This report describes a rapid method for the identification and susceptibility testing of Escherichia coli bacteriuria by use of the Autobac urine screen (AUS), a 5-h indole test, and the AutoMicrobic system E. coli Susceptibility Card (AMS-ECSC). All specimens that were AUS negative at 3 and 5 h were reported as negative. All specimens that were AUS positive at 3 or 5 h were removed from the refrigerator and streaked to a MacConkey-blood agar biplate with a 0.001-ml calibrated loop and incubated at 35 degrees C. The standard method for identification and susceptibility testing consisted of inoculating isolated colonies into the AutoMicrobic system Enterobacteriaceae-Plus Biochemical Card and the AutoMicrobic system General Susceptibility Card. Of 915 specimens, 212 (23.2%) were AUS positive at 3 h, of which 112 (52.8%) were indole positive when tryptophan broth was tested at 5 h. The sensitivity and specificity of this screening method for E. coli bacteriuria were 78.8 and 72.2%, respectively. If contaminated cultures were excluded, specificity was 94.4%. When the indole test was positive, 10 microliters of growth from tryptophan broth was used as inoculum for the AMS-ECSC. AMS-ECSC results were final at 12 h. The sensitivity and specificity of the AMS-ECSC for identification of E. coli were 90.4 and 55.0%, respectively. If contaminated cultures were excluded, specificity was 100%. AMS-ECSC susceptibility results demonstrated an overall agreement of 94.3% with the standard method, with 0.5% very major, 3.4% major, and 1.8% minor discrepancies. The biplate was examined after overnight incubation, and when colonies compatible in morphology with E. coli were present in significant numbers, AMS-ECSC results were reported. When discrepancies were found between biplate and AMS-ECSC results, the biplate was processed in the conventional manner. A rapid method for identifying and performing susceptibility tests for approximately 70% of the specimens with E. coli bacteriuria is described.