INTERNET ANALYSIS OF ELECTRONIC MEDICAL RECORD DATA

Medical University Of South Carolina, Charleston SC

Investigators

Abstract

This revised application for fellowship in Applied Informatics will augment the applicant's knowledge, awareness, and training with the ultimate goal of using information technologies to improve the quality of health care. The fellowship will provide interdisciplinary training in analytical and mathematical sciences in the areas of biostatistics, epidemiology, biomathematical modeling, and biomedical computing through a formal program of study in the Department of Biometry and Epidemiology at MUSC. The applicant will gain technical skills and experience in advanced database design and maintenance. She will learn, develop, apply, and test platform-independent (Web) access tools to the Practice Partner Research Network (PPRNet) database. The PPRNet database contains longitudinal data extracted from electronic medical records of 479,000 patients. The ability to access and analyze clinical process and outcome information through the Web will offer unique opportunities to study epidemiology of disease, evaluate one's own practice performance and drive quality improvement. The specific aim of the research project is to create a practical, secure, platform- independent Internet (Web) interface to a research repository database of longitudinal clinical data extracted from electronic medical records. The PPRNet database will be re-designed, using an efficient and powerful database. Data will be loaded, aggregated, and bridged to common nomenclature dictionaries and a system for update and maintenance of the bridging tables established. Web access and query functions will be developed for research in monitoring improvements in process and outcome measures. These functions will facilitate access to static practice performance assessment reports of these measures, and the capability of PPRNet members to perform dynamic analyses of the database. Appropriate techniques for this access, including those that are most efficient, user-friendly, and secure will be studied, developed, tested, implemented, and evaluated.