Objective: To compare resource use in physician assistant (PA) services versus teaching services for 5 medical diagnostic groups in a large community teaching hospital, while controlling for attending physician.
Methods: The sample was drawn from an administrative database of patients admitted to an internal medicine service in which attending physicians' cases rotated between the PA and teaching services on a preassigned schedule. Diagnoses included in the study were stroke (diagnosis-related group [DRG] 014), pneumonia (DRG 089), acute myocardial infarction (DRG 122), congestive heart failure (DRG 127), and gastrointestinal hemorrhage (DRG 174). Cases discharged between January 1, 1994, and June 30, 1995, were evaluated. Inclusion of cases was based on the Integrated Inpatient Management Model (IIMM). Resource use was measured using direct costs expressed as relative value units (RVUs) for radiology, laboratory, and total resource use, as well as for length of stay (LOS).
Results: After adjustment for admitting physician effects and other covariates, the mixed model analyses indicated that PAs used fewer total ancillary resources for patients with pneumonia and fewer laboratory resources for patients with stroke, pneumonia, and congestive heart failure than did residents, on average. No significant differences were noted in average LOS or use of radiology resources between PA and teaching services. In no cases did PAs use significantly more resources than residents. However, significantly higher mortality among pneumonia cases was detected for PAs.
Conclusions: In the same practice setting, PAs used resources as effectively as, or more effectively than, residents.