P4‐329: Behavioural Neurology Assessment – Revised: Validation in Amnestic Mild Cognitive Impairment (AMCI)

The Behavioural Neurology Assessment-Revised (BNA-R) is an in-depth cognitive assessment tool that is intermediate between short screening tests and lengthy neuropsychological assessments. It provides a relatively quick and reliable assessment of seven cognitive domains: orientation, immediate verbal recall, delayed verbal and visual recall, delayed verbal and visual recognition, visuospatial function, working memory/attention/executive control, and language. Our objective was to validate the BNA-R for diagnosis of aMCI. Participants were assigned to the aMCI or normal control (NC) group based on detailed neuropsychological assessment. The BNA-R was administered within six months before or after this evaluation. To determine concurrent validity, we evaluated the BNA-R's ability to discriminate between aMCI and NC. Evidence for construct validity was obtained by calculating correlations between BNA-R subtests and neuropsychological tests from the aMCI and NC groups. The aMCI (n=50) and NC (n=57) groups did not differ in mean (SD) age: 77.7 ± 6.5 vs 75.3 ± 7.9, education: 15.5 ± 3.4 vs 15.0 ± 3.2 or Full Scale IQ:121.3 ± 14.0 vs 122.3 ± 13.6. There was a difference in male/female ratio: aMCI (27/23) and NC (19/38) (p=.031). With respect to concurrent validity, patients with aMCI scored significantly lower on BNA-R indices of orientation, immediate verbal recall, delayed verbal and visual recall, delayed verbal and visual recognition, and language but not on working memory/attention/executive control or visuospatial function (MANOVA). Logistic regression revealed that indices of immediate verbal recall, delayed verbal and visual recall, visuospatial function and working memory/attention/executive control correctly classified 92% of subjects (sensitivity=.92; specificity=.91). Regarding construct validity, most BNA-R subtests were significantly correlated with the neuropsychological subtests but the largest associations were between subtests of similar cognitive domains and smaller associations were between subtests of dissimilar domains. This demonstrates both convergent and divergent evidence for the constructs tested by the BNA-R. Classification rate of the BNA-R is superior to published data on the MoCA for aMCI vs NC. Setting appropriate BNA-R cut-offs can improve resource utilization by identifying patients who no longer require detailed neuropsychological assessment for a high probability diagnosis of aMCI, or when aMCI is effectively ruled out.