The Word Memory Test genuine memory impairment profile discriminates genuine memory impairment from invalid performance in a mixed clinical sample with cognitive impairment

Objective: The Word Memory Test (WMT) is a memory-based performance validity test (PVT) with adjusted interpretive criteria (Genuine Memory Impairment Profile; GMIP) proposed for those with cognitive impairment (CI). The GMIP has been criticized for poor discriminability; thus, this study sought to validate the GMIP in a mixed clinical sample. Analyses aimed to demonstrate enhanced detection of invalid neuropsychological test performance while minimizing false positives in a sample including patients with mild CI. Method: Data included 116 Veterans who completed the WMT and four criterion PVTs during clinical evaluation. This sample was 37.1% valid-CI, 33.6% valid-unimpaired, and 29.3% invalid per criterion PVTs. Group differences in WMT performance and diagnostic accuracy were assessed. Results: WMT performance significantly differed between validity groups (Wilk's Lambda = .40, F[6, 109] = 27.62, p < .001, eta(2)(p) = .60) with invalid participants scoring significantly lower across all WMT indices, with larger effect sizes for the effort subtests (eta(2)(p) = .44-.55) than memory subtests (eta(2)(p) = .16-.32). CI also had a significant effect on WMT performance (Wilk's Lambda = .65, F[6, 75] = 6.66, p < .001, eta(2)(p) = .35). Those with CI scored lower across all indices, with larger effect sizes for the memory subtests (eta(2)(p) = .29-.30) relative to effort subtests (eta(2)(p) = .18-.20). Standard WMT pass/fail criteria were sensitive (.97), but had unacceptable specificity (. 66). GMIP sensitivity (.94) and specificity (.91) were robust with a DOR of 171.43. Conclusions: WMT indices were more significantly affected by performance validity than memory. However, with CI, GMIP criteria is essential to improve diagnostic accuracy and reduce false positive errors when identifying invalid performance.