Reliability and validity of two four-level emergency triage systems

Area of Science:

• Emergency Medicine
• Healthcare Systems Analysis
• Clinical Triage

Background:

• Effective patient triage is crucial for optimizing emergency department (ED) resource allocation and patient flow.
• Existing triage systems require ongoing evaluation for reliability and predictive validity.

Purpose of the Study:

• To compare the reliability and predictive validity of the four-level Triage Emergency Method (TEM) model against the established I-4L triage system.
• To assess the accuracy of both models in predicting patient admission to the hospital.

Main Methods:

• An observational study involving 10 nurses randomly assigned to use either the I-4L or TEM model for 189 paper-based patient scenarios.
• Weighted kappa statistics were employed to evaluate interrater and intrarater reliability.
• Predictive validity was assessed by the accuracy of triage ratings in predicting patient admission.

Main Results:

• Both I-4L and TEM showed good interrater reliability (kappa=0.73 and kappa=0.79, respectively) and intrarater reliability (kappa=0.82 and kappa=0.78, respectively).
• The accuracy in predicting patient admission was similar for both models (I-4L: 79%, TEM: 77%).
• The proportion of admitted patients per triage level was comparable between the two systems.

Conclusions:

• Both the I-4L and the new TEM model exhibit good reliability and accuracy for triage acuity and admission prediction.
• The TEM model, despite nurses' limited experience, performed comparably to I-4L and offers an added benefit in predicting ED resource utilization.