Errors in nurse-led triage: An observational study

Background: Nurses play a crucial role in correctly prioritizing patients entering emergency departments. However, little is known of the accuracy of nurse-led triage systems. Objectives: (1) To determine the frequency of nurse-led triage errors within the Manchester Triage System; (2) to explore patient, work environment and individual nurse factors associated with triage errors; and (3) to explore associations between triage errors and patient outcomes (i.e., length of emergency department stay, hospitalization, and 7- and 30-day mortality). Setting: This study was conducted in one emergency department in Northern Italy. Participants: A random sample of 5% (n = 1,929) of all eligible patients accessing the emergency department over an 18-month period. Methods: For this retrospective observational study, electronic health record data on triage errors (i.e., incorrect presentational flowchart, specific discriminator and/or priority level) and triage nurses were combined with routine data on patient characteristics, outcomes and the work environment. To explore relationships between these variables, we performed univariate and multivariate logistic regression analyses. Results: We observed triage errors in 16.3% of patients (n = 314). These were significantly associated with patients' emergency department and hospital stays. Analyses revealed that when > one patient was triaged every 15 min (OR: 2.112;95%CI: 1.331-3.354), older patients (OR: 1.009; 95%CI: 1.003-1.015) with > than two chronic conditions (OR: 1.506; 95%CI: 1.091-2.081) and orange or red priority codes (OR: 1.314; 95%CI: 1.046-1.651,) whose triage nurse had previous experience with another triage system (OR: 3.189; 95%CI: 2.455-4.14) had higher odds of triage errors. Conclusion: We provided primary evidence on triage errors. Confirming our findings on the prevalence, nature and consequences of such errors will require further prospective multicenter studies. Considering patient factors (e.g., age, polychronicity) as additional discriminators could make the nurse-led triage process using the Manchester Triage System more accurate. Investigating the roles of triage nurses' training and background and the emergency department work environment on their mental models regarding the triage process will require qualitative research. (C) 2020 Elsevier Ltd. All rights reserved.