Patient Outcomes in a Pediatric Hospital Medicine Service Staffed With Physicians and Advanced Practice Providers

Abstract

OBJECTIVES: Hospitals are employing more nurse practitioners and physician assistants on inpatient pediatric units. With this study, we compared patient outcomes in high-volume inpatient diagnoses on pediatric hospital medicine services staffed by attending physician hospitalists and residents (hospitalist and resident service [HRS]) with 1 staffed by attending physician hospitalists and advanced practice providers (HAPPS). METHODS: A historical cohort study was implemented by using administrative data for patients admitted to HRS and HAPPS from 2007 to 2011 with asthma, bronchiolitis, cellulitis, and pneumonia with severity levels 1 and 2 for all-patient refined diagnosis-related groups. Length of stay, readmission, ICU transfer, and hospital charges were compared. RESULTS: After controlling for clinical, demographic, and socioeconomic differences, the average probability of discharge was 10% greater each day (event ratio [ER] = 1.1 [1.06–1.14]) on HAPPS compared with HRS. By diagnosis, this trend persisted with asthma (ER = 1.07 [1.02–1.12]), cellulitis (ER = 1.2 [1.1–1.3]), and pneumonia (ER = 1.17 [1.08–1.28]) but not for bronchiolitis (ER = 0.99 [0.92–1.06]). Both 3- and 30-day readmissions were higher for HRS discharges with bronchiolitis (odds ratio = 5.9 [1.3–28.6] and 2.0 [1.3–3.3], respectively) but not for the other diagnoses. Hospital charges were 13% higher for patients on HRS than HAPPS. ICU transfers did not differ statistically. CONCLUSIONS: Within the limitations of the design, HAPPS performed at least as well as HRS with respect to length of stay, readmissions, ICU transfers, and charges for 4 of the most common inpatient diagnoses with severity levels 1 to 2. Indicated in these results is that in this configuration, advanced practice providers on pediatric hospitalist services represent a viable model for other institutions to consider and test.