Provider Based Outcomes on an Allogeneic Transplant Service

Introduction:

Allogeneic stem cell transplantation (ASCT) is a potentially curative, although highly morbid procedure for some patients with hematologic malignancies. Care of patients undergoing ASCT is highly specialized, and often complicated. Traditionally, the majority of transplants occur at tertiary academic centers where housestaff (medical residents) provide daily care. However, with new limitations on resident work-hours, alternative models of team based care replacing residents with Advanced Practice Providers (nurse practitioners/physician assistants) have been utilized. In this study we compared outcomes on a housestaff (HS) based allogeneic transplant service to outcomes on an Advanced Practice Provider (APP) based service.

Methods:

A retrospective chart review of ASCT patients admitted to the Hospital of the University of Pennsylvania was performed. Prior to July 2012, all ASCT patients were admitted to a HS service (staffed by residents, a fellow, and an attending). Starting July 2012, all new ASCT patients were admitted to a newly developed APP service (staffed by 2 APPs, an attending, and overnight moonlighters). We evaluated 86 patients admitted to the HS service from May 2011 to May 2012 and 81 patients admitted to the APP service from Oct 2012 to Oct 2013. Patients from June 2012 to September 2012 were excluded due to transition and cross-over during the initial months of the new service. A subset of 27 overflow ASCT patients admitted to the housestaff service from Oct 2012 to Oct 2013 was studied as part of a secondary analysis. For our primary analysis, we compared patient, provider, and cost outcomes pre and post July 2012 on the HS and APP services. Patient outcomes included 100 day relapse free and overall survival rates, length of stay (LOS), 14 & 30 day re-admission rates, in-hospital death rate, ICU transfers, and infectious complications such as pneumonia (PNA), urinary tract infections, bacteremia, and C. Diff colitis. Descriptive and comparative analyses were performed using a t-test for continuous variables, chi-squared and Fisher’s exact test for categorical variables, and a Wilcoxon rank-sum test for ordinal variables.

Results:

Baseline characteristics among patients on the APP and HS services revealed no statistically significant differences in mean age, gender, indication for transplant, donor type (allo-sib vs MUD vs cord graft), or conditioning regimen (myeloablative vs reduced intensity). More patients were transplanted with active disease on the APP service (41% vs 27%, p=0.06), but the difference was not statistically significant. ECOG performance status and a hematopoietic cell transplantation-specific comorbidity index score were also not significantly different between the groups. The rate of PNA was significantly less on the APP service (15% vs 28%, p=0.04) while other infectious complications were similar in both groups. There were several important findings that did not reach statistical significance, including less in-hospital death on the APP service (4% vs 10%, p=0.09), a decrease in ICU transfers on the APP service (8.6% vs 18.6%, p=0.06) and a three day absolute difference in the mean LOS favoring a shorter hospitalization on the APP service (p=0.12) with no difference in 14 or 30 day re-admission rates. 100 day survival rates and relapse free survival rates were similar in both groups with slightly less relapse in the APP group (p=0.08). Lastly, hospital charges and total costs of the hospitalization were not significantly different although there was an increase in testing on the HS service in terms of number of radiological films (5 vs 8, p=0.05) and number of blood cultures ordered (4 vs 7, p=0.03).

Conclusions:

With increasing demands on the physician work force, alternative models of care are increasingly employed. In our study, the development of a specialized inpatient APP allogeneic transplant service suggests that there may be a decreased length of stay, fewer ICU transfers, and fewer in-hospital deaths. There were potential cost-saving benefits due to decreased utilization of laboratory and radiological testing. In conclusion, APPs are a safe and feasible alternative care approach to the traditional HS model for patients undergoing ASCT. Additional studies need to be performed to ensure that both the quality and cost of patient care are maximized as new care models are developed.