Implementing a quality assurance and maintenance framework for standardized clinical terminology data in a...

Area of Science:

• Health Informatics
• Data Quality Management
• Clinical Terminology Standardization

Background:

• High-quality mapping of local terms to standard clinical terminologies is essential for maximizing the value of real-world data (RWD).
• Existing methods may lack scalability and robust quality assurance for multi-institutional RWD.
• A systematic approach is needed to ensure reliable data for research and operations.

Purpose of the Study:

• To apply and evaluate a scalable quality assurance (QA) and maintenance framework for local-to-standard vocabulary mappings.
• To assess the framework's effectiveness across multiple healthcare institutions.
• To enhance the integrity and usability of multi-institutional RWD.

Main Methods:

• Operationalized a framework using six data quality dimensions: validity, uniqueness, completeness, consistency, timeliness, and accuracy.
• Utilized a centralized terminology master (Kakao Healthcare's Universal Data Model) for data from three hospitals.
• Implemented validity checks, uniqueness audits, completeness analyses, and consistency harmonization, with regular updates and expert review for accuracy.

Main Results:

• Resolved issues across all six quality dimensions, correcting format errors and removing redundancies.
• Successfully bridged mapping gaps and minimized semantic discrepancies through inter-institutional harmonization.
• Achieved high-fidelity standardization of terminology across three participating hospitals.

Conclusions:

• The developed scalable framework effectively ensures terminology mapping quality and data integrity.
• This framework provides a robust foundation for reliable research and operational use of multi-institutional RWD.
• It supports enhanced data standardization and value extraction from diverse clinical data sources.