Ten steps to project completion: a short guide for researchers Linkage process summary
All data linkage amendment requests requiring changes to the study design, cohort, data flow, data collections or variables must be reviewed by your Project Manager or other Data Integration Services Centre (DISC) staff prior to lodgement in EthOS.
Please send your completed Amendment Request form to your Project Manager or to [email protected]. Please allow at least two weeks prior to the EthOS amendment submission deadline for our review.
For projects with a Technical Assessment (all projects from approximately early-2019):
Amendment requests without clean and tracked Technical Assessments or that have not been reviewed by DISC staff will not be progressed.
Where re-linkages or other substantial work is required as a result of an amendment request, you may be charged for these services. Please ask us for a quote prior to submitting any amendment application.
Links to AIHW-held and external data collections and variable lists for some of these repositories can be found within our data collections. Not all these collections are able to be linked.
The Data Integration Services Centre will assist with obtaining data custodian approvals for AIHW-held data collections.
A Public Interest Certificate (PIC) is required to be signed by a delegate of the Minister for Health of the Commonwealth for the purposes of paragraphs 130(3)(a) of the Health Insurance Act 1973 and 135A(3)(a) of the National Health Act 1953. The PIC certifies that it is necessary in the public interest to divulge the requested data from Medicare (Medicare Benefits Schedule or Pharmaceutical Benefits Scheme), Centrelink or some other data collections, to the Principal Investigator.
An application for a PIC to be issued by the appropriate government department will be made by the AIHW on behalf of the Principal Investigator once AIHW Ethics Committee approval has been obtained.
Where a project requires Medicare Benefits Schedule, Pharmaceutical Benefits Scheme, Centrelink or certain other data, it is a data custodian requirement that the linked dataset must be stored and analysed within an approved secure research access environment.
Approved secure research access environments include:
The use of a secure research access environment enables release of data to researchers that otherwise would not be possible.
It is recommended that researchers make enquiries about arranging a workspace within one of these environments as soon as possible. There are costs associated with establishing secure research access environment workspaces.
All researchers wishing to access any datasets are required to sign an Undertaking of Confidentiality with the AIHW. This document describes the penalties outlined in the Australian Institute of Health and Welfare Act 1987 (section 29) relating to the unauthorised disclosure of information or attempts to re-identify study participants.
For all projects involving linkages of Pharmaceutical Benefits Scheme data to any non-standard source such as private sector or overseas data collections, additional Departments of Defence and Veterans’ Affairs Human Research Ethics Committee approval is required, unless RPBS information has been excluded.
The AIHW is a member of the Population Health Research Network, an Australian Government collaboration that has built a nationwide data linkage infrastructure capable of securely and safely managing health information from around Australia. This includes Australian Government and state and territory government data.
For multi-state linkage projects, you may choose to manage the project through the Population Health Research Network data linkage online application system.
The Population Health Research Network is an initiative of the Australian Government, conducted as part of the National Collaborative Research Infrastructure Strategy.
For prospective studies or where it is possible to contact participants, the AIHW Ethics Committee expects that, where feasible, researchers will seek written consent from participants to access their Medicare Benefits Schedule or Pharmaceutical Benefits Scheme data from Services Australia.
Please contact Services Australia to obtain their preferred consent form: [email protected]
If this pathway is taken for studies that require only Medicare Benefits Schedule or Pharmaceutical Benefits Scheme data, there is no requirement for AIHW Ethics Committee approval.
These projects are instead approved by Services Australia's External Request Evaluation Committee (EREC).
Visit the AIHW’s Australian Cancer Database (ACD) for information about this collection.
To access ACD information, researchers must seek approval from data custodians in each state and territory. Some jurisdictions also require additional ethics approvals.
Contact details for ACD data custodians.
The approval process can take more than six months, so make sure you allow plenty of time.
If other collections also need to be linked, work can begin while ACD data custodian approvals are finalised.
As the Repatriation Pharmaceutical Benefits Scheme (RPBS) is part of the Pharmaceutical Benefits Scheme (PBS), prescription information relating to Defence personnel and veterans is routinely captured in PBS data requests.
Where data about Defence personnel or veterans are to be analysed or reported separately, approval must also be sought from the Departments of Defence and Veterans’ Affairs Human Research Ethics Committee (DDVA HREC).
Please note that for studies that are not approved by the DDVA HREC, PBS extracts provided to researchers will be recoded to ensure that prescription information cannot be attributed to Defence personnel or veterans.
Approval must be sought from data custodians of state and territory admitted patient and emergency department data collections.
Some states and territories also require additional ethics approvals.
State and territory linkage units can help you with this step. State and territory linkage unit contact details.
If, in addition to your sub-population of interest, you are looking to gain access to records for comparison purposes there are a number of issues you should address in your application.
Under the Beneficence underpinning of the National Statement on the Ethical Conduct in Human Research, the National Health and Medical Research Council requires the balancing of risk with benefit. (See Sections 1.6, 1.7, 2.1.3 and 2.1.4 of the National Statement.) Providing more data than is required to achieve the goals of an analysis could unnecessarily increase risk (for example increase the risk of re-identification or the level of harm should a breach occur).
To this end, Applicants need to provide justification as to why they require access to the data in question. Information relevant to this justification includes:
The volume of data that makes up any comparison group should not be larger than is required to meet the objectives of your study.
When designing your analysis and considering your control/comparison group you should adhere to the following principles:
In determining the required comparison sizes the following approaches may be considered (based on Ury, 1975 and Foppa and Spiegelman, 1997):
There may be studies where these rules will not be sufficient to meet the requirements of the planned comparison (Principle 2). In these cases, you will be required to provide information justifying the need for access to additional data.
One such justification would be to demonstrate the number of control/comparison records needed to achieve the study aims. To do this, first of all determine the statistic and the statistical test that you will use in your analysis. Then consider the power required. That is, determine the size of the difference between the study and control/comparison groups that you want to detect and the level of confidence with which you want to detect it.
Once the required power is determined, along with how the two groups are to be selected and which statistic and statistical test will be used, it is usually possible to calculate the sample size required. One piece of information required to perform this calculation is the variance of the property of interest in the comparison/control group. If the variance can’t be calculated from the data in question then an alternate estimation or approximation is required. This can come from historical data, other data sources or (at worst case) an educated approximation.
In the case that there is more than one property of interest it is necessary to choose one or more design variables that the selection is designed around. These can be selected by importance to the analysis, availability of information about the property (including variance), expected high variance / low power or a combination thereof.
You would include the power requirements and calculations in your technical assessment to support your justification for access to more control/comparison records than indicated by the general rule.
If you need assistance with this aspect of your application the AIHW may be able to help. The availability of assistance will depend on AIHW resourcing levels and would be charged on a cost recovery basis.
Ury, H.K. 1975. Efficiency of case-control studies with multiple controls per case: continuous or dichotomous data. Biometrics 31(3), 643-649.
Foppa I. & Spiegelman D. 1997. Power and sample size calculations for case-control studies of gene-environment interactions with polytomous exposure variable. American Journal of Epidemiology 146(7):596-604.
Please contact your Project Manager (or email [email protected]) to obtain instructions for preparing and formatting your dataset.
Data files must not be sent by standard email.
Data files should be securely sent to the AIHW, using our recommended secure messaging service. Please contact your Project Manager (or email [email protected]) to arrange access.
The Data Integration Services Centre charges for linkage work on a cost–recovery basis. We can provide a quote estimate if you contact us at [email protected].
Here are some quote examples (excluding GST):
Costs increase where:
This website needs JavaScript enabled in order to work correctly; currently it looks like it is disabled. Please enable JavaScript to use this website as intended.
We'd love to know any feedback that you have about the AIHW website, its contents or reports.
The browser you are using to browse this website is outdated and some features may not display properly or be accessible to you. Please use a more recent browser for the best user experience.