Interrater Reliability of a Tool Measuring the Quality of Nursing Triage in the Emergency Department

Abstract

Introduction: Triage plays an essential role in the Emergency Department (ED), helping maintain a safe patient flow. Although assessing the quality of the triage process is crucial, to date, there has been no metrological testing of a tool measuring the quality of nursing triage. Objective: This study aimed to assess the interrater reliability of the Audit Triage Tool (ATT) in Quebec, Canada. Methods: This retrospective cohort study took place in a regional ED. Fifty triages were selected using a systematic random sampling technique with quotas of 10 triages grouped under 5 chief complaints: chest pain, abdominal pain, neurological problems, major blunt trauma and fever. A total of 4 auditors individually applied the 49 criteria of the ATT to 50 triages. The interrater reliability was measured with the intraclass correlation coefficient (ICC), percentage of unanimity (PU) and percentage of agreement (PA). Results: Based on the ICC, 33/49 criteria showed fair (ICC < 0.60) or poor (ICC < 0.40) interrater reliability. This number decreased to 18/49 when adding complementary statistical test: PU < 60 % or PA < 80 %. Further analyses showed that 14/23 explicit criteria reached an ICC > 0.60, comparatively to only 2/26 implicit criteria. Discussion and conclusion: Findings showed that a quarter of the ATT criteria had poor interrater reliability according to various statistical tests. Solutions to improve the reliability of the ATT, mostly regarding the implicit criteria, are needed. Finally, future methodological research on triage quality assessment should focus on a thorough validation of the ATT.