Comparison of Outcome Measures for a Traditional Pediatric Faculty Service and Nonfaculty Hospitalist Services in a Community Teaching Hospital

OBJECTIVE: Several studies have found decreased cost and length of stay for patients who are cared for by pediatric hospitalists compared with traditional faculty models. The objective of this study was to compare cost and length of stay between a faculty group and 2 separate hospitalist groups in a community teaching hospital. This study differs from previous ones in that both the traditional faculty and hospitalist models were in place simultaneously, and the traditional faculty group was employed by the hospital, whereas the hospitalist groups were in private practice. METHODS: A total of 1009 pediatric patients with any of the 11 most frequent diagnosis-related groups were analyzed according to the admitting physician group. Total direct costs and length of stay were computed for 3 separate groups (faculty group, hospitalist group 1, and hospitalist group 2). Linear regression models were used to compare total direct costs and length of stay among the groups. Each model accounted for age, severity index, and payer source. RESULTS: Age, severity index, and physician group were predictive in determining total direct costs and length of stay. There was no significant difference in patient age among the groups, but the faculty group had significantly increased severity indices compared with hospitalist groups 1 and 2 (1.6 +/- 0.7 vs 1.3 +/- 0.6 vs 1.4 +/- 0.6, mean +/- SD). The faculty group had statistically significantly lower total direct costs compared with hospitalist groups 1 and 2 (1781 dollars +/- 1449 dollars vs 1954 dollars +/- 1212 dollars vs 1964 dollars +/- 1495 dollars, mean +/- SD). The faculty group had shorter average length of stay compared with hospitalist groups 1 and 2 (2.6 +/- 2.0 vs 3.1 +/- 2.6 vs 2.9 +/- 2.3, mean +/- SD). The readmission rates among the groups were similar. CONCLUSIONS: Traditional faculty models can be as efficient in terms of total direct costs and length of stay as evolving hospitalist models. This study's results may be unique because the traditional faculty model was composed of general pediatricians instead of a blend of generalists and subspecialists. In addition, the traditional faculty physicians concentrated almost entirely on the care of inpatients while engaged in hospital care.