HEAL-Scot is developing a new, secure, and standardised blueprint to bring together and link data from different domains, such as housing, environment, location, and health.
Although some research has accessed linked data from multiple domains, this is the exception rather than the rule. These studies have had to establish new relationships with those managing data access and create new agreements to allow data sharing. Additionally, Trusted Research Environments (TREs) have had to navigate how to accurately link data across different domains for individual projects to ensure that information relates to the same person before de-identification takes place.
The current situation significantly limits the breadth of complex research into the social determinants of health: exploring the real-world factors that impact on our health and wellbeing.
HEAL-Scot (Housing, Environment and Location Data Linkage in Federated TREs to Map Geospatial Inequalities) aims to create the foundation for TREs to support research requiring access to cross-domain data as standard. This will open up the possibilities for more robust evidence that will inform policies to address health inequalities and positively influence the delivery of public services.
HEAL-Scot is a close partnership between DataLoch and the Health Informatics Centre (HIC) – two Scottish TREs – working together to better understand and map data pipelines and deliver an exemplar research project.
Through this partnership, the team will provide real-world feedback for the following TREvolution innovations:
- Standardised Architecture for Trusted Research Environments (SATRE) v2.0, a framework that defines best-practice standards;
- K8TRE, a SATRE-compliant, practical implementation of a TRE that can be deployed to any cloud-provider as well as on-premises infrastructure to support federated projects across TREs;
- Research Object Crates (RO-Crates), an integrated view of an entire research project – its methods, data, results, and research team; and
- Semi-Automated Checking of Research Outputs (SACRO) to assess initial disclosure risk of research outputs before release from a TRE.
What HEAL-Scot aims to achieve
By the end of the project, the HEAL-Scot team aims to create a secure and reusable way to link housing, environment, and location data to health data that will better enable research into population health. Their goals include:
- Accessing multi-domain data for population research – Securing access to example datasets covering different domains: housing, environment, access to green and blue spaces (e.g. parks, recreational spaces, lakes, beaches, etc), access to health care services, and air quality;
- Linking multi-domain data across Scotland – Standardising multi-domain data extracts from three Scottish regions and enable their secure and accurate linkage to health data; and
- Investigating social factors impacting preventable deaths – Carrying out an exemplar research study focused on social factors that influence preventable and treatable death.
Progress so far
Within the first six months of delivery, the HEAL-Scot team has:
- Developed a shared understanding of the data standards at both DataLoch and HIC to permit a common foundation for linking the data required. This is the basis for developing a technical blueprint for multi-domain data linkage, which is in draft form.
- Successfully sourced an extract of aggregate finance data to be used alongside the health, housing, and location data already hosted by DataLoch and HIC.
- Shared learning with the GROVE (Governance for Household-Level Environment and Health Data) team, a DARE-funded project focused on housing and health data linkage.
- Co-produced a living glossary document covering commonly used technical language to ensure a shared understanding.
- Developed the Tiers of Identifiability overview as a discussion aid for two public-facing workshops focused on the possible confidentiality challenges caused by multi-domain data linkage.
- Planned an exemplar research project that will evaluate the utility of the technical blueprint and TREvolution tool use with non-health data.
In terms of next steps, DataLoch and HIC will link and de-identify a data extract and make this available for the exemplar research project in a single K8TRE workspace.
The HEAL-Scot team is developing a lessons-learned report, which – alongside the technical blueprint and public workshop insights – will inform their final recommendations for TREs looking to link health with data from other domains.
Relevance to TREvolution
The TREvolution programme seeks to build upon TRE best practice and develop secure and standardised ways of working that keep sensitive data safe. The HEAL-Scot team is exploring how specific tools – K8TRE, RO-Crates and SACRO – can be used where non-health and health data is required for a project spanning more than one TRE. As such, the team’s efforts will also contribute to assessing how the federation standards in SATRE v2.0 apply to cutting-edge research needs.
About the DARE UK Research Exemplars
The DARE UK Research Exemplars are short, impactful research projects demonstrating how emerging Trusted Research Environment (TRE) capabilities can support real-world sensitive data research across the UK. Working across multiple TREs and research domains, these projects bring together researchers, TRE operators, public contributors and other partners to apply and evaluate innovations developed through TREvolution. Their role is to test new approaches in live research settings, generating practical evidence on what works well, what challenges remain, and how sensitive data research can be strengthened for the public good.
