A case study in distributed team science in research using electronic health records

Area of Science:

• Health Informatics
• Pharmacovigilance
• Data Science

Background:

• Regulatory barriers hinder cross-jurisdictional health data sharing.
• Pseudonymised routine health data access is crucial for research.
• Developing methods to overcome these barriers is essential for public health studies.

Purpose of the Study:

• To evaluate the risk of new non-vitamin K Target Specific Oral Anticoagulants (TSOACs).
• To establish a collaborative research framework between the Farr Institute in Wales and Scotland.
• To address challenges in accessing and analyzing pseudonymised health data across different platforms and jurisdictions.

Main Methods:

• A federated data and multiple analysts approach was employed.
• Simultaneous access for analysts in Wales (SAIL Databank) and Scotland (eDRIS) was arranged.
• A harmonised algorithm was used to generate comparable study cohorts from distinct data sources.

Main Results:

• A study cohort of 6,829 patients was successfully generated across two countries.
• Source datasets and data types were rigorously reviewed for consistency.
• The federated approach ensured harmonised output despite using separate data repositories.

Conclusions:

• The adopted approach is a simple, efficient, and cost-effective solution for multi-jurisdictional health data analysis.
• This method ensures analytical consistency and coherence across different governance systems.
• It serves as a foundation for a distributed team science infrastructure using Electronic Health Records (EHRs) across the UK and globally.