A comparison of outcomes of general medical inpatient care provided by a hospitalist‐physician assistant model vs a traditional resident‐based model

Abstract

Residency reform in the form of work hour restrictions has forced academic medical centers to develop alternate models of care to provide inpatient care. One such model is the use of physician assistants (PAs) with hospitalists. However, these models of care have not been widely evaluated. To compare the outcomes of inpatient care provided by a hospitalist-PA (H-PA) model with the traditional resident based model. We conducted a retrospective cohort study of 9681 general medical (GM) hospitalizations between January 2005 and December 2006 using a hospital administrative database. We used multivariable mixed models to adjust for a wide variety of potential confounders and account for multiple patient visits to the hospital to compare the outcomes of 2171 hospitalizations to H-PA teams with those of 7510 hospitalizations to resident teams (RES). Length of stay (LOS), charges, readmission within 7, 14, and 30 days and inpatient mortality. Inpatient care provided by H-PA teams was associated with a 6.73% longer LOS (P = 0.005) but charges, risk of readmission at 7, 14, and 30 days and inpatient mortality were similar to resident-based teams. The increase in LOS was dependent on the time of admission of the patients. H-PA team-based GM inpatient care was associated with a higher LOS but similar charges, readmission rates, and inpatient mortality to traditional resident-based teams, a finding that persisted in sensitivity analyses.