Use linked data assets
Data linkage refers to the bringing together of information from more than one source that relates to the same person. This allows researchers to tell a bigger story than analysing data from just one source. For example, comparing the data on women who received the HPV vaccination with data on women who developed cervical cancer provided strong evidence that overall, the vaccination was effective in reducing cervical cancer.
This strategy includes using linked data assets to provide a more complete picture of people’s use of services and health outcomes. By linking information over time, these assets support analysis of care pathways, including how people access and move through the health system.
This approach enables more detailed and policy‑relevant insights than can be obtained from a single data source and can improve understanding of patterns of care and outcomes.
This strategy includes undertaking more in-depth analysis and targeted analytical projects to answer complex questions about care pathways, patient journeys and sexual and reproductive health outcomes.
A range of data linkage assets are held or being developed by the AIHW and other data linkage authorities, including the ABS.
Key data assets that could provide greater insights into sexual and reproductive health include:
- the AIHW's National Health Data Hub (NHDH), a data asset that draws together core administrative health, disability, welfare and aged care data sets
- the ABS’s Person Level Integrated Data Asset, a secure data asset that combines information on health, education, government payments, income and taxation, employment and population demographics (including the Census) over time.
Understanding care pathways, symptoms and outcomes related to endometriosis: an example of how data linkage could be used
Linking hospital, perinatal MBS and PBS data with primary health care and other data could show how people with symptoms consistent with endometriosis move through the health system over time.
Linked data could help to show:
- when people first seek care for symptoms of endometriosis
- how often they see different health professionals
- whether symptoms are managed in primary care or lead to hospital admission
- how long it takes to receive a diagnosis of endometriosis, and the factors associated with faster diagnoses
- what treatments are provided, such as medicines or surgery, and the association with certain health outcomes
- what are the fertility experiences and outcomes, such as use of medically assisted reproduction, live births, pregnancy loss or involuntary childlessness
- what are the longer term impacts on health and wellbeing.
Linking hospital data across different types of care also allows related pregnancy outcomes to be analysed. For example, analysis could examine whether people with endometriosis are more likely to experience pregnancy loss, and what care they receive before and afterwards.
Data linkage is complex and requires strong governance, privacy protections and technical infrastructure. Access to linked data may be limited by approvals, data availability and data quality, and linked data may still be constrained by the scope and limitations of the original data sources. As a result, it is recommended that data linkage be undertaken where feasible and alongside other data sources to support robust and appropriate interpretation of findings.