Standardisation and Reproducibility
Responsibilities within the Institute
The division Studies and Statistical Consultancy develops and implements tools and workflows which allow reproducible analysis and transparent publications. We actively support the introduction and application of these techniques throughout the Institute.Selected activities
Current projects of the division Studies and Statistical Consultancy already implement an automated controlled analysis workflow from data import to publication-ready table and figure delivery with version management.
We implemented reproducible data management workflows by transforming clinical study data captured in REDCap into the internationally recognized CDISC data standards (SDTM for standardized raw study data and ADaM for analysis-ready datasets), thereby enabling transparent, reproducible, and traceable statistical analyses.
This workflow was applied in the HP alone vs HP+LTA for pbHH randomized pilot trial: Hiatoplasty with ligamentum teres augmentation vs dorsal hiatoplasty alone for the surgical treatment of postbariatric, symptomatic hiatal hernias.
It was also adapted to the randomized comparison of online interventions in the WESPA/GDDD trial: An online program with individualized vs automated support for significant others of depressed individuals.
We teach how to employ artificial intelligence / chatbots (LLMs) in the planning process of reproducible data science workflows.