DMP: Data sharing

[K-Beicher]

Data will not be made available without application
Metadata will be made available through our website for people to browse, this will be set up using Seedcase Flower
We expect metadata to be published as we start to collect the data, we will be able to do a dress rehearsal with the data from the feasibility study, which we don't expect many researchers to use subsequently as this data is mainly for course correcting the main study
Most applications will come from people associated with the study (we expect)
All applications must detail why they need to specific data points (especially if they belong in a sensitive category)
Once an application has been approved data will be made available in a new project folder on GenomeDK where the raw data must remain, only scripts and results should be exported
It will also be possible to ask for a new data set to be uploaded to Denmark Statistics (DS), where it will be available in the SCDA project database
No data should be used directly from REDCap for analysis purpose
Some data from the feasibility study will need to be shared and analysed before we start the main study, this will need to be thought through and arranged as soon as possible (Daniel to think about this)
Data from the main study will as a rule not be shared until the data collection has finished for all participants
Some analysis will be possible using the Clinician Portal and the Reporting Module from LIVA, this should be encouraged to monitor progress for participants but should (as a rule) not be used for publication
Any script created to extract sub-data sets for individual analysis projects on GDK will be kept in the main study project folder
There is a data sharing agreement in place between the five centres
We have yet to make a decision on any IP or license for shared data
A guidance document on how to apply for data should be written

[danielibsen]

Data analysis using data from the feasibility study will be done before all of this is finished. We must work on this on a case-by-case basis. People still need to get approval through the publications panel. I think we should aim for the system to run well after the feasibility study has finished.