The data management approach of the ADRC is an integrated statistics, data management and data protection solution. The Data Management Core receives, stores, catalogues, tracks and integrates data generated by the Cores and Projects in the ADRC and provides expert advice for the statistical analysis of data components or integrations. At the center of the data management core for the ADRC is the overarching aim to meet the needs of NACC at every level, from data acquisition to data storage, formatting, and reporting. Implementation of this data management system is a dynamic process that aims to meet the growing and diverse needs of the ADRC. These dynamic changes include readiness to respond to changes in not only national standards for research data acquisition and maintenance (e.g., HIPAA) but also to the necessary changes and growth that occurs to NACC. Every effort has been made to systematize and unify all data collection by the clinical and neuropathology cores. This unification serves to enhance the activities of the cores, the access to the generated data by the projects, and the accuracy and ease with which data from the cores can and is reported to NACC. Thus, all ADRC clinical core sites collect a common data set in a common and identical format. This data (Standardized Clinical Dementia Evaluation (SCDE)) is completed by investigators at each site and completed forms are then input into the Data Warehouse by the Data Management core personnel. The Data Warehouse provides for multiple levels of data integrity checking, including range and logic. Once the SCDE results have been input, generating a NACC report is designed to be a turn-key operation. Similarly, all data from the neuropathology core are coded onto CERAD neuropathology battery forms that are enhanced to guarantee inclusion of all NACC neuropathology variables. Again these data are input into the Data Warehouse where data integrity is highly controlled allowing for accurate data for research use and for NACC reporting. The Data Warehouse is also designed to accommodate more idiosyncratic data sets such as those generated by Projects 1-3. By design, all data input into the Warehouse MUST include a set of common identifier fields. The use of these common identifier fields allows for the integration and cross-fertilization of project and core based data. All of the projects in this proposal will also require considerable statistical data analysis and a high level of statistical analytic sophistication. This statistical analysis advice is provided by Dr. Schmeidler who has been the statistical expert for the ADRC for over 14 years. Dr. Schmeidler is not only familiar with all of the data collected and analyzed by members of the ADRC in the past, but he has been actively involved in the planning of the currently proposed studies and in their experimental design.
Showing the most recent 10 out of 555 publications
