The following is a summary of “A new Bayesian method for the estimation of emergency nurses’ thresholds and agreement in the context of telephone triage,” published in the February 2025 issue of Emergency Medicine by Vicovaro et al.
Triage is the process designed to ensure that patients receive a level and quality of care corresponding to the urgency of the conditions.
Researchers conducted a retrospective study to examine the application of a new decision-making model in telephone triage.
They applied a model to estimate the nurse’s Belonging Threshold (BT), which measured the minimum severity of an emergency scenario, prompting the nurse to activate a rescue vehicle with emergency devices. The BT served as an indicator of the nurse’s tendency to over- or under-triage. The model also provided accurate estimations of the agreement level among nurses and between nurses and reference experts, accounting for variability in BTs among nurses.
The results showed that the mean BT for 21 emergency nurses was 0.53 (SE = 0.03), close to the ideal of 0.50, indicating no significant over- or under-triage at the group level, 3 nurses exhibited under-triage tendencies, while 1 showed over-triage. Analysis of the 2019 data revealed balanced triage, with 40–60% of codes being white or green. A positive correlation (ρ = 0.58, P = 0.007) was found between BT and the percentage of white or green codes. Cohen’s κ values above 0.60 indicated moderate to high agreement with reference experts for 13 nurses. A Spearman’s correlation (ρ = 0.78, P < 0.001) confirmed a strong positive relationship between agreement with experts and colleagues. Excluding nurses with low expert agreement, the mean Cohen’s κ between nurses increased to 0.55 (SE = 0.03). Analysis of emergency scenarios identified over-triage in scenario 6 and under-triage in scenario 14, highlighting areas for focused training.
Investigators concluded the model was useful in identifying nurses who could benefit from training to improve protocol consistency and pinpointed challenging emergency scenarios for priority code assignments.
Source: frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2025.1477844/full
