Rush Alzheimer Disease Center (RADC), and they were designed to be used in joint analyses to maximize sample size. ROS subjects live in communities distributed throughout the U.S., while MAP subjects live in communities in the Chicago metropolitan area. We refer to the joint dataset as “ROSMAP”. Cross-sectional assessment of cognitive performance at the last clinical evaluation can be used in analyses with neuropathology or brain omics data; however, the trajectory of cognitive decline is a more pertinent trait for drug discovery as this is the clinical outcome of interest in the vast majority of clinical trials both in the preclinical and clinical AD space. The primary trait that captures this trajectory of decline is the “Global Cognitive Slope”. It is derived from the annual neuropsychologic evaluation of each subject. 19 different neuropsychologic tests are common between ROS and MAP (of the 21 tests deployed by one or the other study), and these data are collapsed into a single “Global Cognitive Score.” The longitudinal Global Cognitive Scores are then used in a random effects model to estimate person-specific annual rates of cognitive decline controlling for known confounders such as demographics and years of education. The approach used in constructing these
