电子病历数据库在临床研究中的应用及偏倚控制

Database of Electronic Health Records: Application in Clinical Research and Bias Control

电子病历数据库(electronic health records,EHRs)的出现为大样本临床队列研究提供了低成本、高效率的愿景和机遇,但利用EHRs开展临床研究仍然存在误区和挑战。本文回归流行病学的本质,通过实例探讨EHRs的特征,EHRs中可获得的信息与临床研究中暴露变量、结局变量及协变量之间的对应关系,如何利用EHRs进行患病率及发病率估计、开展疗效和政策评价以及在临床研究中对选择偏倚、混杂偏倚尤其是未测量混杂偏倚的控制等,旨在促进EHRs与临床研究的整合,提高临床研究的效率及质量。

The emergence of electronic health records(E H R s)provides a good opportunity for c lin ic a l studies to he carried out on large samples w ith high efficiency and at low costs.H ow ever,m isunderstanding and inappropriate application of EHRs in c lin ic a l research are common.This paper hrings us hack to the essentials of epidem iology.We use a num her of examples to discuss:the characteristics of E H R s,the relationship of data domains of EHR s w ith th e ir corresponding variahles(exposure,outcomes,and covariates)in epidem iological research,and how to use EHRs to estimate prevalence and incidence and to evaluate the effectiveness of treatment and p o lic y.We also focus on the selection hias and confounding controls,especially unmeasured confounding controls.H o p e fu lly,this paper would contrihute to the integration of EHRs and c lin ic a l research and to the im provement of efficiency and q u ality of c lin ic a l research.