Infrastructure

Governments around the world appreciate that linked administrative data can provide a unique resource for monitoring, evaluating and improving services. At Data Linkage Services, we are committed to developing dedicated data linkage infrastructure to automate routine linkages to provide reliable, valuable services and products using both Health and non-health related data.

The WA Data Linkage System (WADLS)

The WA Data Linkage System (WADLS) is a system of linkages within and between data collections, maintained and operated by Data Linkage Services.

Since its modest beginnings in 1995, the WADLS has grown to become one of the most comprehensive, high quality and enduring linkage systems worldwide. Today, this data linkage infrastructure contains over 150 million records spanning over fifty routinely linked datasets.

Many of these datasets represent the Health sector, however a variety of non-Health datasets are also linked. Hundreds of other “project specific” datasets have also been linked in the WADLS over its lifetime.

The WADLS comprises a large database of demographic information, linkage keys and associated metadata, along with a linkage software package, Data Linkage System No. 3 (DLS3), developed and supported by the Department of Health’s in-house Systems Team.

Data Engineers in the Data Linkage team implement linkage strategies using DLS3 and carry out a variety of quality checking protocols to ensure linkages are of a high quality.

More information about the WADLS and linkage quality is available in the following article:

  • Eitelhuber, T. W., Thackray, J., Hodges, S. and Alan, J. (2018) Fit for purpose – developing a software platform to support the modern challenges of data linkage in Western Australia. International Journal of Population Data Science, 3(3). doi: 10.23889/ijpds.v3i3.435.

The WA Health Enterprise Linked Data Warehouse

Data Linkage Services at the Department of Health have developed a consolidated Linked Data Warehouse to streamline application and approvals processes and reduce the complexity of data delivery for more timely provisioning of linked data.

The development of a whole-of-Government data linkage capability required expansion of the datasets maintained as part of the Western Australian Data Linkage System (WADLS) and the Custodian Administered Research Extract Server (CARES) to enable the creation of a ‘single’ linkage system for commonly used WA Government cross-sectoral data. 

WA Health Enterprise Linked Data Repository- basic schematic

Applications for data held entirely within the Linked Data Warehouse may be classified as a ‘low complexity’ request. The following datasets are currently available within the Warehouse:

  • Births
  • Deaths
  • Deaths (COD URF)
  • Electoral Roll
  • Child Development Information System (CDIS)
  • Emergency Department Data Collection (EDDC)
  • Mental Health Inpatient Data (MHIP)
  • Midwives Notification System (MNS)
  • Hospital Morbidity Data Collection (HMDC)
  • Trauma Registries
  • WA Cancer Registry (WACR)
  • WA Notifiable Infectious Diseases Database (WANIDD)
  • WA Register of Developmental Anomalies (WARDA)
    • Birth Defects
    • Cerebral Palsy

The Custodian Administered Research Extract Server (CARES)

The Custodian Administered Research Extract Server (CARES) was initiated in 2011 as a collaboration between Data Linkage Services and Data Custodians at the Department of Health to streamline the data extraction process and hasten data delivery for approved projects.

CARES went live in 2013, and was the standard method for preparing linked data extracts for ten years. In 2023, extraction processes were transitioned from CARES to the WA Health Enterprise Linked Data Warehouse.

More information on CARES can be found at:

  • Eitelhuber, T. , Davis, G. , Rosman, D. and Glauert, R. (2014), Western Australia unveils advances in linked data delivery systems. Australian and New Zealand Journal of Public Health, 38: 397-398. doi:10.1111/1753-6405.12258
  • Eitelhuber, T. and Davis, G.: The custodian administered research extract server: “improving the pipeline” in linked data delivery systems. Health Information Science and Systems 2014 2:6. doi.org/10.1186/2047-2501-2-6