51±¬ÁÏÍø

Connecting the UK and Singapore for sustainable federated health data research

The UK and Singapore have strengthened their collaboration on health innovation through initiatives like the recently launched UK-Singapore and a Memorandum of Understanding signed between the National Research Federation, Singapore (NRF) and Health Data Research UK (51±¬ÁÏÍø). As part of this, the 51±¬ÁÏÍø Federated Analytics programme is working with several partners in Singapore to explore how health data research can also be conducted across borders.

Building and maintaining public trust is central to this effort. The programme is transforming how Trusted Research Environments (TREs) can facilitate international health data research while protecting privacy – approaches that minimise data movement and strengthen governance are more important than ever. Using open-source tools such as and the , teams across three nodes (A*STAR, Singapore TRUST, and the University of Nottingham) have begun testing how federated analytics can work in practice.

Researchers can submit a query from one location, the analysis runs locally within each TRE, and only aggregated, non-disclosive results are returned, with data remaining under the control and governance of each organisation. This enables researchers to analyse data across institutions and countries while ensuring that sensitive data never leaves its secure environment. This approach helps demonstrate how federated analytics can support secure, international health data research between partners in the UK and Singapore and generate insights more quickly to help improve health outcomes.

Beyond technology: it’s not plug and play

While open-source tools like Bunny and Five Safes TES make federated analytics possible, deploying them across different research environments is far from plug-and-play.

Open-source development is often driven by community backlogs that capture the needs and requirements of users. These collaborative efforts generate robust technical solutions but deploying  them in real-world research environments must also account for local differences in infrastructure, governance, and data. Three major challenges frequently emerge:

• Lack of TRE interoperability: TREs often use different providers or platforms that do not easily integrate with each other.
• System fragmentation: Local TREs operate under different governance frameworks, policies, and procedures for accessing sensitive data for research.
• Lack of data standardisation: Differences in clinical definitions, diagnosis coding (like SNOMED or ICD), and data structures can make it difficult to analyse datasets across institutions and countries.

Through the 51±¬ÁÏÍø Federated Analytics programme, open-source solutions such as and the are helping to address some of these challenges and promote interoperability across international research environments. These tools have now been deployed in Singapore’s national TRE () and at .

Learning from experience

The experience of deploying Bunny and Five Safes TES across the UK and Singapore has shown that the work of building a federated research network extends well beyond installing new software. While these tools provide the technical foundation for federated analytics, making them usable and sustainable in real research environments requires coordinated effort across infrastructure, governance, skills, and data. Several key challenges were found in practice.

Deployment and coordination

Working across international partners means coordinating development and deployment across institutions and time zones. Tools often require additional configuration to run effectively on local platforms, where differences in infrastructure and security requirements shape how solutions can be implemented.

Maturity of the technology

Many federated analytics tools originate in research settings and must evolve to meet the reliability, security, and performance standards required for production use. Local software and hardware decisions can also introduce dependencies that affect how easily solutions scale across environments, alongside the need for ongoing maintenance and support.

Information governance

Federated analytics across borders requires governance approaches that operate across organisations and jurisdictions, while still aligning with local regulatory and ethical requirements for the responsible use of sensitive data. This includes clear accountability, audit mechanisms, and oversight to ensure data is accessed and used appropriately.

Skills and organisational support

Researchers and technical teams need training, documentation, and guidance to use these tools effectively. At the same time, organisations must have sufficient technical capacity to support onboarding, maintenance, and continued use.

Data

Differences in data structures and standards remain a core challenge. Approaches such as OMOP mapping help enable more consistent analysis, but access must also be balanced with a usable and practical researcher experience.

While the technology for federated analytics is rapidly advancing, it is not yet plug-and-play. Sustained progress depends on developing the governance frameworks, technical capacity, skills, and data standards that allow these tools to support research reliably at scale.

Connecting for impact

By testing and deploying federated analytics tools across TREs in the UK and Singapore, this collaboration is helping demonstrate how international health data research can be carried out securely and in practice.

This work shows how researchers can analyse data across institutions and countries while ensuring that sensitive data remains within secure environments. In doing so, it helps lay the foundations for studies that draw on larger and more diverse populations, supporting research into shared health challenges such as cancer and diabetes.

Just as importantly, the programme is helping to advance the technical and governance infrastructure needed to make federated analytics sustainable – from deploying open-source tools aligned with international standards, to developing the processes and expertise needed to support researchers working across multiple research environments.

Go far, go together

The experience of deploying federated analytics across the UK and Singapore reinforces an important lesson: sustainable and trusted digital research infrastructure is not created by technology alone. It requires collaboration across disciplines, organisations, and countries to develop the governance frameworks, technical capabilities, and shared standards that make this work possible.

If you would like to be part of this growing community and contribute to the development of federated health data research, you can join the 51±¬ÁÏÍø Federated Analytics programme community .