Abstract：
BACKGROUND: The Parkland Trauma Index of Mortality (PTIM) is an integrated, machine learning 72-h mortality prediction model that automatically extracts and analyzes demographic, laboratory, and physiological data in polytrauma patients. We hypothesized that this validated model would perform equally as well at another level 1 trauma center.
METHODS: A retrospective cohort study was performed including ∼5000 adult level 1 trauma activation patients from January 2022 to September 2023. Demographics, physiologic and laboratory values were collected. First, a test set of models using PTIM clinical variables (CVs) was used as external validation, named PTIM+. Then, multiple novel mortality prediction models were developed considering all CVs designated as the Cincinnati Trauma Index of Mortality (CTIM). The statistical performance of the models was then compared.
RESULTS: PTIM CVs were found to have similar predictive performance within the PTIM + external validation model. The highest correlating CVs used in CTIM overlapped considerably with those of the PTIM, and performance was comparable between models. Specifically, for prediction of mortality within 48 h (CTIM versus PTIM): positive prediction value was 35.6% versus 32.5%, negative prediction value was 99.6% versus 99.3%, sensitivity was 81.0% versus 82.5%, specificity was 97.3% versus 93.6%, and area under the curve was 0.98 versus 0.94.
CONCLUSIONS: This external cohort study suggests that the variables initially identified via PTIM retain their predictive ability and are accessible in a different level 1 trauma center. This work shows that a trauma center may be able to operationalize an effective predictive model without undertaking a repeated time and resource intensive process of full variable selection.
METHODS: A retrospective cohort study was performed including ∼5000 adult level 1 trauma activation patients from January 2022 to September 2023. Demographics, physiologic and laboratory values were collected. First, a test set of models using PTIM clinical variables (CVs) was used as external validation, named PTIM+. Then, multiple novel mortality prediction models were developed considering all CVs designated as the Cincinnati Trauma Index of Mortality (CTIM). The statistical performance of the models was then compared.
RESULTS: PTIM CVs were found to have similar predictive performance within the PTIM + external validation model. The highest correlating CVs used in CTIM overlapped considerably with those of the PTIM, and performance was comparable between models. Specifically, for prediction of mortality within 48 h (CTIM versus PTIM): positive prediction value was 35.6% versus 32.5%, negative prediction value was 99.6% versus 99.3%, sensitivity was 81.0% versus 82.5%, specificity was 97.3% versus 93.6%, and area under the curve was 0.98 versus 0.94.
CONCLUSIONS: This external cohort study suggests that the variables initially identified via PTIM retain their predictive ability and are accessible in a different level 1 trauma center. This work shows that a trauma center may be able to operationalize an effective predictive model without undertaking a repeated time and resource intensive process of full variable selection.
摘要：
背景：Parkland创伤死亡率指数（PTIM）是一个综合的，机器学习72小时死亡率预测模型，自动提取和分析人口统计，实验室,和多发性创伤患者的生理数据。我们假设这个经过验证的模型在另一个1级创伤中心的表现同样好。
方法：对2022年1月至2023年9月的5000名成人1级创伤激活患者进行了回顾性队列研究。人口统计,收集生理和实验室值.首先,使用PTIM临床变量（CV）的一组测试模型被用作外部验证，名为PTIM+。然后，考虑到所有指定为辛辛那提创伤死亡率指数(CTIM)的CV,开发了多种新的死亡率预测模型.然后比较模型的统计性能。
结果：发现PTIMCV在PTIM+外部验证模型中具有相似的预测性能。CTIM中使用的最高相关CV与PTIM的高度重叠，模型之间的性能相当。具体来说,对于48小时内死亡率的预测（CTIM与PTIM）：阳性预测值为35.6％对32.5％，负预测值为99.6%对99.3%，灵敏度分别为81.0%和82.5%，特异性为97.3%对93.6%,曲线下面积为0.98对0.94。
结论：这项外部队列研究表明，最初通过PTIM确定的变量保留了其预测能力，并且可以在不同的1级创伤中心获得。这项工作表明，创伤中心可能能够运行有效的预测模型，而无需进行重复的时间和资源密集型的全变量选择过程。
方法：对2022年1月至2023年9月的5000名成人1级创伤激活患者进行了回顾性队列研究。人口统计,收集生理和实验室值.首先,使用PTIM临床变量（CV）的一组测试模型被用作外部验证，名为PTIM+。然后，考虑到所有指定为辛辛那提创伤死亡率指数(CTIM)的CV,开发了多种新的死亡率预测模型.然后比较模型的统计性能。
结果：发现PTIMCV在PTIM+外部验证模型中具有相似的预测性能。CTIM中使用的最高相关CV与PTIM的高度重叠，模型之间的性能相当。具体来说,对于48小时内死亡率的预测（CTIM与PTIM）：阳性预测值为35.6％对32.5％，负预测值为99.6%对99.3%，灵敏度分别为81.0%和82.5%，特异性为97.3%对93.6%,曲线下面积为0.98对0.94。
结论：这项外部队列研究表明，最初通过PTIM确定的变量保留了其预测能力，并且可以在不同的1级创伤中心获得。这项工作表明，创伤中心可能能够运行有效的预测模型，而无需进行重复的时间和资源密集型的全变量选择过程。
