Assessing the Performance of Confirmatory Cognitive Tests in Screening Subjects for Early Alzheimer's Disease Clinical Trials

Our analysis demonstrated the utility of confirmatory scales to screen fail subjects lacking objectifiable cognitive decline. Depending on the instrument used, between a quarter and a third of subjects were deemed not eligible. However, substantial differences in performance between the instruments used were identified, especially in potential study participants with MMSE scores of 26 or below. While the FCSRT and WMS-LMII performed in an almost linear manner over the tested MMSE range, the RBANS performed in a curvilinear manner. Additionally, for those with MMSE scores below 25, the RBANS only identified a small subset of subjects as ineligible. Those weighing implementation of these instruments in a clinical trial should consider the presented data in the context of the MMSE inclusion criteria, subject and rater burden, and the goals of the trial.