A study was undertaken to determine whether hospital bed utilization is different between teaching and nonteaching hospitalist services. During a 2-year period, the average length of stay (ALOS), case-mix index, readmissions within 30 days of discharge, and percentage of 1-day stays were compared between these services. To ameliorate confounders, the ALOS was recalculated (ALOS revised) after excluding patients with length of stay of more than 20 days and those originally admitted to the intensive care unit. Six thousand ninety-nine patients were discharged by the teaching service (group 1), 14 348 by the corporate hospitalist (group 2), and 2634 by the community-based practitioner group (group 3). The mean ALOS revised was significantly lower for the teaching service (3.64 ± 2.22 days) compared with the other 2 services (4.53 ± 1.48 and 4.73 ± 1.02, respectively [P < .001]), although the mean case-mix index was significantly higher for the teaching service compared with the other services. The mean percentage of readmissions within 30 days of discharge and 1-day stays were not significantly different between group 1 and groups 2 and 3. Hospital bed utilization in the teaching service was superior to the nonteaching hospitalist services. Further research should explore the reasons for the differences between these models of care.