Strategy & Planning Department, Ambulance Victoria, Australia.
Injury. 2012 May;43(5):573-81. doi: 10.1016/j.injury.2010.10.003. Epub 2010 Nov 11.
Pre-hospital trauma triage criteria are used to expedite the transport of severely injured patients to major trauma services. The current Victorian adult pre-hospital trauma triage criteria consist of physiological, anatomical and mechanistic elements. The purpose of this study was to evaluate the performance of the current triage criteria and, if necessary propose refined criteria to improve the under and over-triage rates.
The study was conducted in Melbourne, Victoria, which has a fully integrated State Trauma System. Trauma data was sourced from the pre-hospital Victorian Ambulance Clinical Information System and the Victorian State Trauma Registry. Confirmed major trauma was defined at hospital discharge as one or more of death, ISS>15, ICU ventilation or urgent surgery. Data was matched through probabilistic linkage. The triage criteria were evaluated using multivariate logistic regression and classification tree modelling. Diagnostic statistics, including sensitivity and specificity were calculated to assess triage performance.
Over 12-months there were 1166 'confirmed major trauma' patients and 44,166 'non-major trauma' patients. Evaluation showed the current triage criteria needed refinement, and multiple revised pre-hospital trauma triage models were constructed. Based on the best overall combination of diagnostic statistics, a revised model was chosen with a sensitivity of 97.8% (vs. 95.3% in the current model), a specificity of 82.7% (vs. 62.7%) and an accuracy of 83.0% (vs. 63.4%). The over-triage rate was 17.3% (vs. 37.3%) and the under-triage rate was 2.2% (vs. 4.7%).
Evaluation showed that the specificity and sensitivity of the current trauma triage criteria could be improved. The implementation of a revised triage model should identify more confirmed major trauma patients. Likewise, over-triage of non-major trauma patients to major trauma services would be significantly reduced. The refined criteria should also decrease discretionary decision-making by paramedics in the field.
院前创伤分诊标准用于加快严重受伤患者向主要创伤服务的转运。目前维多利亚州成人院前创伤分诊标准包括生理、解剖和机械因素。本研究的目的是评估当前分诊标准的性能，并在必要时提出改进分诊率的细化标准。
该研究在墨尔本进行，墨尔本拥有一个完全整合的州创伤系统。创伤数据来自院前维多利亚救护车临床信息系统和维多利亚州创伤登记处。在医院出院时确定为严重创伤的标准为死亡、ISS>15、ICU 通气或紧急手术。通过概率链接匹配数据。使用多变量逻辑回归和分类树模型评估分诊标准。诊断统计数据，包括灵敏度和特异性，用于评估分诊性能。
在 12 个月内，有 1166 名“确诊严重创伤”患者和 44166 名“非严重创伤”患者。评估表明，目前的分诊标准需要细化，并构建了多个修订后的院前创伤分诊模型。基于诊断统计数据的最佳整体组合，选择了一个修订后的模型，其灵敏度为 97.8%（当前模型为 95.3%），特异性为 82.7%（当前模型为 62.7%），准确性为 83.0%（当前模型为 63.4%）。过度分诊率为 17.3%（当前模型为 37.3%），漏诊率为 2.2%（当前模型为 4.7%）。
评估表明，当前创伤分诊标准的特异性和灵敏度可以提高。实施修订后的分诊模型应能识别更多确诊的严重创伤患者。同样，将非严重创伤患者过度分诊到主要创伤服务的情况将显著减少。细化的标准还应减少现场护理人员的自由裁量决策。
