Verdier C, Flory A
Laboratoire d'Ingénierie des Systèmes d'Information, Institut National des Sciences Appliquées de Lyon, France.
Methods Inf Med. 1994 Dec;33(5):496-501.
A new way is presented to build an information system addressed to problems in epidemiology. Based on our analysis of current and future requirements, a system is proposed which allows for collection, organization and distribution of data within a computer network. In this application, two broad communities of users-physicians and epidemiologists-can be identified, each with their own perspectives and goals. The different requirements of each community lead us to a client-service centered architecture which provides the functionality requirements of the two groups. The resulting physician workstation provides help for recording and querying medical information about patients and from a pharmacological database. All information is classified and coded in order to be retrieved for pharmaco-economic studies. The service center receives information from physician workstations and permits organizations that are in charge of statistical studies to work with "real" data recorded during patient encounters. This leads to a new approach in epidemiology. Studies can be carried out with a more efficient data acquisition. For modelling the information system, we use an object-oriented approach. We have observed that the object-oriented representation, particularly its concepts of generalization, aggregation and encapsulation, are very usable for our problem.
提出了一种构建针对流行病学问题的信息系统的新方法。基于我们对当前和未来需求的分析,提出了一种系统,该系统允许在计算机网络内收集、组织和分发数据。在这个应用中,可以识别出两个广泛的用户群体——医生和流行病学家,每个群体都有自己的观点和目标。每个群体的不同需求使我们采用以客户-服务为中心的架构,该架构提供了这两个群体的功能需求。由此产生的医生工作站有助于记录和查询有关患者的医疗信息以及来自药理学数据库的信息。所有信息都经过分类和编码,以便为药物经济学研究进行检索。服务中心从医生工作站接收信息,并允许负责统计研究的组织使用在患者就诊期间记录的“真实”数据。这导致了流行病学的一种新方法。可以通过更高效的数据采集来开展研究。为了对信息系统进行建模,我们使用面向对象的方法。我们观察到,面向对象的表示,特别是其泛化、聚合和封装的概念,对于我们的问题非常有用。