Semantic integration of medication data into the EHOP Clinical Data Warehouse

Area of Science:

• Medical Informatics
• Health Data Science
• Pharmacovigilance

Background:

• Reusing medication data is vital for medical research but integrating it into clinical data warehouses (CDW) presents significant challenges.
• Semantic integration requires robust methods to handle the complexity of prescription data.

Purpose of the Study:

• To develop a reliable and scalable method for integrating prescription data into the French clinical data warehouse, EHOP.
• To address the challenges of semantic integration of medication data.

Main Methods:

• Utilized PN13/PHAST as a semantic interoperability standard for the Extract, Transform, Load (ETL) process.
• Stored prescription data as documents within the CDW.
• Employed Theriaque, a drug knowledge database (DKDB), for fine-grained annotation and semantic querying.

Main Results:

• The system demonstrated high performance in integrating prescription data.
• Precision varied from 52% to 100% depending on the specific use case.
• Recall consistently achieved 100% across all evaluated scenarios.

Conclusions:

• Semantic interoperability standards and DKDB are crucial for successful drug data integration.
• A document-oriented approach combined with fine-grained annotation enhances data integration.
• The proposed method offers a scalable solution for medication data management in CDWs.