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

Area of Science:

• Health Informatics
• Data Management
• Information Systems

Background:

• Clinical patient data holds significant potential for reuse.
• Clinical Data Warehouses (CDWHs) are key infrastructure for data aggregation.
• Effective management of data processes is crucial for data utility.

Purpose of the Study:

• To develop a modeling approach for managing Extract, Transform, Load (ETL) processes.
• To facilitate metadata reuse between IT management, CDWHs, and secondary data users.
• To improve the maintenance and management of data integration pipelines.

Main Methods:

• Conducted an expert survey and literature review to identify requirements and existing techniques.
• Developed a novel ETL modeling technique by extending 3LGM^2 and integrating openEHR information models.
• Evaluated the developed technique through exemplary modeling and a second expert survey.

Main Results:

• Identified six existing modeling techniques from 15 reviewed publications.
• Developed and validated a hybrid modeling technique combining 3LGM^2 and openEHR.
• Seven experts participated in the evaluation, confirming the approach's utility.

Conclusions:

• The developed modeling approach aids in managing and maintaining ETL processes.
• It serves as a valuable interface between regular IT management, CDWHs, and secondary data users.
• Enhances the accessibility and usability of clinical data for secondary purposes.