Garrett Keim, MD (he/him/his)
Children's Hospital of Philadelphia
Philadelphia, Pennsylvania
Disclosure information not submitted.
Neethi Pinto, MD,
Assistant Professor
Children's Hospital of Philadelphia
Philadelphia, Pennsylvania
Disclosure information not submitted.
Jason Christie, MD
Professor of Medicine (Pulmonary and Critical Care) and Epidemiology
University of Pennsylvania School of Medicine, United States
Disclosure information not submitted.
Nadir Yehya, MD, MSCE
Children's Hospital of Philadelphia
Cherry Hill, NJ
Disclosure information not submitted.
Title: Readmission after Pediatric Acute Respiratory Failure: A Common Occurrence
Introduction: Acute respiratory failure (ARF) requiring invasive mechanical ventilation (IMV) occurs in ~30% of critically ill children. Most of these children survive to hospital discharge, but little is known about their outcomes after discharge. We examined the relationship between patient/hospitalization characteristics and 1-year hospital readmission among ARF survivors.
Methods: Children < 18 years old with an index hospitalization between January 2013 - December 2017 including respiratory illness diagnostic codes or procedural codes for IMV or endotracheal intubation were identified in the IBM MarketScan database. Subjects with < 30 days of insurance coverage prior to admission and neonatal diagnostic codes were excluded. The primary outcome was 1-year readmission. Continuous and categorical data were compared using non-parametric and chi-squared tests, respectively. Cox proportion hazard models were used to test association with readmission.
Results: Among 11,484 admissions, the in-hospital mortality rate was 9.9%. Readmission was 28.7%, and mortality was 2.3% at 1 year. Index length-of-stay (LOS) (10 vs 7 days, p< 0.001) and total hospital cost ($50,037 vs. $30,479, p< 0.001) were higher among readmitted children. Readmitted children were more likely to have a complex chronic condition (CCC). Children with a respiratory CCC were at highest risk of readmission, followed by non-respiratory CCC, and lowest among those without CCC (Log-rank test for trend, p< 0.0001). Comorbidities were the strongest predictors of readmission (respiratory CCC HR= 3.08, 95% CI (2.73-3.47), non-respiratory CCC HR 1.81, 95%CI (1.63-2.02)). Additionally, LOS > 7 days (HR=1.47, 95%CI (1.32-1.64)) and total hospital cost >$50,000 (HR 1.25, 95%CI (1.09-1.43)) predicted readmission.
Conclusions: Hospital readmission at 1-year is common after pediatric ARF and more frequent than readmission among the general pediatric intensive care unit population. Chronic complex conditions and complexity of index hospitalization predict readmission. This is the largest study to characterize risk factors for readmission after pediatric ARF; interventions to reduce hospital readmission will likely be of highest yield if focused on ARF patients with complex chronic conditions and longer lengths of stay.