Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Utrecht, the Netherlands; London School of Hygiene and Tropical Medicine, London, UK.
J Intern Med. 2014 Jun;275(6):562-9. doi: 10.1111/joim.12211.
A solid foundation of evidence of the effects of an intervention is a prerequisite of evidence-based medicine. The best source of such evidence is considered to be randomized trials, which are able to avoid confounding. However, they may not always estimate effectiveness in clinical practice. Databases that collate anonymized electronic health records (EHRs) from different clinical centres have been widely used for many years in observational studies. Randomized point-of-care trials have been initiated recently to recruit and follow patients using the data from EHR databases. In this review, we describe how EHR databases can be used for conducting large-scale simple trials and discuss the advantages and disadvantages of their use.
循证医学的前提是有确凿的干预措施效果证据。此类证据的最佳来源被认为是随机试验,因为随机试验能够避免混杂。然而,随机试验并不总能在临床实践中估计疗效。多年来,数据库一直在广泛用于观察性研究,这些数据库将来自不同临床中心的匿名电子健康记录(EHR)进行整理。最近,已经启动了随机即时护理试验,利用 EHR 数据库中的数据招募和跟踪患者。在这篇综述中,我们描述了如何使用 EHR 数据库进行大规模简单试验,并讨论了使用这些数据库的优缺点。