Utilization of the emergency room at an urban community hospital is studied in a format designed to accomplish three complementary objectives: 1) to characterize a sample of individual patients, rather than an unweighted sample of visits, 2) to estimate the number of individuals served during a specified period and the magnitude of the relationship between these patients and the utilization of other hospital services, and 3) to introduce the patient's “frequency-of-visit” as an important variable in the analysis of emergency room utilization.Some specific findings are: 1) the vast majority of patients who used the emergency room did so very infrequently; 46,527 visits were made in one year by an estimated 34,286 different patients; 2) an estimated 2,586 patients made three or more visits during the year; a disproportionately large number of these “high-frequency” users were black, low-income, and from inner-city areas; a relatively small percentage of their visits were for accidental injury, 3) approximately 53 per cent of the hospital's inpatient admissions and 68 per cent of the inpatient days were generated by patients who also made at least one emergency room visit during the year studied.