Are 5-level triage systems improved by using a symptom based approach?—a Danish cohort study

Background: Five-level triage systems are being utilized in Danish emergency departments with and without the use of presenting symptoms. The aim of this study was to validate and compare two 5-level triage systems used in Danish emergency departments: “Danish Emergency Process Triage” (DEPT) based...

Full description

Bibliographic Details
Main Authors: Brabrand, M. (Author), Dahlin, J. (Author), Fløjstrup, M. (Author), Kongensgaard, F.T (Author), Lassen, A. (Author)
Format: Article
Language:English
Published: BioMed Central Ltd 2022
Subjects:
ED
ICU
Online Access:View Fulltext in Publisher
LEADER 03148nam a2200289Ia 4500
001 10.1186-s13049-022-01016-2
008 220510s2022 CNT 000 0 und d
020 |a 17577241 (ISSN) 
245 1 0 |a Are 5-level triage systems improved by using a symptom based approach?—a Danish cohort study 
260 0 |b BioMed Central Ltd  |c 2022 
856 |z View Fulltext in Publisher  |u https://doi.org/10.1186/s13049-022-01016-2 
520 3 |a Background: Five-level triage systems are being utilized in Danish emergency departments with and without the use of presenting symptoms. The aim of this study was to validate and compare two 5-level triage systems used in Danish emergency departments: “Danish Emergency Process Triage” (DEPT) based on a combination of vital signs and presenting symptoms and a locally adapted version of DEPT (VITAL-TRIAGE) using vital signs only. Methods: This was a retrospective cohort using data from five Danish emergency departments. All patients attending an emergency department during the period of 1 April 2012 until 31 December 2015 were included. Validity of the two triage systems was assessed by comparing urgency categories determined by each triage system with critical outcomes: admission to Intensive care unit (ICU) within 24 h, 2-day mortality, diagnosis of critical illness, surgery within 48 h, discharge within 4 h and length of hospital stay. Results: We included 632,196 ED contacts. Sensitivity for 24-h ICU admission was 0.79 (95% confidence interval 0.78–0.80) for DEPT and 0.44 (0.41–0.47) for VITAL-TRIAGE. The sensitivity for 2-day mortality was 0.69 (0.67–0.70) for DEPT and 0.37 (0.34–0.41) for VITAL-TRIAGE. The sensitivity to detect diagnoses of critical illness was 0.48 (0.47–0.50) for DEPT and 0.09 (0.08–0.10) for VITAL-TRIAGE. The sensitivity for predicting surgery within 48 h was 0.30 (0.30–0.31) in DEPT and 0.04 (0.04–0.04) in VITAL-TRIAGE. Length of stay was longer in VITAL-TRIAGE than DEPT. The sensitivity of DEPT to predict patients discharged within 4 h was 0.91 (0.91–0.92) while VITAL-TRIAGE was higher at 0.99 (0.99–0.99). The odds ratio for 24-h ICU admission and 2-day mortality was increased in high-urgency categories of both triage systems compared to low-urgency categories. Conclusions: High urgency categories in both triage systems are correlated with adverse outcomes. The inclusion of presenting symptoms in a modern 5-level triage system led to significantly higher sensitivity measures for the ability to predict outcomes related to patient urgency. DEPT achieves equal prognostic performance as other widespread 5-level triage systems. © 2022, The Author(s). 
650 0 4 |a Chief complaint 
650 0 4 |a ED 
650 0 4 |a ICU 
650 0 4 |a Mortality 
650 0 4 |a Surgery 
650 0 4 |a Symptoms 
650 0 4 |a Triage 
650 0 4 |a Triage systems 
650 0 4 |a Vital signs 
700 1 |a Brabrand, M.  |e author 
700 1 |a Dahlin, J.  |e author 
700 1 |a Fløjstrup, M.  |e author 
700 1 |a Kongensgaard, F.T.  |e author 
700 1 |a Lassen, A.  |e author 
773 |t Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine