IJPDS International Journal of Population Data Science

Trusted Research Environments (TREs) are like safe havens for health data, ensuring that each one of us will be able to contribute anonymously to valuable research insights generated from day to day interactions with health care systems. TREs protect our information by offering both legal foundations and technology tools. But there's a catch: not all data is in one place, different data flows into different TREs. For example, if you move from Wales to England, there will most likely be an unknown gap in your electronic health records which is the data you contribute to research. This can be more critical in cases where people develop or suffer from a rare condition and are interested in getting timely insight into diagnosis or treatment for their conditions.  

All TREs have high security in five main areas known as “five-safes”; however, not all TREs follow the same approach when it comes to operationalising these five rules. Imagine you're in a library, and each book is locked away in its own separate room. If you need information from multiple books, you'd have to go into each room individually. This can be time-consuming and frustrating. The same problem can occur in the world of research where collaboration is a key. But what happens when data gets trapped in separate silos, making it hard for researchers to work together?

Professor Simon Thompson and his team, are focusing on technical approaches in running multi-TRE analyses and as part of their ongoing work they have looked in detail how five-safe rules were operationalised in 15 major UK TREs that had contributed to multi-TRE research. Their work showed that although there are major commonalities in how TREs follow five-safes, fine differences and lack of standards are causing data segregation across UK-TREs.

Lead researcher, Fatemeh Torabi, said “In a world where knowledge and research are key to solving pressing issues, a standard approach is like the bridge that brings data owners and researchers together. By breaking down data silos, it fosters collaboration, enhances data security, optimises resources, and promotes transparency. This means a brighter future where researchers can work together seamlessly to tackle the world's challenges.”

The situation is like having those locked rooms in the library. A standard model is like the librarian who holds the keys to all the rooms in the library.

Prof Thompson said “We stand at the forefront of a transformative era in data-driven research. This article unveils the remarkable journey we've embarked upon to ensure that the existing governance models, the cornerstone of TREs, evolves into a robust and standardised foundation.”

Prof Lyons added “It is fantastic to see work published. It shows that the team at Swansea University continue to lead on innovative approaches to support the global health research community through combining data from many centres whilst still protecting privacy.”

Emma Squires also commented, “We believe that every individual’s health data holds a potential key to unlocking groundbreaking insights. TREs provide the secure space where this potential can flourish, safeguarding privacy and facilitating research. However, the challenge lies in achieving a suitable level of governance for a wide range of data.”

Fatemeh Torabi, Emma Squires, Professor Ronan A. Lyons and Professor Simon Thompson, Population Data Science, Swansea University

Click here to read the full open access article

The initiative is part of the UK Secure Research Platforms (UK SeRP) one of the largest TRE providers within the UK. The work was funded and supported by Dementias Platform UK and Alzheimer’s Disease Data Initiatives.

Torabi, F., Orton, C., Squires, E., Heys , S., Hier, R., Lyons, R. A. and Thompson, S. (2023) “The Common governance model: a way to avoid data segregation between existing trusted research environment”, International Journal of Population Data Science, 8(4). doi: 10.23889/ijpds.v8i4.2164.