Date created: | Last Updated:

Category: Project

Description: # Background The ReAD-StuD project performs a quantitative analysis and deduplication of registered studies that were performed in Germany. This study had the aim to join all German trials from clinicaltrials.gov (via AACT), the International Trials Registry Platform (ICTRP) and the German Clinical Trials Register (DRKS). # Data Data from the respective platforms was downloaded at the beginning of 2021. The resulting data is available as a csv-file in all_trials_de.csv. Since the table is too large to browse it on OSF, we have created a random sample of 1000 trials from that table which is in all_trials_de_sample1000.csv. This table can be produced using the supplied scripts. # Scripts To reproduce all_trials_de.csv and the plots from our corresponding article, the four scripts should be executed in order. 1 - Data: Load the data files from AACT, ICTRP and DRKS and create data frames for the individual data bases 2 - Random Forest: Train a Random Forest model for identifying duplicates between the data bases 3 - Join: Join the three data frames using primary IDs and the Random Forest model to create all_trials_de.csv 4 - Plots: Plots for our corresponding paper which illustrate overlaps between data bases, pre-registration trends and the number of trials in all data bases over time