This prospective case-controlled study was performed to compare the time intervals of a consult emergency department (ED) admission process with an emergency medicine (EM) service admission process. During March 1994, the consultant services admitted 307 patients for hospitalization at an urban tertiary academic ED with an EM residency; in April 1994, the EM service admitted 264 patients. The times measured were: 1) triage to examination room; 2) room to first physician contact; and 3) emergency physician contact to admit request. Data analysis was by mode and Wilcoxon tests. We analyzed 537 evaluable admissions. The mode for consultant process was 205 min, and the mode for emergency medicine admissions was 158 min from first physician contact to admit request, for a 47-min difference. All patients arrived stable to an inpatient bed; none was transferred to the Intensive Care Unit or to an operating room in the first 24 h. Concordance of the ED admitting impression and the hospital discharge diagnosis was 99% (259/264). We conclude that in selected tertiary academic EDs, admission of all patients by the EM service is more efficient than a consultant-admission process. Outcomes show the EM admission process may be employed safely and with accurate patient diagnosis.