Montefiore Einstein Robust Geriatric Normative Project (MERGER-NP): Base rates of score discrepancies, cognitive dispersion, and impairment thresholds on the RBANS and select neuropsychological tests

<b>Objective:</b> This study developed and validated multivariate base rates for interpreting neuropsychological tests in older adults. We extended the clinical utility of the Montefiore Einstein Robust Geriatric (MERGER) norms by quantifying score discrepancies, cognitive dispersion, and low scores at different impairment thresholds on the RBANS and select neuropsychological tests. <b>Method:</b> The normative (MERGER) sample comprised 420 community-residing older adults. Base rates were calculated for select score discrepancies, cognitive dispersion (intraindividual standard deviation [ISD] and coefficient of variation [CoV]), and low scores at different impairment thresholds. Clinical validation analyses examined the sensitivity of base rate-derived cutoffs for cognitive dispersion and low scores at different impairment thresholds in identifying MCI and dementia, fixing specificity at 90% by selecting cutoffs corresponding to a 10% base rate in the MERGER sample. <b>Results:</b> Convenient look-up tables provide base rates for score discrepancies, dispersion, and low scores. Validation analyses showed that impairment thresholds and CoV from the full battery were highly effective at identifying dementia (sensitivity up to 87%, AUC up to 0.96) and moderately sensitive to MCI (sensitivity up to 54%, AUC up to 0.80). ISD, however, demonstrated modest sensitivity (23-31%) with lower AUCs (0.63-0.75). Diagnostic ability for domain-specific dispersion indices was generally lower than for the full battery. <b>Conclusions:</b> These multivariate base rates enhance the clinical interpretation of neuropsychological data in older adults by quantifying the likelihood of observed performance patterns, specifically score discrepancies, cognitive dispersion, and low score thresholds. The high sensitivity of impairment thresholds highlight their potential as diagnostic markers of dementia.