The Centre for Victorian Data Linkage (CVDL) has now fully implemented its new cloud-based technical platform, ensuring that researchers, health services and government have access to timely linked data from around 40 administrative and clinical datasets.
The CVDL has moved from a largely manual linkage and integration infrastructure to an automated Microsoft Azure system with new linkage software, LinXmart incorporated.
Previously the CVDL was only able to update the Integrated Data Resource (IDR) on an annual basis due to the amount of manual intervention required, but the automated processes and data pipelines now enable monthly updates.
This means that data is much more current, and new datasets can be incorporated much more quickly. This increased currency is of great benefit to linked-data users, who can be confident that they have access to the most up to date data available.
The CVDL has also incorporated value-added components to the IDR, to make the complex linked data more user-friendly for researchers and analysts. These include summary tables of person demographics and service usage, and derived variables which draw on characteristics from across the various datasets.
CVDL is also piloting other value-added components, including a familial linkage table, which facilitates identification of family relationships in the linked data.
The CVDL has developed an updated version of its secure data access environment, the Victorian data Access Linkage Trust (VALT). VALT enables dissemination of linked data within a secure-based cloud environment, with analysis undertaken via Virtual Machines (VMs).
VALT moved from the beta version to full implementation in 2021 and there are now around 100 active project VMs running in VALT. Feedback from users indicates that the VMs are powerful and suitable for undertaking complex analysis of large datasets.
The updated version of VALT, known as VALT2, enables the linked data to be stored on a remote server rather than held locally within the project VMs. This is particularly useful and efficient where projects involve multiple researchers and analysts as the single database of linked data can be accessed by more than one VM.
If you have any questions relating to the VALT system or our plans for implementation, contact Mark Sipthorp, Technical Manager, Centre for Victorian Data Linkage via [email protected].