Composite Endpoints Sensitive to the Earliest Stages of Alzheimer's Disease Lead Investigator: Steven Edland Institution : UCSD E-Mail : sedland@ucsd.edu Proposal ID : 613 Proposal Description: Composite endpoints defined as the weighted average of component scales will be constructed from the C1/C2 neuropsychometric battery. The performance of optimally weighted composites (as defined in Ard et al, Pharmaceut. Statist. 2015, 14 418???426) will be characterized, with focus on the reproducibility of weight estimates when calculated from subsets of the data, and with investigations of power to detect subtle cognitive decline in normal elderly, e.g., the power to detect differences in rate of decline in E4+ versus E4- controls, and the power to detect differences in rate of decline in subjects who do and do not convert to CDR>0 on followup visits.