ESI – Every Score Inaccurate

June 6, 2018

Written by Clay Smith

Spoon Feed
Emergency severity index (ESI) scores were highly variable and inaccurate in this multicenter, international study.  The most important thing to take away is to not be falsely reassured by low-acuity ESI scores.  View them with healthy skepticism.

Why does this matter?
ESI scores are routinely used in EDs across the US and abroad.  Clinical decisions and room placement decisions are often made in view of these scores based on anticipated resource utilization.  High scores are low acuity, low resource; low scores are high acuity, resource-intensive.  But are the scores accurate?

Flip a coin
This was a multicenter cross-sectional study of 87 nurses from the US, UAE, and Brazil comparing assigned vs actual ESI scores on 25 AHRQ standardized triage case scenarios.  Overall accuracy was 59%; interrater reliability was modest.  Twice as many patients were undertriaged rather than overtriaged (27.6% vs 13.2%).  Scores were more accurate in medium acuity patients at 76%; accuracy in high acuity scenarios was only 44%, 54% in low acuity patients.  Performance was even worse in pediatric patient scenarios.  Avoid the cognitive error of false reassurance with a low-acuity ESI score.  Don’t be lulled to sleep.  Also, it looks like we need a better triage scoring system – maybe A.I. could help.

Source
Accuracy and Reliability of Emergency Department Triage Using the Emergency Severity Index: An International Multicenter Assessment.  Ann Emerg Med. 2018 May;71(5):581-587.e3. doi: 10.1016/j.annemergmed.2017.09.036. Epub 2017 Nov 24.

Peer reviewed by Thomas Davis