Privacy-Preserving Linkage of Distributed Pseudonymised Datasets in a Virtual European Rare Disease Platform

Area of Science:

• Health Informatics
• Bioinformatics
• Data Science

Background:

• Secondary data use is crucial for rare disease (RD) research due to data scarcity.
• Distributed and pseudonymised data formats pose challenges for effective data linkage.
• The European Joint Programme on Rare Diseases (EJP RD) aims to build an infrastructure supporting secondary data use.

Purpose of the Study:

• To present a conceptual framework for integrating a Privacy-Preserving Record Linkage (PPRL) service with the EJP RD Virtual Platform (VP).
• To facilitate the secure and privacy-compliant linking of distributed pseudonymised rare disease datasets.
• To enhance data discoverability and enable accurate patient cohort analysis within the EJP RD infrastructure.

Main Methods:

• Developed a concept for linking a PPRL service to the EJP RD VP.
• Implemented a Federated Discovery service (FDP) within the PPRL service for resource discovery (Level 1).
• Configured the PPRL service to act as an individual or catalog endpoint for data discoverability (Level 2).

Main Results:

• The proposed solution enables patient counting across PPRL-supported resources.
• It accurately counts unique patients, avoiding duplication across datasets.
• The system facilitates counting of patients registered in multiple resources, crucial for cohort analysis.

Conclusions:

• The PPRL service concept offers a viable solution for linking distributed, pseudonymised rare disease data within the EJP RD VP.
• This integration enhances the secondary use of rare disease data by addressing privacy and data distribution challenges.
• The system is currently being prepared for deployment, promising to advance rare disease research infrastructure.