Breckenkamp J, Spallek J, Kraywinkel K, Krieg V, Schwabe W, Greiner W, Damm O, Hense H, Razum O
Universität Bielefeld.
Gesundheitswesen. 2012 Jul;74(7):e52-60. doi: 10.1055/s-0031-1286274. Epub 2011 Oct 19.
The aim of this study was to examine whether a linkage of data sets of the Epidemiological Cancer Registry North Rhine-Westphalia (EKR-NRW) and the Medical Service of the Health Insurances (MDK WL) is feasible, which problems may occur during execution and what could be a solution strategy.
Data of the EKR-NRW were used to identify all cancer cases in the district of Munster, for which an expert opinion regarding nursing needs from the period 2004-2008 was available at the MDK WL. For this purpose factually anonymised data of the EKR-NRW and the MDK WL were linked by means of a semi-automatic probabilistic record linkage.
Data linkage yielded 18 877 cancer cases but required great technical and temporal input. The processing of management data and the use of "routine channels" (ISDN connection), which provide the necessary safety for data transfer, but are designed for a considerably smaller amount of data, accounted for this. The interface problem (converting data from text format to a hierarchical XML format) can be solved with Excel or SAS.
A record linkage with factually anonymised data from the MDK WL and the EKR-NRW is feasible. This allows, among other things, quantifying the needs for nursing care in persons with a cancer diagnosis.
本研究旨在探讨北莱茵-威斯特法伦州癌症流行病学登记处（EKR-NRW）与医疗保险医疗服务机构（MDK WL）的数据集进行关联是否可行，实施过程中可能出现哪些问题以及解决策略是什么。
利用EKR-NRW的数据识别明斯特地区的所有癌症病例，MDK WL有2004年至2008年期间关于护理需求的专家意见。为此，通过半自动概率记录关联将EKR-NRW和MDK WL实际匿名化的数据进行关联。
数据关联产生了18877例癌症病例，但需要大量的技术和时间投入。管理数据的处理以及使用“常规渠道”（ISDN连接）为此负责，该渠道为数据传输提供了必要的安全性，但设计用于处理的数据量要小得多。接口问题（将数据从文本格式转换为分层XML格式）可以用Excel或SAS解决。
将MDK WL和EKR-NRW的实际匿名化数据进行记录关联是可行的。这尤其能够量化癌症诊断患者的护理需求。
