Abstract 16636: Patient and Hospital-level Predictors of Early Unplanned Pediatric Cardiac Intensive Care Unit Readmissions
Background: Unplanned readmissions to pediatric cardiac intensive care units (CICU) are associated with higher mortality and greater resource utilization. Patient and hospital factors leading to CICU readmission remain unknown and differences in readmission rates across hospitals have not been reported.
Methods: We examined initial CICU encounters in the Pediatric Cardiac Critical Care Consortium (PC4) registry from 10/2013 to 3/2016. The primary outcome was unplanned CICU readmission <48 hours from CICU discharge. Patient, encounter, and hospital factors were compared. Those associated with the outcome (p<0.1) were included in a multivariable logistic regression model with treating hospital as a fixed effect. Patient factors associated with readmission (p<0.05) formed a model to calculate adjusted hospital readmission rates.
Results: Of 11,673 included CICU encounters, 62% were surgical (v. medical) and 70% were planned (v. unplanned). There were 175 (1.5%) unplanned CICU readmissions within 48 hours, most commonly for respiratory (32%) or cardiac (27%) conditions. Univariate comparisons showed no association between unplanned readmission and hospital factors including nurse hours per patient day, 24/7 in-house CICU attending and use of early warning scores on ward units. In multivariable analysis, only extracardiac anomaly (OR 1.4 [95%CI 1.1-2.0]) greater surgical complexity (OR 1.8 [95%CI 1.2-2.8]), and treating hospital (p=0.01) were associated with early unplanned readmission. After adjusting for patient-level factors, we observed four-fold variation in unplanned readmission rates across hospitals (Figure).
Conclusions: Patient factors related to surgical complexity and extracardiac anomalies are associated with early unplanned CICU readmission. Future quality improvement depends on uncovering hospital practices impacting the wide variation in readmission rates we identified across hospitals.
Author Disclosures: A.H. Smith: None. V. Anand: None. M. Banerjee: None. K.E. Bates: None. M.A. Brunetti: None. D.S. Cooper: None. J. Lehrich: None. K.P. Mistry: None. S.K. Pasquali: None. A.Y. Shin: None. S. Tabbutt: None. M.G. Gaies: Research Grant; Significant; 1K08HL116639 (PI: Gaies).
- © 2016 by American Heart Association, Inc.