|Page tools: Print Page Print All RSS Search this Product|
Data integration brings together information from multiple sources to help answer important questions about people, families, places, businesses or life events. Combining health survey results with Census and administrative government data can help provide a better picture of the health of Aboriginal and Torres Strait Islander peoples.
Linked (or integrated) data that is used for analysis in data integration projects is de-identified, meaning it does not contain identifying information like names and addresses. Other important measures are taken to ensure the data is not likely to identify an individual and all information is protected under the Census and Statistics Act.
More information about data integration can be found on the ABS website at www.abs.gov.au/dataintegration.
The Multi-Agency Data Integration Project (MADIP)
One way the ABS currently integrates data is through the Multi-Agency Data Integration Project (MADIP). MADIP is a secure approach for combining information from across areas of government like healthcare, education, government payments, personal income tax, Census and other population demographic data over time.
Current MADIP Data Sources
The ABS is one of six Australian Government agencies that work together to manage MADIP. The MADIP allows authorised researchers to better understand socio-economic outcomes and trends to help inform policy, services and programs to make sure they are valuable to the people and communities who need them. Information from MADIP cannot be used for compliance purposes, like checking if you are paying enough tax, or whether you are eligible for government payments.
More information about MADIP can be found on the ABS website at www.abs.gov.au/madip.
Privacy and Security
The ABS takes data security very seriously and prioritises protecting our respondents’ privacy. We fully comply with the Privacy Act 1988, the Australian Privacy Principles and all other relevant legislation.
Data is only integrated where there is a clear public benefit and can only be used for statistical and research purposes. The ABS enables integration of the datasets that are part of MADIP, but these datasets are not held together. They are stored separately and information from them is only brought together for specific approved projects and the ABS uses a secure environment for access and storage. Authorised researchers are only granted access to anonymous information, and only the information they need for their project. The ABS checks data outputs before they are shared. The ABS does not release information in a manner likely to enable the identification of an individual.
Access to MADIP data is only available to authorised users. Before anyone can access integrated data, the project must be approved by all relevant data custodians and the ABS. All projects must meet the requirements of the ‘Five Safes Framework’ which is in line with international best practice and designed to identify and manage any risks to keep the information safe and secure.
The Five Safes Framework covers separate but related dimensions:
Opportunity to integrate Health Survey and administrative data
In December 2019, the first results from the 2018-19 NATSIHS were released. There is a lot of interest from researchers about combining NATSIHS data with government administrative data, such as through MADIP. Currently, no NATSIHS or National Aboriginal and Torres Strait Islander Social Survey (NATSISS) data are used in any data integration project.
The NATSIHS cannot collect everything about peoples’ health, or associated social and economic circumstances. It also represents one point in time, which means there can be limitations in using it to assess outcomes and impact, and telling stories about peoples’ life journeys. Integrating NATSIHS data would mean more information is available with the survey data, which would allow researchers to see more complete pictures of people’s health and wellbeing. For example, a topic that could be explored is whether use of medical services and medications follows the best care pathways (eg. using Medicare and PBS data). Data integration could also enable more complex analysis and open up more possibilities for local level data. We would also be in a better position to explore what information can and can’t be gained from administrative data to reduce the burden of future surveys, reduce survey length, or create space in surveys for the data that communities need.
In your response to this paper, the ABS is seeking your input on the following data integration items:
These documents will be presented in a new window.