We conducted a prospective study comparing the diagnostic yield of standard clinical examinations and of comprehensive microbiological studies in establishing the etiology of genitourinary (GU) symptoms and the prevalence of GU tract infections in 204 women seen in a university-based family practice. Two thirds were initially seen with GU symptoms and one third were seen for routine examinations. In each case we obtained demographic and historical information, a physical examination, and a variety of laboratory tests, including research procedures not commonly available. Diagnoses considered were urinary tract infection and sterile pyuria; trichomonal, yeast, and nonspecific vaginitis; and GU tract infection with Neisseria gonorrhoeae, Chlamydia trachomatis, or herpes simplex virus. Using strategies commonly employed in practice, we reached a diagnosis in only 34% of symptomatic women, a figure rising to 66% when selected, nonroutine laboratory examinations were added.