Interobserver agreement in emergency department triage

Area of Science:

• Emergency Medicine
• Nursing
• Healthcare Quality

Background:

• Triage is a critical process in emergency departments (EDs) for prioritizing patient care.
• Previous studies indicated poor agreement between in-person and telephonic triage assessments.
• Understanding factors influencing triage reliability is essential for improving patient outcomes.

Purpose of the Study:

• To quantify interobserver reliability in emergency department (ED) triage.
• To assess the impact of vital signs on triage decision-making.
• To contextualize prior findings on agreement disparities between in-person and telephonic triage.

Main Methods:

• Prospective observational study with a randomized crossover design.
• Two independent, sequential in-person ED intake interviews conducted by experienced triage nurses.
• Nurses assigned one of five triage designations before and after receiving vital signs.

Main Results:

• Poor agreement was observed between nurses' initial triage designations (53% agreement, kappa=0.30).
• Knowledge of vital signs did not enhance agreement (49% agreement, kappa=0.25).
• The study included 201 patients and 19 experienced triage nurses.

Conclusions:

• Significant interobserver variability exists among certified triage nurses using a standardized scale.
• The study indicates that the communication method (in-person vs. telephone) is not the primary driver of triage unreliability.
• Findings suggest systemic issues in triage scale design or application contribute to poor agreement.