Data linkage in the Statistical Services Branch
Frequently Asked Questions
-
Requesting only ready-linked data from the Master Linkage File (MLF) means we can process your request faster than those involving longer time periods and new data sets. For data sources and time periods covered by the MLF please refer to: Data Collections available for Linkage.
For further factors associated with time taken to complete a linkage request refer to this timeliness matrix.
-
This depends on both the workload of the linkage team, and the complexity of your request. A simple linkage of a small cohort (say, 1,000 patients) to the master linkage file may take 2-3 days depending how many variables are requested in the output. Linkage of several datasets over a longer period of time than that covered by the master linkage file, with a large number of variables and data manipulation can take months.
You should allow a number of months in your project plan for approval and data linkage.
For further factors associated with time taken to complete a linkage request refer to this timeliness matrix.
-
Ideally we need the following items for a successful linkage: study/patient ID, first name, surname, address, post code, date of birth, sex. We can work with most formats – text files (.csv, .txt) and Excel are common. However, a long list of patients in a Word file would not be a useful format. Files must be password protected, and can be sent by email if file size allows. See the Data Linkage Information Sheet about providing data to SSB for more details.
-
The data, as described in the approved PHA, will be provided either by a password protected file or via a secure method of transfer such as Kiteworks. Other secure methods of transfer should be discussed during the PHA negotiations. The password, if necessary, can be obtained on receipt of the output. We can accommodate most output formats. We tend to supply generic text formats (.csv, .txt) that can be read by packages such as SPSS and R. For more information about the format of the data see Data Linkage Training Resources.
-
The scope and coverage in the Master Linkage File (MLF) varies for each collection. The Data collections available for linkage page outlines each collection's start dates within the MLF. Requesting only ready-linked data from the MLF means we can process your request more efficiently.
-
You will need to submit amendments for both ethics and PHA approval. Changes to the PHA include additional researcher names, additional data items and/or changes to the time period of the data requested. See How do I submit an update to my approved PHA? for more details.
-
Anyone who will have access to a linked data file must be named on the PHA form. Please list individual names and position titles in Sections 3 and 4 of the PHA form. Position titles alone are not acceptable to the Data Custodian. Data files and/or passwords will only be provided to persons named on the PHA form. It is illegal for those who are named on the PHA form to allow anyone who is not listed in the PHA to access the unit record level linked data. Should additional researchers or staff on your project require access to the linked data, you will need to contact Office of Research and Innovation (ORI).
-
For work to commence on your extract, the Statistical Services Branch must have a PHA application signed by the relevant Data Custodians and a letter of approval signed by the Director-General/Delegate relating to your project. Following these approvals, the commencement of work depends on a number of factors, such as existing workload and whether we are in receipt of all the data sources needed for the project. If you have a strict deadline, it is best to plan ahead and allow plenty of time for the approval process and extract to be completed.
-
The data requested in the PHA is for the use of researchers named on the PHA and the project described within the PHA. The extracted data cannot be used for additional projects outside of the scope of the PHA. Should additional researchers wish to access the data (for the purposes of the project described in the PHA) then an amendment must be submitted to the Data Custodian. If researchers wish to use the data for purposes outside of the original PHA, then a new PHA should be submitted to the Data Custodian for approval.
-
You can submit your PHA to DLQ@health.qld.gov.au before you have ethics approval. The Data Custodian can review your PHA regardless of the status of your ethics application. However, you will need evidence of ethics approval before you are given approval to access the data.
-
Data can be described as 'potentially identifiable' if it contains items that identify (or could possibly lead to identification of) an individual. Examples of these items are: Names and addresses, Contact details (e.g. telephone numbers, email addresses), Date-based data items (e.g. birth data, admission date, discharge date, date of death), demographic items (e.g. age, sex, country of birth, indigenous status), facility (e.g. hospital name), identifying numbers (e.g. medical record numbers, private health insurance numbers), rare characteristics such as one person aged 109 years, one of quintuplets, long length of stay etc., a combination of sex, date of birth and postcode/SA2/SLA. If only one or more of the persons or organisations are potentially identifiable, then the data set is considered potentially identifiable. If you are still unsure, contact the Statistical Services Branch.
-
No. Ethics approval for a research project is not an automatic approval for a data custodian to release confidential health information. For Queensland Health to release confidential health information for the purposes of research, a researcher must obtain: approval from the relevant data custodian/s (to release health information), approval from the relevant data custodian/s (to release health information), approval from a Human Research Ethics Committee and Approval from the Director-General (or delegate). This process is governed by the Queensland Health Office of Health and Medical Research - further information on this process is contained on the Health and Medical Research Unit web site.
-
The Data Custodians for various QH data collections are listed in the Queensland Department of Health and HHS Data Custodian List. SSB can only authorise the release of data collected in the Queensland Hospital Admitted Patient Data Collection (QHAPDC), Queensland Health Non-Admitted Patient Data Collection (QHNAPDC), Perinatal Data Collection (PDC) as well as Death Registry and Birth Registry data.
-
You only need approval from the SSB Data Custodian if you are accessing corporate databases in the SSB such as the Queensland Hospital Admitted Patient Data Collection (QHAPDC) or the Perinatal Data Collection (PDC), or if you require SSB to link datasets for your project. Even if you are not accessing SSB datasets but you anticipate that SSB will be providing linkage services, you must consult with SSB.
Please note that the SSB has no jurisdiction over hospital charts or local databases maintained at the hospital. If you wish to access these you must negotiate directly with the relevant hospital or Hospital and Health Service staff.