B - 135 Wisconsin Card Sorting Test 64-Card Computer Version (WCST-64-CV) Failure to Maintain Set (FMS) as an Embedded Performance Validity Indicator in a Mixed Clinical Sample

Abstract Objective The current objective was to determine the sensitivity and specificity of the Wisconsin Card Sorting Test 64-card Computer Version (WCST-64-CV) Failure to Maintain Set (FMS) as an embedded validity indicator. Methods Data were from an archival database of an academic medical center neuropsychology service. Three hundred sixty-eight adult participants completed the WCST-64-CV and Word Memory Test (WMT). Participants were defined as noncredible if all three of their WMT Immediate Recall (IR), Delayed Recall (DR), and Consistency (CNS) scores were 82.5% or below. Participants were defined as credible if all three of IR, DR, and CNS scores were above 82.5%. Forty-seven participants with scores both above and at/below 82.5% were excluded (N = 321). The sample was mean age 45 years (standard deviation [SD] = 17), mean education 13 years (SD = 2), 50% female, and 77% Caucasian. Primary diagnoses were 56.8% epilepsy and 18.1% mild cognitive impairment. No participants had dementia or an acute stroke. Results Twenty-eight participants were noncredible (8.7%) and 293 were credible (91.3%). One hundred sixty-five participants got 0 FMS (51.4%), 95 participants got 1 (29.6%), 41 participants got 2 (12.8%), and 20 participants got 3 or more (6.2%). The lowest cut-off that achieved at least 90% specificity was failure defined as ≥3. The specificity was 93.5% and the sensitivity was 3.5%. A cut-off of ≥2 achieved 80.8% specificity and 17.8% sensitivity. A cut-off of ≥1 achieved 50.8% specificity and 42.8% sensitivity. Conclusions There is no cut-off of WCST-64-CV FMS that achieves adequate specificity and sensitivity as an embedded validity indicator in this mixed clinical sample.