文献速递重症综合管理与镇静镇痛

地理、时间与COVID-19大流行数据集偏移下的死亡预测性能:全球开源疾病严重程度评分模型的外部验证

Mortality Prediction Performance Under Geographical, Temporal, and COVID-19 Pandemic Dataset Shift: External Validation of the Global Open-Source Severity of Illness Score Model.
PubMed 文献 · Crit Care Explor · 2025-06-04
一句话结论GOSSIS-1在日本ICU外部验证中保持强区分度(AUC 0.908),但校准随疾病类别和大流行时期显著变化。

该研究在日本重症患者数据库(2015—2022年)中外部验证GOSSIS-1模型,并与APACHE-II、APACHE-III及本地化JROD模型比较死亡预测性能。GOSSIS-1区分度AUC为0.908,与APACHE-III(0.908)和JROD(0.910)相当;校准方面SMR为0.89(95%CI 0.88-0.90),优于APACHE-II(SMR 0.39,95%CI 0.39-0.40)和APACHE-III(SMR 0.46,95%CI 0.46-0.47),接近JROD(SMR 0.97,95%CI 0.96-0.99)。但神经系统疾病和创伤的校准欠佳,且2020年COVID-19大流行期间性能全面下降,GOSSIS较APACHE-III尤为明显。

要点

  • GOSSIS-1在日本ICU队列中区分度AUC 0.908,与APACHE-III(0.908)及JROD(0.910)相当。
  • GOSSIS-1校准SMR 0.89(95%CI 0.88-0.90),优于APACHE-II(0.39,95%CI 0.39-0.40)和APACHE-III(0.46,95%CI 0.46-0.47),接近JROD(0.97,95%CI 0.96-0.99)。
  • GOSSIS-1在神经系统疾病和创伤类别校准欠佳。
  • 2015—2019年模型表现时间稳定,2020年COVID-19大流行期间性能下降,GOSSIS较APACHE-III更明显。
  • 复杂模型如GOSSIS-1的校准可因大流行等卫生系统压力而发生改变。

适用人群与场景

  • 日本ICU患者,数据来自日本重症患者数据库(care Japanese Intensive care PAtient Database),时间跨度为2015年至2022年。

研究设计与关键结果

  • 研究类型为外部验证,验证队列为日本重症患者数据库2015—2022年数据,比较GOSSIS-1与APACHE-II、APACHE-III及JROD模型的死亡预测性能。
  • GOSSIS-1区分度AUC 0.908,APACHE-III AUC 0.908,JROD AUC 0.910。
  • GOSSIS-1校准SMR 0.89(95%CI 0.88-0.90),优于APACHE-II(SMR 0.39,95%CI 0.39-0.40)和APACHE-III(SMR 0.46,95%CI 0.46-0.47),接近JROD(SMR 0.97,95%CI 0.96-0.99)。
  • GOSSIS-1在神经系统疾病和创伤类别校准欠佳。
  • 2015—2019年模型表现时间稳定,2020年COVID-19大流行期间性能下降,各疾病类别普遍受影响,GOSSIS较APACHE-III尤为明显。

证据类型、等级与局限

  • 证据类型:外部验证研究。
  • 局限:原文未报告局限性。

对床旁实践的改变

  • GOSSIS-1在日本ICU人群中保持强区分度,但神经系统疾病和创伤患者校准欠佳,用于这些类别风险预测时需谨慎。
  • COVID-19大流行期间GOSSIS-1校准下降,提示在卫生系统压力时期使用该模型需注意性能变化。

提炼说明

  • 本条依据论文摘要整理;摘要未涉及「诊断标准与关键阈值」「处置流程与药物剂量」「红旗征与不能漏的情况」「指南或共识出处」,故未列出这些栏目;全文请点「阅读原文」。
英文原文(摘要)· 点击展开

BACKGROUND: Risk-prediction models are widely used for quality of care evaluations, resource management, and patient stratification in research. While established models have long been used for risk prediction, healthcare has evolved significantly, and the optimal model must be selected for evaluation in line with contemporary healthcare settings and regional considerations.

OBJECTIVES: To evaluate the geographic and temporal generalizability of the models for mortality prediction in ICUs through external validation in Japan.

DERIVATION COHORT: Not applicable.

VALIDATION COHORT: The care Japanese Intensive care PAtient Database from 2015 to 2022.

PREDICTION MODEL: The Global Open-Source Severity of Illness Score (GOSSIS-1), a modern risk model utilizing machine learning approaches, was compared with conventional models-the Acute Physiology and Chronic Health Evaluation (APACHE-II and APACHE-III)-and a locally calibrated model, the Japan Risk of Death (JROD).

RESULTS: Despite the demographic and clinical differences of the validation cohort, GOSSIS-1 maintained strong discrimination, achieving an area under the curve of 0.908, comparable to APACHE-III (0.908) and JROD (0.910). It also exhibited superior calibration, achieving a standardized mortality ratio (SMR) of 0.89 (95% CI, 0.88-0.90), significantly outperforming APACHE-II (SMR, 0.39; 95% CI, 0.39-0.40) and APACHE-III (SMR, 0.46; 95% CI, 0.46-0.47), and demonstrating a performance close to that of JROD (SMR, 0.97; 95% CI, 0.96-0.99). However, performance varied significantly across disease categories, with suboptimal calibration for neurologic conditions and trauma. While the model showed temporal stability from 2015 to 2019, performance deteriorated during the COVID-19 pandemic, broadly reducing performance across disease categories in 2020. This trend was particularly pronounced in GOSSIS compared with APACHE-III.

CONCLUSIONS: GOSSIS-1 demonstrates robust discrimination despite substantial geographic dataset shift but shows important calibration variations across disease categories. In particular, in a complex model like GOSSIS-1, stresses on the health system, such as a pandemic, can manifest changes in model calibration.

来源:Crit Care Explor,查看原文。版权归原作者所有。

参考来源

  1. Tohyama T, McCoy LG, Ishii E, et al. Mortality Prediction Performance Under Geographical, Temporal, and COVID-19 Pandemic Dataset Shift: External Validation of the Global Open-Source Severity of Illness Score Model. Crit Care Explor. 2025. PMID: 40466201. doi:10.1097/CCE.0000000000001275

供执业医师抢救与诊疗参考,不替代临床判断;剂量请结合体重、肝肾功能和本院药品规格复核,以指南原文为准。