To investigate the effect of concurrent antibiotic therapy on bacterial isolation, 1001 consecutive urine cultures obtained from hospitalized patients were evaluated. Of the cultures, 121 (12.1%) had positive results. The administration of antibiotic therapy substantially changed urine culture results in uncatheterized patients. Only one (0.7%) of 135 cultures obtained from patients without a recent history of catheterization was positive when antibiotics were being administered. In contrast, 43 (8.6%) of 501 urine cultures obtained from uncatheterized patients not receiving antibiotic therapy were positive. These results indicate that concurrent antibiotic therapy should be considered before ordering urine cultures in uncatheterized patients. Substantial savings would result if, in this situation, cultures were limited to patients with urinary tract symptoms or a history of urinary tract pathology.