Mining data, gathering variables and recombining information: the flexible architecture of epidemiological studies
Mining data, gathering variables and recombining information: the flexible architecture of epidemiological studies
复制标题
DOI：
10.1016/j.shpsc.2008.09.008
复制
发表时间：
2008-12-01
影响因子：
0.9
通讯作者：
Bauer, Susanne
中科院分区：
文献类型：
--
作者：
Bauer, Susanne
自20世纪后半叶以来，生物医学研究越来越多地使用流行病学方法来建立人口层面的经验证据。本文基于丹麦的案例研究，介绍流行病学研究中数据的实践。我提出了一个认识论的记录链接，邀请探索流行病学研究的异质组合。专注于数据收集，采样和链接，我研究如何数据组织和处理成为生产力超越其收集的背景。该案例研究着眼于1976年建立的一个当地人口数据库，以调查预防心血管疾病的可能性，30年后，该数据库被用于测试乳腺癌病因学的假设。对于两个乳腺癌调查的基础上相同的核心数据集，我遵循的基本记录链接的做法，并描述如何研究对象，如分子标记物成为相关的公共卫生，通过信息网络。流行病学关联研究作为一种工具，将不同的背景纳入统计风险估计，从而为研究以及临床测试和公共卫生政策配置选项。(C)2008爱思唯尔有限公司保留所有权利。
Since the second half of the twentieth century, biomedical research has made increasing use of epidemiological methods to establish empirical evidence on a population level. This paper is about practices with data in epidemiological research, based on a case study in Denmark. I propose an epistemology of record linkage that invites exploration of epidemiological studies as heterogeneous assemblages. Focusing on data collecting, sampling and linkage, I examine how data organisation and processing become productive beyond the context of their collection. The case study looks at how a local population database established in 1976 to investigate possibilities for the prevention of cardiovascular disease is used thirty years later to test hypotheses on the aetiology of breast cancer. For two breast cancer investigations based on the same core data set, I follow the underlying record linkage practice and describe how research objects such as molecular markers become relevant with respect to public health through information networking. Epidemiological association studies function as tools that performatively enrol different contexts into statistical risk estimation, thereby configuring options for research as well as for clinical testing and public health policy. (C) 2008 Elsevier Ltd. All rights reserved.