OBJECTIVE: To provide reference for safe and rational use of drugs in the clinic and further drug safety risk evaluation. METHODS: 8 795 cases were collected from ADR monitoring network in our region from 2007 to 2010, and determined quantitatively by using proportional reporting ratio (PPR), Reporting odds ratio(ROR)and Bayesian confidence propagation neural network (BCPNN). The suspicious signal were compared and analyzed, and the quantitative signal detection method based on autonomous reporting system database and the scientific ADR signal detection way were explored. RESULTS & CONCLUSION: 3 common methods were all adopted to detect 869 types of Drug-ADR and 66 types of Drugs-ADR quantitatively. Part of signal could be detected. The results of PRR and ROR method were similar, and BCPNN method was different from them significantly. The signal strengths of 3 methods were not identical.%目的:为指导临床安全、合理用药和进一步评价药品安全风险提供参考.方法:通过收集我区2007-2010年药品不良反应(ADR)监测网络ADR报告8 795份,运用频数法中的比例报告(PRR)法、报告比值比(ROR)法和贝叶斯判别可信区间递进神经网络模型(BCPNN)法进行定量检测,对生成的可疑信号进行比较分析,探索以自发呈报系统数据库为基础的定量信号检测方法的运用和科学的ADR信号检测途径.结果与结论:通过对869种Drug-ADR组合和66种Drugs-ADR组合进行3种常用方法信号定量检测,发现对部分已知信号能够检出,PRR法、ROR法结果相近,BCPNN法差别较大,3种方法信号强度各不相同.
展开▼
