基于多源信息构建药品不良反应基准数据库  

Methodology for Constructing Benchmark Database of Adverse Drug Reactions Based on Multi-Source Information

作　　者：聂晓璐[1,2,3] 孙凤 阎爱侠[4] 彭晓霞 詹思延[2] NIE Xiaolu;SUN Feng;YAN Aixia;PENG Xiaoxia;ZHAN Siyan(Center for Clinical Epidemiology&Evidencebased Medicine,Beijing Children’s Hospital,Capital Medical University,National Center for Children’s Health,Beijing 100045,China;Department of Epidemiology and Biostatistics,School of Public Health,Peking University,Beijing 100091,China;Hainan Institute of Real World Data,Qionghai Hainan 571437,China;College of Life Science and Technology,Beijing University of Chemical Technology,Beijing 100029,China)

机构地区：[1]国家儿童医学中心,首都医科大学附属北京儿童医院临床流行病与循证医学中心,100045 [2]北京大学公共卫生学院流行病与卫生统计学系,北京100091 [3]海南省真实世界数据研究院,海南琼海571437 [4]北京化工大学生命科学与技术学院,北京100029

出　　处：《中国药物警戒》2025年第1期10-15,36,共7页Chinese Journal of Pharmacovigilance

基　　金：国家自然科学基金资助项目(82204149);北京市医院管理中心“青苗”计划专项经费资助(QML20231204);北京儿童医院科研苗圃计划(3-1-014-01-04);海南博鳌乐城国际医疗旅游先行区真实世界研究专项计划项目(HNLC 2022RWS015)。

摘　　要：目的总结现有主要药品不良反应(ADR)基准数据库构建信息来源,并以药源性血小板减少(DITP)为例综合多源数据构建ADR基准数据库,以期为今后开展计算模拟研究和指导临床安全用药提供参考。方法比较介绍现有各主要ADR基准数据库的信息来源优势与局限性;利用多源信息汇总构建DITP基准数据库,使用Kappa值评价各来源数据信息的一致性;利用药品解剖学、治疗学及化学分类法(ATC)编码分析DITP风险药物在解剖学分类中的分布情况与差异。结果利用美国食品药品监督管理局(FDA)推荐的多源信息来源方法构建了包含1765种药物的DITP基准数据库(DITPst)。在DITPst数据库中,按照ATC编码解剖学分类划分,最常发生DITP的药物为抗肿瘤及免疫调节类药物,77.17%(196/254)可引起DITP。结论利用多源信息构建ADR基准数据库可为药物研发开展计算模拟研究及药品上市后安全与合理用药提供数据参考。Objective To summarize the primary data sources used in constructing benchmark databases for adverse drug reactions(ADR)and to demonstrate a comprehensive,multi-source approach to building an ADR benchmark database using drug-induced thrombocytopenia(DITP)as an example so as to provide a reference for future computational modeling studies and to guide safer clinical drug use.Methods The advantages and limitations of the data sources used in existing ADR benchmark databases were compared and analyzed.A benchmark database for DITP was constructed by integrating data from multiple sources,and the consistency of these data sources was evaluated using the kappa statistic.The distribution and variability of DITPrisk drugs were analyzed based on the Anatomic Therapeutic Chemical(ATC)classification system.Results Using the FDA-recommended multi-source integration approach,a DITP benchmark database(DITPst)comprising 1765 drugs was constructed.Analysis of the anatomical classification of drugs within the DITPst database revealed that antineoplastic and immunomodulating agents were the most frequently associated with DITP,with 77.17%(196/254)of these drugs identified as causing these ADR.Conclusion Constructing ADR benchmark databases using multi-source information provides a valuable data reference for computational modeling in drug development as well as for ensuring post-marketing drug safety and promoting rational drug use.

关 键 词：药品不良反应 药源性血小板减少 基准数据库 多源信息 计算模拟研究 抗肿瘤药物 免疫调节类药物 安全合理用药 

分 类 号：R994.1[医药卫生—毒理学] R978[医药卫生—药学]