In a study to determine the risk factors for urinary tract infection in college-aged women, women who presented with acute urinary tract infection to the student health service were compared to women without bacteriuria who presented with complaints of other acute illnesses. Among women who were sexually active, the following multivariate adjusted odds ratios (95% confidence intervals) were found: intercourse in the previous 48 hours, 58.1 (11.9 to 284.1); intercourse only in the previous 3 to 7 days, 9.1 (1.9 to 44.1); diaphragm use in the previous 48 hours, 8.4 (3.4 to 21.1); urination after intercourse, 0.5 (0.3 to 0.9); and past history of urinary tract infection, 2.7 (1.5 to 5.0). Several other factors previously postulated to be related to urinary tract infection were found not to be associated, including oral contraceptive use, tampon use, and direction of wiping after a bowel movement. When the women with symptomatic bacteriuria were compared to women with asymptomatic bacteriuria, the results were similar, except diaphragm use and urination after intercourse were no longer associated with urinary tract infection. When the women with asymptomatic bacteriuria were compared to women without symptoms and without bacteriuria, diaphragm use remained the only statistically significant risk factor. These findings should be taken into account in attempts to prevent urinary tract infection, as well as in subsequent studies of this disease.