ENABLING CLINICAL RESEARCH IN EMERGENCY AND ACUTE CARE MEDICINE THROUGH AUTOMATED DATA EXTRACTION

迄今为止，在急诊医学中进行护理质量评估研究的可能性与可持续性问题发生了冲突。访问艾德的大量患者和经常困扰这些部门的人员短缺使得无法实现临时数据收集。填补临床研究需求和可靠数据可用性之间的差距的唯一方法是直接从ED电子健康记录（EHR）中提取此类数据，避免专门的数据收集。实现这一目标将使分布式临床研究成为可能，而现在这种研究过于局限于学术中心，并允许利用临床信息来解决众多的研究问题。然而，从电子病历中获得一致的数据是一项复杂的任务。虽然电子健康记录中登记的小部分数据是结构化的（例如实验室测试结果和生命参数），但有关患者状况的大部分有用信息都包含在自由文本中（例如体征和症状的存在、疑似和确诊诊断、既往史等）。此外，由于对艾德患者进行主动随访是不可行的，因此还需要依赖现有数据源，以便在最适当的时间间隔内针对感兴趣的研究问题测量患者的结局。该提案有三个主要目标：1）开发新的技术解决方案，从不同电子患者文件中包含的结构化和非结构化数据中提取可靠的临床信息; 2）FAIRify（即使数据可查找、可解释、可互操作和可重复使用）为临床医生、研究人员、卫生政策制定者和公民建立的数据库，同时尊重欧洲和国家立法; 3）在两个相关用例中试点利用已建立的数据库：i）评估艾德患者住院的倾向，以及ii）开发供公民和政策制定者使用的仪表板，以提高ED的护理质量。

To date, the possibility of doing research on quality-of-care assessment in emergency medicine has clashed against sustainability problems. The vast number of patients visiting an ED and the staff shortage that often afflicts these departments make ad hoc data collections unattainable. The only way to fill the gap between the need for clinical research and the availability of robust data is to directly extract such data from the EDs electronic health records (EHRs), avoiding dedicated data collection. Achieving this goal would enable distributed clinical research, which is now too much restricted to academic centres, and allow to leverage of clinical information to address a multitude of research questions. Nonetheless, obtaining consistent data from EHRs is a complex task. While a small part of the data registered in EHRs is structured (such as lab test results and vital parameters), most of the useful information on patients' conditions is variably contained in free text (e.g. presence of signs and symptoms, suspected and confirmed diagnosis, anamnesis, etc.). Moreover, as a proactive follow-up of ED patients is unfeasible, relying on the existing data sources is also necessary to measure the outcome of the patients at the most appropriate time interval for the research question of interest. This proposal has three main aims: 1) to develop new technical solutions to extract reliable clinical information from structured and unstructured data contained in different electronic patient files; 2) to FAIRify (i.e. making data Findable, Accessible, Interoperable, and Re-usable) the established databases for clinicians, researchers, health policymakers and citizens while respecting the European and national legislations; 3) to pilot the exploitation of the established databases in two relevant use cases: i) assessment of ED propensity to hospitalise a patient, and ii) development of a dashboard to be used by citizens and policymakers to improve the quality of care in ED.