 0.89 sensitivity, 0.92 specificity, and a Receiver Operating Curve (ROC) of 0.957. The current empirically-derived cutscores and logit equation for the WMS-IV may be an additional consideration in analyzing database validity and noncredible performance in mTBI personal injury litigants ages 18-69.