Derivation of an embedded Rey Auditory Verbal Learning Test performance validity indicator

Clin Neuropsychol. 2012;26(8):1397-408. doi: 10.1080/13854046.2012.728627. Epub 2012 Oct 18.

Abstract

This study derived an embedded performance validity indicator for the Rey Auditory Verbal Learning Test (AVLT) using an archival dataset. Participants aged 20 to 65 (N = 167) who reported traumatic brain injury and completed at least two performance validity tests were included. The group who passed all performance validity measures (n = 68) demonstrated higher scores on all AVLT trials than the group who failed two or more validity indicators (n = 62). Bayesian model averaging was used to identify the optimal combination of AVLT variables for group discrimination; Total Learning and Recognition raw scores were selected. Logistic regression using these variables showed excellent discrimination with an area under the curve of. 85. The resulting AVLT performance validity index demonstrated sensitivity of. 55 with specificity of. 91. Further study of this index is warranted and cross-validation is recommended prior to clinical use.

MeSH terms

  • Adult
  • Aged
  • Area Under Curve
  • Bayes Theorem
  • Brain Injuries / diagnosis*
  • Brain Injuries / physiopathology
  • Disability Evaluation
  • Female
  • Humans
  • Logistic Models
  • Male
  • Middle Aged
  • Neuropsychological Tests / standards*
  • Neuropsychology / legislation & jurisprudence*
  • Retrospective Studies
  • Sensitivity and Specificity
  • Severity of Illness Index
  • Verbal Learning / physiology*
  • Young Adult