论文部分内容阅读
目的探索一种通过挖掘临床微生物数据早期预警医院感染暴发的方法,使感染控制工作更有目的有重点地开展,提高工作效率。方法采用移动百分位数法,以某种病原菌一周累计检出数的P95作为预警阈值(若P95<3则以3作为阈值)。若累计检出数超过阈值则产生预警信号,SAS程序自动绘制时序图,再人工对时序图进行甄别,然后对相关病历进行回顾性调查,判断是否属于暴发。结果基于某医院2013年微生物数据,通过本方法共产生126个预警信号,经过图形甄别筛选出8次疑似暴发信号,回顾性调查确认其中5次属于暴发。结论本方法具有简单、高效、成本低等特点,是一种预警院感暴发的有效工具。
Objective To explore a method of early warning of outbreaks of nosocomial infections by mining clinical microbiological data so as to make infection control work more purposeful and focused and improve work efficiency. Methods The mobile percentile method was used to detect the P95 of some pathogens in a week as the early warning threshold (if P95 <3, the threshold was 3). If the cumulative number of detection exceeds the threshold alarm signals are generated, SAS program automatically draw the timing diagram, and then manually identify the timing diagram, and then the relevant medical records were retrospectively investigated to determine whether the outbreak. Results Based on the microbiological data of a hospital in 2013, a total of 126 early-warning signals were generated by this method. Eight suspected outbreak signals were screened out by graphic screening, and five of them were confirmed as outbreaks by retrospective survey. Conclusion The method is simple, efficient and low cost, which is an effective tool for early warning of outbreaks.