606

Abstract

Introduction: Many trauma systems have added additional services such as acute care surgery. These additional services have been undertaken for improved financial stability and to offer new surgical opportunities for trauma physicians. This has increased the work burden for existing trauma systems. With new restricted resident work hours, there has been potential for inpatient care to suffer due to these combined factors. Advanced Practice Professionals (APP); represented by physician assistants and nurse practitioners, were added to our trauma system and patient logistics were located to one floor. This was done to help offset the increased work load that accompanied the new additional services. Patient readmissions are used to evaluate the quality of inpatient care and are currently being used by the Center for Medicare and Medicaid Services. APP are increasingly being used to staff hospital provider positions and they are considered a safe and cost effective alternative. Patient readmissions offer the ability to judge the effects of changes in staffing and patient logistics. It is unknown what effect if any the new changes would have on the quality of trauma inpatient care in the face of these new responsibilities. Methods: Trauma readmissions were evaluated from 2001-2012. Elective surgeries and non trauma readmissions were excluded. Length of stay (LOS), age, Injury Severity Score (ISS) and disposition "not home" were compared. A full complement of APP's was obtained in 2008 and clustering in 2010. An overdispersed Poisson regression with offset by number of yearly admissions was used to evaluate the effect of APP complement and patient clustering of the trauma service on readmission days with statistical significance < 0.05. Results: Over an 11 year period, the trauma population increased from 1,392 admits to 1,848. During this time, average age increased from 36 to 44 and ISS decreased from 13.4 to 11, while average mortality decreased from 5.7% to 5.3%, LOS dropped from 7.2 days to 6.2, and disposition "not home" declined from 36.4% to 30.8% . Clustering showed a 14% reduction in readmissions (p=0.119, 95% CI 4% increase to 28% decrease) and APP staffing had a 37% reduction in readmissions (p<0.001, 95% CI 25 to 47% decrease). When clustering, which had a p = 0.119, was removed from the model, APP staffing showed a 40% reduction in readmissions (95% CI: 30-49% reduction). Conclusions: Addition of APP to the trauma staff resulted in increased quality of service which was delivered during an era of increasing clinical responsibilities for the trauma division and increasingly restricted resident work hours. This is revealed through decreased readmissions and LOS. Similar improvements may be expected on other services with utilization of APP into staffing models. Although centralization of trauma patients did not show significance, it did not result in a decline in quality of care and should be evaluated further.