and the final version is expected to have full capability for users to search by the MeSH terms as well as by key words.

Review of the recommendations as a whole revealed that it would be helpful to categorize types of recommendations. In particular, staff used a two-tiered system that distinguishes “major knowledge gaps” from “knowledge gaps” (see Appendix D for criteria used). In addition, some of the research recommendations fit into a third category called “research methods.” Examples include recommendations related to databases and nutrient assays.

Review of all the recommendations also made it apparent that it would be useful to identify cross-cutting topics. Obvious topics include age groups (e.g., adolescents, children, infants, elderly), biomarkers, dietary supplements, lactation and pregnancy, and the Tolerable Upper Intake Levels (ULs) and toxicity. Additional suggestions for cross-cutting topics would be welcome.

In developing the database, staff used two types of software—Excel and Access—both from Microsoft.³ Each has advantages and disadvantages. Excel is widely available, many investigators use the software for a variety of purposes, and most users find data entry to be quite easy. However, Excel does not allow Boolean searches and has limited printing options. Access is less widely available, less familiar, and less user-friendly than is Excel. However, Access has a number of features that are highly desirable in a searchable database and that are not available in Excel. These features include capabilities to conduct tailored queries, save useful queries, and produce custom printouts of desired information.

To make the database as user-friendly as possible, staff placed two releases on the Institute of Medicine (IOM) website at www.iom.edu/ DRIresearch2006 and sought feedback from users. The remainder of this chapter summarizes how to access and use the database. Additional information about the database appears in Appendix D.