Evaluation of Early Warning Scores in Predicting ICU Admission in Emergency Department Patients

Authors

  • Mohsin Hayat Assistant Professor Pediatric Department Medical Teaching Institute Lady Reading Hospital Peshawar
  • Mohammad Irshad Assistant Professor Pediatric Department Medical Teaching Institute Lady Reading Hospital Peshawar
  • Ashfaq Ahmad Associate Prof department of Pediatric Saidu Medical College, Swat Pakistan
  • Rashida Sadiq Women Medical Officer Services Hospital Peshawar
  • Muhammad Aftab Consultant Pediatrician Blackpool Victoria Teaching Hospital NHS England Whinny Hayes Road Blackpool FY38NR
  • Hunain Khan House Officer Pakistan Institute Of Medical Sciences, Islamabad, Pakistan

DOI:

https://doi.org/10.52783/rev-alap.75

Keywords:

Early Warning Scores, ICU, Emergency, Prediction

Abstract

Background:

The emergency department relies on Early Warning Scores (EWS) as essential assessment methods for tracking patients who face clinical deterioration. The implementation of scoring systems leads to improved emergency department patient prioritization alongside enhanced care delivery that potentially decreases patient mortality numbers. Most studies should evaluate the prediction accuracy of these tools for ICU admissions since they have extensively spread throughout healthcare facilities.

Objectives:

to assess and compare MEWS, NEWS and REMS systems in their ability to forecast ICU hospital admissions for emergency department patients.

Study design: A Retrospective Observational Study.

Place and duration of study. Department of pediatric MTI,LRH Peshawar  January  from July 2023 to December 2023

Methods: this study conducted  at Department of pediatric MTI,LRH Peshawar  from January  from July 2023 to December 2023. The study enrolled adult patients who came to the ED then received admission at ICU or general wards. Studiers performed MEWS, NEWS and REMS calculations at the time of patient arrival. The study compared predictive performance by plottingROC curves along with AUCs through DeLong’s statistical testing.

Results:

50 enrolled patients we included 120 were in the ICU and 180 received care in general wards. The average patient age in ICU admission was 65.2 years with standard deviation of 12.6 years which differed significantly from 57.4 years with standard deviation of 13.8 years in non-ICU admission (p = 0.001). The success rate of NEWS for risk prediction reached 0.85 while REMS achieved 0.80 and MEWS came in at 0.74. The use of NEWS scoring with at least 5 points led to a significant relationship with ICU hospital admission.

Conclusion:

Among ED patients NEWS delivered better predictive value for ICU admission than both MEWS and REMS systems. The implementation of NEWS into ED triage procedures would improve fast critical illness detection which enables both timely interventions and resource optimization. Further study involving larger metacentric studies must be conducted to support large-scale implementation.

 

Downloads

Published

2024-12-31

Issue

Section

Research Articles