Modeling of ETL-Processes and Processed Information in Clinical Data Warehousing.

Background: Literature describes a big potential for reuse of clinical patient data. A clinical data warehouse (CDWH) is a means for that.

Objectives: To support management and maintenance of processes extracting, transforming and loading (ETL) data into CDWHs as well as to ease reuse of metadata between regular IT-management, CDWH and secondary data users by providing a modeling approach.

Methods: Expert survey and literature review to find requirements and existing modeling techniques. An ETL-modeling-technique was developed extending existing modeling techniques. Evaluation by exemplarily modeling existing ETL-process and a second expert survey.

Results: Nine experts participated in the first survey. Literature review yielded 15 included publications. Six existing modeling techniques were identified. A modeling technique extending 3LGM and combining it with openEHR information models was developed and evaluated. Seven experts participated in the evaluation.

Conclusion: The developed approach can help in management and maintenance of ETL-processes and could serve as interface between regular IT-management, CDWH and secondary data users.