Utility of various WAIS-IV Digit Span indices for identifying noncredible performance validity among cognitively impaired and unimpaired examinees

Abstract

Objective: Wechsler Adult Intelligence Scale-Fourth Edition (WAIS-IV) Reliable Digit Span (RDS), RDS-revised (RDS-R), and age-corrected scaled score (ACSS) are validated and commonly used embedded performance validity tests (PVTs), though existing validation studies have largely examined younger (approximate ages 19–35) patients with mild traumatic brain injury or those without cognitive impairment. This study compared the classification accuracy of RDS, RDS-R, and ACSS in a mixed clinical sample of relatively older (M age = 54.61) veterans with and without neurocognitive impairment. Method: During a comprehensive neuropsychological evaluation, 113 clinically-referred veterans completed the WAIS-IV Digit Span subtest and the following criterion PVTs: Dot Counting Test, Word Choice Test, and Test of Memory Malingering. Those with ≤1 criterion PVT failure were classified as valid (n = 87), whereas those with ≥2 failures were classified as noncredible (n = 26). Among valid participants, 49% were cognitively impaired. Results: RDS, RDS-R, and ACSS all significantly predicted validity group membership with respective areas under the curve (AUCs) of .79, .81, and .85, and optimal cut scores of RDS ≤ 5, RDS-R ≤ 9, and ACSS ≤ 5. Lower accuracy and AUCs were observed for the valid-cognitively impaired subsample across indices, but to a greater degree for traditional RDS. ACSS evidenced maximal sensitivity/specificity for the total sample (≤5; .62/.87), cognitively unimpaired subsample (≤5; .62/.95), and cognitively impaired subsample (≤4; .39/.86). Conclusions: ACSS yielded better classification accuracy and sensitivity/specificity than RDS and RDS-R. While all three indices have utility as embedded PVTs, ACSS ≤ 5 may be most robust to cognitive impairment while identifying noncredible performance.