Latest news

In clinical trials, a data dictionary is understood to be a catalogue of descriptions of the parameters to be collected, i.e. all variable names with data types, formats and possible response options and their coding. Without a data dictionary, it is impossible for third parties to understand shared data sets. In many projects, a tabular description of the variables is published in text form, but the description of the individual columns is often vague and misleading.

During the workshop, an already agreed minimal information model for data dictionaries was presented [1], which has already examined and standardised many common formats. Afterwards, it was discussed in smaller groups whether the model sufficiently covers the requirements from practice and in which form the model could be implemented and used in the future. The participants agreed that such coordination is also important with regard to European initiatives such as EOSC and EHDS. The new EFMI working group "FAIR data HRPO" [2] will deal with this and other topics.