Pediatric ED Saves: Analyzing the ED Screen of Direct Admissions

Research output: Contribution to journalArticlepeer-review

Abstract

Introduction: Direct admissions (DAs) are a routine hospital entry portal with few guidelines to assess patient safety during this process. This study assessed the effectiveness of an institutional screen for patients presenting as DA. It investigated patient variables that may predict appropriateness for DA and those at high risk for deterioration. Methods: The study includes patients who received the institutional screen between June 1, 2019, and May 31, 2020. We placed charts into three groups: group 1 (stable), group 2 (unstable), and group 3 (stable then transferred to pediatric intensive care unit within 6 hours). We assessed effectiveness by calculating sensitivity, specificity, and predictive values. We used comparative analysis between groups to identify patients safe for DA and those at high risk for deterioration, Results: The screen was 80% sensitive and 100% specific, predicting 97.7% of stable patients. Of the 652 charts reviewed, 384 met the inclusion criteria. Group 1 (31.60, 26.45%, 5.23%) had lower respiratory rate, respiratory diagnosis, and oxygen requirement compared to group 2 (45.00, 78.13%, 15.63%) and group 3 (44.50, 75.00%, 50.00%). For SpO2, group 1 (98.70) was higher than group 2 (96.03). For the Pediatric Early Warning Score, group 2 (1.72) was higher than group 1 (0.31) and group 3 (0.63). Conclusions: The institutional screen is an effective tool to identify patients presenting as DA needing immediate emergency department intervention and/or pediatric intensive care unit care.

Original languageEnglish (US)
Pages (from-to)E678
JournalPediatric Quality and Safety
Volume8
Issue number4
DOIs
StatePublished - Jul 7 2023

ASJC Scopus subject areas

  • Pediatrics, Perinatology, and Child Health

Fingerprint

Dive into the research topics of 'Pediatric ED Saves: Analyzing the ED Screen of Direct Admissions'. Together they form a unique fingerprint.

Cite this