The Data and Statistical Analysis Core (Data Core) provides two essential functions for Advancing Reliable Measurement in Alzheimer's Disease and cognitive Aging (ARMADA). The Data Core will 1) provide the data management infrastructure to permit data processing, secure housing, and sharing, and 2) serve as a central unit to direct, manage, and coordinate both psychometric and statistical analysis. The Core consists of an expert group of database managers and analysts capable of processing and analyzing diverse data types. We will develop and operate a central data repository for data collected at participating sites using the English and Spanish versions of from the NIH Toolbox for Assessment of Neurological Behavior and Function (NIHTB) batteries (cognition, emotion, sensory, motor), and the 3 proposed new tests. The Data Core will use established National Alzheimer's Coordinating Center (NACC) database mechanisms, pipelines and procedures to receive, curate, store and provide the data to the analysis team and other researchers. A key component of the Data Core is the NACC [U01 AG016976], which has been receiving data from all NIA Alzheimer's Disease Centers (ADC) for the last 18 years. These data include a standardized longitudinal clinical evaluation, a neuropsychological test battery (UDS), and ancillary data including voluntarily submitted MRI/PET imaging (DICOM), and other biomarker data. The UDS-based clinical evaluations will be important in defining the comparison groups at six of the nine sites (e.g., older normal controls [NCO], normal controls in the oldest of old [NCOO], amnestic mild cognitive impairment [aMCI], and early AD). The NIHTB Cognitive Battery will be validated against the UDS neuropsychological battery and new measures will be added to potentially extend the NIHTB (NIHTB+) to improve early detection of clinically significant cognitive change. To make meaning of these data, the Data Core comprises a diverse set of analysts, including statisticians, psychometricians, and neuro-imaging analysts. The ARMADA statistical analysis team draws heavily upon personnel from the ADCs and the Patient-Centered Assessment Resource (U2C CA186878), which is currently responsible for norming, scoring, and enhancing the NIHTB. The Data Core will validate and extend the NIHTB and NIHTB+ by conducting psychometric and statistical analyses to determine psychometric properties, structural validity, construct validity, and longitudinal validity of the existing NIHTB and NIHTB+ measures. Validation analyses will be conducted across clinical groups (AD, aMCI), NCO, NCOO, participants from underrepresented groups, and Spanish-speakers. The vision of the ARMADA Data Core is one in which data are efficiently and centrally merged and pre-processed, followed by the execution of clearly specified analysis plans and theory-driven hypotheses, thereby building the validity evidence for the use of the NIHTB+ in the detection and treatment of AD.
| Besser, Lilah; Kukull, Walter; Knopman, David S et al. (2018) Version 3 of the National Alzheimer's Coordinating Center's Uniform Data Set. Alzheimer Dis Assoc Disord 32:351-358 |
