Storyline: Making data available

DATE: 23-08-2024 STATUS: ADOPTED

 

Precondition

• There is a dataset that a dataholder wants to offer for multiple use. It can be

  • research data (e.g. generated by means of an EDC tool) or a result dataset of a central or federated analysis

    • The data is generally already made FAIR in Storyline: Generating research data

    • The terms of use have already been agreed in Storyline: Generating research data

  • Healthcare data transformed into available research data

Process model

1. The dataproducer asks the Local review committee data providing whether and under which conditions the data may be offered for multiple use.

2. The local review committee data providing determines, based on the pre defined data policy, whether and under which conditions the data may be offered.

3. The data governance committee

   • Indicates for previously defined data how to make it FAIR, which coding and modeling to apply if the data is not or not sufficient FAIR

4. The data producer produces, creates and stores the original health data on the basis of agreements made and guidelines for, among other things, storage of data, ownership, intellectual property, terms of use, selection and retention. The data producer preferably produces the data FAIR (see note above)

5. The Data Preparator and Data Producer use an existing (reusable) interface or create new reusable interface (preferably in collaboration with the software supplier data producer), between the data producer and the data preparator.

6. Based on the data-centric principle, the Data Preparator separates the original health data from the original application and prepares the original health data, where necessary, for multiple use (including for use for research and innovation) in a persistent data platform.  

   • If the data producer has not made the data sufficiently FAIR, the Data Preparator will, where possible, further make it FAIR (see note above)

   • If the data producer has made the data sufficiently FAIR, the Data Preparator will implement the data-centric principle from a persistent data platform, which acts as a transport layer towards the data provider.

   • The data preparator ensures that the metadata of the dataset is up-to-date

7. The data provider sets up the dataset delivery process in accordance with the authorization metadata of the dataset recorded by the data preparer and, if necessary, sets up a target transformation (e.g. if a pseudonymization still needs to be performed).

8. The data provider registers and publishes the metadata at a FAIR data point of their choice, either as a separate application or integrated into an application with data repository or metadata catalog functionality.

9. The data provider registers the FAIR data point used with a FAIR data point register (see e.g. FAIR Data Point) if this has not already been done. It is the responsibility of the data guide to regularly and/or triggered by a search query to retrieve the latest state of affairs of the metadata from the FAIR data points. (see Storyline: Search data in metadata

10. The data provider makes the dataset immediately available if the dataset is public.

Postcondition

• The data is available on a persistent data platform

• The data is made FAIR

• The data is

  • readily available if it's public

  • available after an approved request and additional processing (e.g. data minimisation and pseudonymisation) by the data provider

• The metadata of the data is published and can be found

Open

Process diagram''Making data available”