Cosgrave Nicole, Saleh Sooad, Ong Woei Shan, Frydenlund Juliane, Williams David J, Cahir Caitriona
Department of Medicine, RCSI University of Medicine and Health Sciences, Dublin, Ireland.
Department of Geriatric and Stroke Medicine, Beaumont Hospital, Dublin, Ireland.
Eur J Clin Pharmacol. 2025 Jan;81(1):93-110. doi: 10.1007/s00228-024-03774-7. Epub 2024 Nov 18.
Adverse drug reactions (ADRs) are common and result in significant morbidity, mortality, and associated hospital costs. Models predicting ADRs in older adults were previously found to lack reliability and validity. This systematic review and meta-analysis aim to provide an updated, comprehensive quality assessment and analysis of ADR-risk prediction tools in older adults.
Standard databases and citations were searched (2012 to 2023) and studies which developed and/or validated an ADR prediction model for use in older adults were included. Four studies from a previous systematic review were also included. The TRIPOD (transparent reporting of a multivariable prediction model for individual prognosis or diagnosis) checklist was used to evaluate each study. Random effects models were used to derive a pooled discrimination estimate (area-under-the-receiver-operation curve; AUROC) for model development and validation studies.
Eight studies, describing 6 ADR-risk prediction models met the inclusion criteria). In the meta-analysis, the pooled AUROC was 0.75 (95% CI 0.57, 0.87; I-squared = 96.88%) for model development studies and 0.73 (95% CI 0.52, 0.87; Isquared = 90.19%) for externally validated studies. Studies had poor adherence (range 34-50%; median 46.5%; IQR 12%) to TRIPOD guidelines.
The studies identified through this systematic review exhibit poor adherence to TRIPOD guidelines which may question the investigational rigor and the usability of the models. This underscores the urgent need to develop a validated, robust, and reliable tool worthy of implementation and testing in a real-world setting to gauge its impact and usability effectively.
药物不良反应（ADR）很常见，会导致严重的发病率、死亡率以及相关的医院费用。先前发现，预测老年人药物不良反应的模型缺乏可靠性和有效性。本系统评价和荟萃分析旨在对老年人药物不良反应风险预测工具进行更新、全面的质量评估和分析。
检索了标准数据库及文献（2012年至2023年），纳入了开发和/或验证用于老年人的药物不良反应预测模型的研究。还纳入了先前一项系统评价中的四项研究。使用TRIPOD（个体预后或诊断多变量预测模型的透明报告）清单来评估每项研究。随机效应模型用于得出模型开发和验证研究的合并鉴别估计值（受试者操作特征曲线下面积；AUROC）。
八项研究描述了6种药物不良反应风险预测模型，符合纳入标准。在荟萃分析中，模型开发研究的合并AUROC为0.75（95%CI 0.57，0.87；I²=96.88%），外部验证研究的合并AUROC为0.73（95%CI 0.52，0.87；I²=90.19%）。研究对TRIPOD指南的遵循情况较差（范围34 - 50%；中位数46.5%；四分位间距12%）。
通过本系统评价确定的研究对TRIPOD指南的遵循情况较差，这可能会质疑研究的严谨性以及模型的可用性。这凸显了迫切需要开发一种经过验证、强大且可靠的工具，值得在现实环境中实施和测试，以有效评估其影响和可用性。
