POINT OF CARE DELIVERY OF RESEARCH EVIDENCE

这是对最初在中审查的拨款申请的修订 1999年7月。研究表明，平均需要近20年的时间 让研究证据进入临床实践。依靠被动 传播信息以保持卫生专业人员的知识更新是 在一个全球环境中注定要失败，在这个环境中，大约有200万名医疗人员 研究论文每年发表一次。这个修订项目的目的是 是：(1)制定政策，自动选择可信和 实质性证据，以及(2)将患者数据与临床证据相匹配，以及 (3)直接向临床决策点提供高质量的证据 制作。这项拟议的项目是由来自 密苏里大学、哥伦比亚大学和杜克大学。系统化 评审以评估资格标准、结果变量和总体 将对已公布证据的质量进行审查。自动取证 筛选系统将开发并选择随机临床试验 证据将从文献中提炼出来。抽象将包含 匹配的资格标准，对临床医生的建议，以及 建议向患者发出通知。一个决策支持系统，自动地 将创建与患者数据相匹配的文献。基于 抽象的文献证据，基于Web的智能代理将搜索 病人数据库中的护理失误。提醒建议和证据 他们身后的人将被送到适当的临床医生那里。提醒标准将 将通过随机临床试验进行开发和评估。此外，还有 将是杜克大学的分包合同，杜克大学将专注于开发 使用结构的临床决策支持系统，电子存储 证据逻辑模块(由本项目中的其他研究人员创建)以 向临床医生提供循证建议。这个决策支持系统将 在一项随机对照试验中进行评估，该试验评估 关于遵守护理标准的制度。

This is a revision of a grant application originally reviewed in July of 1999. Studies document that it takes an average of nearly two decades for research evidence to reach clinical practice. Relying on the passive diffusion of information to keep health professionals' knowledge up to date is doomed to failure in a global environment in which about two million medical research articles are published annually. The purpose of this revised project is to: (1) develop policies for the automated selection of credible and substantial evidence, and (2) match patient data with clinical evidence and (3) directly deliver high-quality evidence to the point of clinical decision making. The proposed project is a collaboration of researchers from the University of Missouri, Columbia University, and Duke University. Systematic reviews to evaluate eligibility criteria, outcomes variables, and overall quality of published evidence will be conducted. An automated evidence filtering system will be developed and selected randomized clinical trial evidence will be abstracted from the literature. The abstractions will contain eligibility criteria for matching, a recommendation to the clinician, and a suggested notice to the patient. A decision-support system to automatically match the literature with patient data will be created. Based on the abstracted literature evidence, an intelligent web-based agent will search the patient databases for lapses in care. Reminder recommendations and evidence behind them will be sent to the appropriate clinician. Reminder standards will be developed and evaluated with a randomized clinical trial. In addition there will be a subcontract to Duke University which will focus on developing a clinical decision support system that uses structures, electronically stored evidence logic modules (created by the other investigators in this project) to deliver evidence-based advice to clinicians. This decision support system will be evaluated in a randomized controlled trial that assesses the impact of the system on adherence with care standards.