药物不良反应信号检测.pdf

上海市 ADR 数据库中神经系统药物不良反应的信号检测 发表时间：2009-9-18 9:37:29 来源：创新医学网推荐 刘媛 1 程能能 1? 杜文民 2 王永铭 1 (1 复旦大学药学院 药理学教研室 上海 200032; 2 上海市药品不良反应监测中 心 上海 200040) LIU Yuan1, CHENG Neng-neng 1△ , DU Wen-ming2 , WANG Yong-ming1 (1 Department of Pharmacology, School of Pharmacy, Fudan University , Shanghai , 200032 ; 2 Shanghai Center f or Adverse Drug Reaction Monitoring, Shanghai , 200040 , China) 【摘要】背景 在药品不良反应(ADR)监测数据库中存在 大量已经明确因果关系的神经系统 ADR 报告, 目前的 ADR 信号定量检测方法不区分数据库中的已知或未知报 告。 目的考察已知 ADR 报告对 ADR 信号检测效率的影 响,并对上海市 ADR 数据库进行神经系统 ADR 信号检 测。 方法　建立 ADR 信息库以自动判别 ADR 数据库中 的已知ADR 报告，采用 PRR 算法对上海市 2004-2007 年 ADR 自发呈报数据库进行信号检测，并计算可解释度 以评价检出的信号。结果 1. 建立了包含 1261 种药 物,4238 种药物-事件组合 (drug-event combination, DEC) 的 神经系统 ADR 信息库；2.筛除已知 ADR 报告能提高神经 系统 ADR 信号检出率 24.64% ；3.检测计算共产生神经系 统 ADR 信号 98 条，包括卡培他滨-咽下困难、吗替麦考 酚酯-视觉异常、头孢他啶-精神病等值得关注的信号。 结 论　筛除已知 ADR 报告能提高神经系统 ADR 信号的检 测效率。 Abstract Background The signal detection for adverse drug reaction (ADR) is usually based on an ADR spontaneous reporting system (SRS) database containing a considerable proportion of known ADR reports. And currently no algorithm of signal detection can generate difference in results between the known and new ADR reports. Purpose To examine the efficiency of ADR signal detection based on SRS database while known ADR reports were excluded partially or totally and to identify potential ADR signals of nervous system. Methods An ADR database of nervous system was established, which was used to remark known ADR reports from SRS automatically. ADR signals of nervous system were generated for possible hazards from proportional reporting ratios (PRR). A new measure representing the association between a drug and an AE was created and applied as reference to assist signal evaluation. Results 1. An ADR database of nervous system containing 1261 drugs and 4138 DECs （Drug-effect combinations ） was established. 2. Excluding the reports carrying only known ADRs from Shanghai SRS dat