Interoperability of population-based patient registries

Area of Science:

• Health Informatics
• Data Science
• Cancer Registries

Background:

• Population-based patient registries contain rich health data valuable for secondary use.
• Previous attempts at patient registry interoperability have yielded limited success, lacking a unifying solution.
• Achieving full interoperability is crucial for maximizing the utility of diverse health datasets.

Purpose of the Study:

• To demonstrate a practical solution for patient registry interoperability using federated, semantic metadata registries.
• To showcase how an existing framework can be implemented gradually and independently within registry domains.
• To illustrate the extension of the ISO/IEC 11179 standard for enhanced data interoperability.

Main Methods:

• Implementation of a federated, semantic metadata registry framework.
• Application of linked open data principles to extend the ISO/IEC 11179 standard.
• Development of automated data extraction scripts for diverse registry content models.
• Demonstration using European population-based cancer registries.

Main Results:

• Successful retrieval of aggregated core datasets from cancer registries.
• Feasibility shown for retrieving aggregated full datasets and record-level data.
• Demonstrated syntactic and semantic interoperability of data elements.
• Validation of a gradual and independent implementation strategy.

Conclusions:

• The federated, semantic metadata registry framework offers a viable solution for patient registry interoperability.
• The approach facilitates automated data search, retrieval, and linkage across registries.
• This infrastructure lays the groundwork for leveraging registry data in artificial intelligence applications.
• The method supports both syntactic and semantic data integration, enhancing secondary data usage.