基于HIS数据的药源性心律失常自动监测模块规则建立与验证  

Establishment and validation of automatic monitoring module for drug-induced arrhythmia based on data of HIS

作　　者：郭海丽 郭代红[1] 高奥 张博[1] 李伯妍 王嘉熙 石廷永 GUO Hai-li;GUO Dai-hong;GAO Ao;ZHANG Bo;LI Bo-yan;WANG Jia-xi;SHI Ting-yong(Department of Pharmacy,Medical Supplies Center of Chinese PLA General Hospital,Beijing 100853,China;College of Pharmacy,Chongqing Medical University,Chongqing 400016,China;Kanglianda Software Corporation,Beijing 100028,China)

机构地区：[1]解放军总医院医疗保障中心药剂科,北京100853 [2]重庆医科大学药学院,重庆400016 [3]北京康联达软件有限公司,北京100028

出　　处：《中国药物应用与监测》2022年第3期176-181,共6页Chinese Journal of Drug Application and Monitoring

基　　金：2017年军事医学创新工程重点项目(17CXZ010);中国研究型医院学会专项课题-临床重点药品的使用监测和评价研究专项(Y2021FH-YWPJ01)。

摘　　要：目的:在临床ADE主动监测与智能评估警示系统-Ⅱ(ADE-ASAS-Ⅱ)中建立药源性心律失常自动监测模块,探究药物与心律失常的关系,为临床合理用药提供参考。方法:建立心律失常模块的关键词词集,采用文本分类技术识别提取上述词集,优化报警规则;扩大监测样本量,对阳性患者的性别、年龄、相关药物等数据进行分析以验证模块。结果:最终确定关键词28个,模块最佳设置条件的阳性预测值(PPV)为10.82%,召回率(R)为99.18%。利用该模块监测患者33007例,其中阳性622例,发生率1.88%,PPV7.88%。65岁及以上患者占比50.32%;涉及药物共64类、133种,以喹诺酮类抗菌药物、钙通道阻滞剂、抗真菌药物为主;药源性心律失常类型多样,主要是期外收缩、心动过速、QTc间期延长。结论:本研究构建的药源性心律失常自动监测模块,可以高效、精准、快速获取真实世界大样本用药人群中的目标病例。监测结果显示住院人群发生药源性心律失常属于常见,相关患者的年龄分布、药物类别、类型分类等与有关研究结果基本一致。Objective: To establish a drug-induced arrhythmia module in adverse drug event active surveillance and assessment system Ⅱ(ADE-ASAS-Ⅱ) and provide references for clinical rational drug use. Methods: The key words set of the drug-induced arrhythmia module were established after identifyings and extracting by text classification technology. The alert rule setting of the module was optimized. The sample size of monitored inpatients was expanded to analyze the gender, age and related drugs of positive patients in order to validate the module. Results: A total of 28 key words were identified. In the optimal alert rule setting of the module, the positive predictive value(PPV) and the recall rate(R) were 10.82% and 99.18% respectively. The module was used to monitor 33 007 patients, of which there were 622 positive patients with an incidence rate of 1.88% and a PPV of 7.88%.Among them, 50.32% were over 65 years old. 64 categories of 133 drugs were involved, and the top 3 drugs were quinolones,calcium channel blockers and antifungal drugs. There were various kinds of drug-induced arrhythmia, in which extrasystole,tachycardia and QTc interval prolongation were the most important types. Conclusion: In this study, the automatic monitoring module of drug-induced arrhythmia could efficiently, accurately and quickly obtain target cases in a large sample of patients in the real world. Monitoring results showed that the incidence of drug-induced arrhythmia was common in hospitalized patients. The age distribution, drug category and type classification of related patients were basically consistent with relevant research results.

关 键 词：药源性心律失常 自动监测 文本分类技术 药品不良反应 真实世界研究 

分 类 号：R969.3[医药卫生—药理学]