The efficacy of single-dose (cefaclor, 2 g orally) and multidose (cefaclor, 250 mg orally three times a day for 10 days) antibiotic regimens in the therapy of acute uncomplicated urinary tract infections (UTI) in nonpregnant women were compared. The patient's clinical status and results of urine cultures were compared in retrospect with the results of the antibody-coated bacteria (ACB) test and C-reactive protein (CRP) test in order to determine if either test would predict the patient's response. Overall, 10 of 30 patients (33 percent) and 18 of 22 patients (81 percent) given single doses and multidoses, respectively, had negative urine cultures four weeks after completion of therapy. A negative urine culture at four weeks correlated with a negative ACB test utilizing the less inclusive criteria for negativity (less than 5 bacteria with fluorescence in 5 minutes of search) but not with a negative ACB test utilizing the more inclusive criteria (less than 10 percent bacteria with fluorescence) or with a negative CRP test. The cure rate in the ACB-negative single-dose group (7 of 9 patients) utilizing the less inclusive criteria for negativity was similar to the cure rate in the ACB-negative multidose group (8 of 10 patients). This study suggests that the ACB test, if properly standardized, might permit identification of a population of patients with UTI who would respond to single-dose cefaclor therapy.