Five projects awarded to Scottish Regional Safe Havens to simplify access to data

The project will assist in the further development of data federation, federated analytics and Trusted Research Environment (TRE) development for machine learning workflows. HIC operates a TRE for the Scottish Government and NHS Tayside and NHS Fife Health Board Regions delivering secure, research managed ‘access to’ data under robust governance control.  

Dr Christian Cole of the Health Informatics Centre, Dundee, said: “This funding from Research Data Scotland will enable and continue collaborative projects in HIC with the Scottish Save Haven Network to deliver advances in AI capability and national laboratory data. We look forward to continuing the strong relationship between HIC, RDS and the wider Scottish data communities.”

Professor Roger Halliday, Interim CEO of Research Data Scotland, said: “We’re delighted to fund these projects that share RDS’s vision to create the conditions for data-driven research and innovation to happen more systematically across Scotland.

“These projects will help researchers and users of data to identify common standards, test new approaches and help spread good practice to simplify access to data in a safe and secure way that ultimately save time, money and lives.”

The funding comes from RDS as part of its new System Development Fund, which has been set up to take forward the shared agenda for simplifying and aligning access to data, and improving data services for researchers.

The other four projects to be funded include:

Dataloch – Development of an agreed national standard for the de-identification of personal tags within Electronic Health Data (EHR)  



DataLoch is a secure data service that has been developed in partnership by the University of Edinburgh and NHS Lothian. DataLoch’s approach is to put data at the centre of responses to health and social care system challenges and thereby improve services through research, innovation and planning. This project, led by Dr. Arlene Casey, receives funding of £170,000 and will be in collaboration across the Scottish Safe Haven Network focusing on agreeing national standards and methods for de-identifying unstructured health records for use in research. Enabling access to these types of records, while maintaining patient confidentiality, has the potential to greatly improve health research across Scotland. 

University of Edinburgh – Led by Professor Gillian Raab, this project receives funding of £30,000 and will review and evaluate methodology on how to measure the disclosure risks from synthetic data. 

West of Scotland Safe Haven, delivering two streams in tandem: 

• Technical – providing advice to stakeholders on best practice in health data settings
• Governance – providing best practice guidelines on the risks posed by synthetic data 

The West of Scotland Safe Haven is a partnership between NHS Greater Glasgow and Clyde and the University of Glasgow, providing safe, ethically approved access to pseudonymised NHS health datasets on a secure ISO-accredited data analytical platform. This project, led by Dr Charlie Mayor receives funding of £200,497 and is set to deliver, in collaboration with DataLoch in Edinburgh, best practice guidance with synthetic health data. The creation of synthetic versions of real datasets offers opportunities to better protect patient privacy, whilst offering valuable access to realistic health data for research purposes 

DASH – Development and delivery of five separate federation projects  

The Grampian Data Safe Haven (DaSH) covers the geographical region of Aberdeen City, Aberdeenshire and Moray, Scotland and allows researchers to contribute to the public benefits of heath research through analysis of local, pseudonymised, patient-level data whilst ensuring individual patient privacy and confidentiality are safeguarded by rigorous principles of data security, patient confidentiality and patient anonymity. Projects awarded to DaSH from the Systems Development Fund receive £185,000 and will be coordinated by Katherine O'Sullivan. The five sub-projects are looking across the spectrum from establishing best practice, acceptable de-identification metrics for patient data to the validation of machine learning and creation of synthetic health datasets based on popular unconsented Scottish patient datasets. 

The full value of all projects awarded to Regional Safe Havens from RDS Systems Development Fund is £763,337.