ENHANCED FOOD COMPOSITION RESEARCH DATA BASE

EPIDEMIOLOGICAL RESEARCH ON DIETARY PRACTICES IS MAKING A SIGNIFICANT CONTRIBUTION TO A REDUCTION IN CANCER MORTALITY. HOWEVER, PROGRESS IS IMPEDED BY CURRENT NUTRITION DATABASES THAT DIFFER SIGNIFICANTLY IN FOOD COMPONENTS OF INTEREST IN CANCER RESEARCH. THESE DISAGREEMENTS ARE CAUSED BY FOOD CODING DISCREPANCIES, USE OF DIFFERENT DATA, AND MANY OTHER CAUSES. THE ENHANCED FOOD COMPONENT RESEARCH DATA BASE WILL ALLEVIATE THIS PROBLEM THROUGH USE OF THE U.S. FOOD AND DRUGADMINISTRATION FACTORED FOOD VOCABULARY (FFV), HIGH-SPEED DATA RETRIEVAL MECHANISMS, AND TEXT FILES TO ASSIST IN THE INTERPRETATION OF RESULTS AND IN COMBINING DATA SOURCES. THE SYSTEM HAS MODEST HARDWARE RESOURCES REQUIREMENTS AND CAN BE RUN ON MANY LAPTOP COMPUTERS. THE SOFTWARE ARCHITECTURE IS MODULAR AND FACILITATES THE INTEGRATION OF ADDITIONAL FOOD DATABASES (E.G., INTERNATIONAL FOOD DATABASES). DURING PHASE I, A SYSTEM CAPABLE OF HANDLING U.S. DEPARTMENTOF AGRICULTURE HANDBOOK 8 DATA USING EITHER THE FFV OR A USER-DEFINED THESAURUS WAS BUILT. THE SYSTEM WAS ALSO TESTED AT SITES WHERE EPIDEMIOLOGICAL DATA ARE BEING ENTERED. IN PHASE II, THE SYSTEM WILL BE FULLY IMPLEMENTED WITH ACCESS TO MULTIPLE DOMESTIC AND FOREIGN DATABASES, IN-DEPTH EXPERT ASSISTANCE FOR BOTH THE FFV AND INTERPRETATION OF NUTRIENT DATA RESULTS, FLEXIBILITY OF OUTPUT FORMATS, AND THE CAPABILITY TO EASILY UPDATE EXISTING RECORDS.