Software Technology for Cancer Epidemiology and Prevention

The goal of the proposed project is the creation of a web-based software application that efficiently and flexibly integrates and extracts multiple-sourced epidemiological data for a wide array of study designs. We envision software that links existing databases at appropriate levels of aggregation, most notably at person-level, and assists researchers in dynamically applying inclusion and exclusion criteria to create cohorts or select cases and controls, merging and splitting datasets, selecting fields for export, visualizing the process and results, and exporting datasets to statistical analysis software. We propose to build this software on our patented technology platform which provides the advantages of easy extensibility for additional data sources and applications, continuous upgrade while the system is live, software modification without core engine rebuild, and a modular design approach which makes simultaneous modification of domain logic and user interface possible. In Phase I, we will analyze the databases and data dictionary for best approaches to data linkage and extraction, create usability diagrams to guide the software development, and describe our findings in a written feasibility and approach report. Database linkage, dataset manipulation, and data import into statistical software is currently time-consuming and difficult to reproduce; a software tool that expedites the path from data collection to data analysis will be critical for improved response time from study to implemented protective measures for prevention of disease.