People Research Data Commons

The challenge

The data that health researchers need is distributed across government, research and health service sectors. Researchers and innovators face challenges knowing whether useful data exists, who holds it, and how to access it.

For this focus area, we aim to make the health data, software and models researchers need to improve healthcare more findable, accessible, reusable and interoperable (FAIR). 

Our activities

We’re developing a Discovery Framework that describes a consistent set of informatics practices – including for metadata, identifiers, vocabularies and metadata syndication – to enable a national view of health data. Health Data Australia will serve as a catalogue of Australian health data assets and related research outputs by acting as a central metadata aggregation point. ARDC partners from academia, government, National Collaborative Research Infrastructure Strategy (NCRIS) facilities, health services and industry can adopt and adapt the Framework to suit their specific needs and objectives.

For our next steps, we’re working with:

• the academic research sector to support the efficient collection, management and sharing of health studies data to improve healthcare via our Health Studies Australian National Data Asset (HeSANDA) activity
• key government agencies to facilitate discovery and streamline access to health data for researchers via our Government Health Data Assets activity
• NCRIS facilities for consistency in their data collection, curation and access standards to make their data findable and accessible as national data assets via our NCRIS FAIR Data and Secure Access activity.

The challenge

Health information of patients and research subjects must be kept secure and not disclosed during the process of research or service improvement. Data custodians therefore require researchers to use secure environments for data analysis. Researchers don’t always have access to secure environments for the data they want to share or access, or when secure environments do exist, they can inadvertently create data silos that impede national-scale multidisciplinary collaborations.

Our activities

Through our Trusted Research Environments activity, we’ve established a National Trusted Research Environment (TRE) Framework in response to calls for a national conversation around a TRE’s core features, a reference architecture, and principles to underpin consistency and interoperability. The ARDC is harnessing expertise from a number of previous TRE projects such as SeRP, ERICA and CADRE. The Framework provides a common blueprint and offer best-practice guidelines beneficial to many universities and research institutions seeking accreditation under the Office of the National Data Commissioner (ONDC) Data Scheme.

Aligned with the Framework, we’re establishing networks of secure access services that will:

• provide TRE coverage for secondary use of data in strategic academic research areas, such as clinical trials and cohorts
• be consistent, appropriate, interoperable and comprehensive across government, NCRIS, and the broader research sector for the data researchers need.

We’re also looking to establish a community of practice to support data custodians and the users and providers of TREs.

The challenge

To harness the potential of medical records, there is a growing push for systematic, secure and ethical access to them for research. However, challenges remain in governance, ethics and the inconsistent formats across healthcare settings. This highlights the need for standardisation, leading to the development of common data models like the Observational Medical Outcomes Partnership (OMOP), which improves the use of healthcare data for research.

Building on our previous work to transform hospital electronic medical records (EMRs) into an international gold standard OMOP Common Data Model (CDM), we’re working with researchers, health services, infrastructure providers and policymakers to promote and support standardisation of data structures, common medical concept definitions, and common identifiers for concepts, people, resources, etc. across different data sources. Such a common data model across Australia’s health system enables system-wide research questions to improve healthcare.

Our activities

Through our Data Integration activity, we’re developing a national framework that will outline opportunities for data contributors in research or healthcare to easily adopt common data models, ontologies and identifiers and join up data to increase its value.

The next steps are to:

• convene a national alliance of research and health services to roll out common data models nationally and establish research access pathways to medical record data as part of ongoing national infrastructure
• support national and international standards for data collection during health studies
• establish and operate infrastructure for discovering and querying standardised data.

The challenge

Applying advanced analytics techniques like machine learning to national sensitive data collections across multiple secure data repositories raises data challenges that require innovative solutions.

Our activities

Through our Advanced Analytics in Healthcare activity, we completed 2 defining projects in advanced analytics in August 2024. 

In the first project, we worked with the Australian Data Science Network (ADSN) to co-develop a National Infrastructure Framework for Health Analytics. The framework project captures, documents and prioritises AI infrastructure needs in:

• underpinning hardware infrastructure
• national reference data assets
• underpinning federated and foundational infrastructure
• shared tools and platforms
• coordinated socio-technical systems (STS) assets on a national level – including training, policy and guidelines, culture, and community support. 

In parallel with the Framework project, which elicits challenges and needs broadly, a Pathfinder project was co-developed to delve deeper into the well-known challenge for machine learning that sensitive data cannot leave the premises. This project laid the foundation for federated learning infrastructure.

The 2 projects informed us about the constituting parts of the future infrastructure architecture for advanced analytics in healthcare. Based on the projects, we are developing the infrastructure architecture and a roadmap. With these plans ready, we are running a coordinated set of infrastructure development activities in collaboration with:

• universities, medical research institutes and national research collaborations
• government data custodians and health systems players
• NCRIS facilities.