Egoda Kapuralalage Thilini Nisansala, Chan Ho Fai, Dulleck Uwe, Hughes James A, Torgler Benno, Whyte Stephen
School of Economics and Finance, Queensland University of Technology, 2 George St, Brisbane, QLD 4000, Australia.
School of Economics and Finance, Queensland University of Technology, 2 George St, Brisbane, QLD 4000, Australia; ARC Training Centre for Behavioural Insights for Technology Adoption, Queensland University of Technology, Brisbane, QLD 4000, Australia; Centre for Behavioural Economics, Society & Technology, Queensland University of Technology, Brisbane, QLD 4000, Australia.
Am J Emerg Med. 2025 Jun;92:60-67. doi: 10.1016/j.ajem.2025.02.043. Epub 2025 Feb 27.
In emergency medicine, triage decisions are critical for ensuring patient safety and optimizing resource usage. Such decisions involve a complex interplay of rational and analytical thinking, combined with an intuitive and humanistic approach. However, the influence of cognitive biases on triage decisions remains poorly understood.
Between February 20 and June 27, 2023, we conducted an online scenario-based survey with 78 triage-competent Registered Nurses in the emergency department at Princess Alexandra Hospital in Australia. Co-designed with nurse educators and nursing academics, the survey included domains covering demographic information, tailored diagnostic tests to capture the presence of cognitive biases and risk-taking behavior, and six vignettes requiring triage using the Australasian Triage Scale. Logistic mixed-effects and multivariate Poisson regression models were performed to identify the influence of cognitive biases and risk-taking behavior on triage decision accuracy.
We identified negative framing bias (82.5 %), anchoring bias (82 %), and availability bias (62.8 %) as the most prevalent cognitive biases among triage nurses. After adjusting for age, sex, education, and triage work experience, no statistically significant associations were observed between cognitive biases or risk-taking behavior and triage accuracy. This indicates that cognitive biases may have a limited influence on well-trained nurses. However, age, sex, and triage work experience were found to be significant predictors of inaccurate triaged decisions.
Our study provides preliminary evidence that cognitive biases and risk-taking behavior are not associated with triage accuracy among well-experienced and trained emergency triage nurses. Further research is required to fully understand the impact of cognitive biases on emergency triage decisions.
在急诊医学中，分诊决策对于确保患者安全和优化资源利用至关重要。此类决策涉及理性与分析思维的复杂相互作用，同时还需结合直观和人文主义方法。然而，认知偏差对分诊决策的影响仍知之甚少。
2023年2月20日至6月27日期间，我们对澳大利亚亚历山德拉公主医院急诊科78名具备分诊能力的注册护士进行了一项基于情景的在线调查。该调查与护士教育工作者和护理学者共同设计，涵盖人口统计学信息、用于捕捉认知偏差和冒险行为存在情况的定制诊断测试，以及六个需要使用澳大利亚分诊量表进行分诊的病例 vignettes。采用逻辑混合效应模型和多变量泊松回归模型来确定认知偏差和冒险行为对分诊决策准确性的影响。
我们发现消极框架偏差（82.5%）、锚定偏差（82%）和可得性偏差（62.8%）是分诊护士中最普遍的认知偏差。在对年龄、性别、教育程度和分诊工作经验进行调整后，未观察到认知偏差或冒险行为与分诊准确性之间存在统计学上的显著关联。这表明认知偏差对训练有素的护士影响可能有限。然而，年龄、性别和分诊工作经验被发现是分诊决策不准确的重要预测因素。
我们的研究提供了初步证据，表明认知偏差和冒险行为与经验丰富且训练有素的急诊分诊护士的分诊准确性无关。需要进一步研究以充分了解认知偏差对急诊分诊决策的影响。
