Home
Sharing with third parties
Research data may only be shared with an external commercial party if the study subjects have provided informed consent for this. You should not hand over exclusive rights to reuse or publish your research data to commercial publishers or agents without retaining the rights to make the data openly available for reuse.
If your research involves the exchange or linkage of privacy-sensitive data with (one or more) third parties, always follow the guideline for secure data linkage. To make sure you work in the best way possible a number of tools have been developed.
Secure Data Linkage Toolkit:
- Privacy Assessment tools
- Decision Tree
- NL Data Linkage Form
- Standardized linking procedures incl. Best Practice examples
- Trusted Third Party (TTP) (explanation by ELSI Helpdesk)
For more information, please also read ‘Data delen en koppelen’ at the ELSI helpdesk’.
Frequently Asked Questions
-
Is there a checklist for data sharing?
A Dutch checklist for sharing data has been developed by Radboud UMC.
-
What are the required pre-conditions for data sharing?
The captured data from one or more sources are made available in a research environment.
The research environment is well defined in terms of:
- purpose;
- consent;
- ownership and governance;
- infrastructure and architecture (e.g., a distributed virtual research environment).
and argued with respect to:
- fit for purpose;
- level of data sensitivity;
- level of completeness;
- level of security.
-
What are the Levels of Data Sensitivity?
Level 5 - Extremely sensitive information about individually identifiable people
Level 4 - Very sensitive information about individually identifiable people
Level 3 - Sensitive information about individually identifiable people
Level 2 - Benign information about individually identifiable people
Level 1 - De-identified research information about people and other non-confidential research information
-
What determines successful data linkage from a research perspective?
The success in linkage is determined by a number of factors (described in detail in Biolink):
- Choice in linking variables: name and address, sex, date of birth
- Choice of algorithm: either looking for deterministic (exact matching) variables or probabilistic (similar) variables.
- Amount of data overlap
- Amount of errors in the dataset
-
What are real-life examples of data linkage?
- The paper ‘Record linkage for health studies: three demonstration projects’ describes three data linking projects:
- The Netherlands Twin Register (NTR) linked with the Achmea Health Database (AHD).
- The KOALA cohort with a number of pharmacies in the database of the Stichting Farmaceutische Kengetallen (SFK).
- The population register (Basisregistratie Personen, BRP, formerly the Gemeentelijke Basisadministratie) and employment register (ER, in Dutch: Werknemersbestand).
- The paper ‘Using record linkage to construct a population-‐based cancer patient cohort with multiple disease outcomes’ describes methods and experiences of the Netherlands Cancer Institute in the construction of a population-based breast cancer survivor cohort linked with cardiovascular disease and mortality registries.