The indirect costs of graduate medical education.

Abstract

This analysis, aimed at measuring the indirect costs associated with hospital teaching programs, found that university teaching hospitals were 33 per cent more costly than nonteaching hospitals with respect to direct hospital costs (excluding overhead) after adjustment for differences in case mix using diagnosis-related groups (DRGs). This study found major teaching hospitals (not administered by a university) to be 18 per cent and minor teaching hospitals 9 per cent more costly than nonteaching hospitals. These cost differentials were due primarily to the greater intensity of services provided in teaching settings rather than to the cost per unit of service. Inclusion of the full costs of physician services reduced the cost differences among teaching categories. Although teaching hospitals, especially university teaching hospitals, are demonstrably more expensive than nonteaching hospitals for the same types of cases, researchers and policy makers must consider the physician-substitution effect. This is particularly important in the light of the current controversy over the integration of physician costs for inpatient services into Medicare's DRG-based prospective-payment system.