At AMIA 1997, we reported on the design and development of a new computer-based tool, called QID, for empiric antibiotic decision support. QID was designed to help physicians identify the antibiotic regimens with the highest probability of covering the pathogens that are most likely to be present in individual patients. QID creates a list of antibiotics, ordered by potential benefit in treatment, for a patient with a suspected infection before culture results are available. Since our initial publication, a "before and after" study has been done using 20 internal medicine residents and the same number of internal medicine attendings. In order to test the hypothesis that physician's would make more appropriate empiric antibiotic choices with the aid of QID, we chose University of Utah physicians and had each evaluate four infectious disease cases that were abstracted from medical record infectious disease cases. Immediately following their initial review and determination of antibiotic therapy for each case, the study participants were presented with QID's antibiotic recommendations on the same case to see if this information would change their initial drug regimen. The tool was shown to have a greater impact on the most difficult cases but statistically improved scores overall (p < .001). Details of our study design and results are presented.