¹Research Division, Federal Institute for Drugs and Medical Devices (BfArM), Bonn, Germany
²Department of Clinical Medicine, Department of Clinical Epidemiology, and Center for Population Medicine, Aarhus University, Aarhus, Denmark
³Department of Bioinformatics, Fraunhofer Institute for Algorithms and Scientific Computing (SCAI), Sankt Augustin, Germany
⁴INFARMED, National Authority of Medicines and Health Products, I.P., Health Technology Assessment Department (DATS), Lisbon, Portugal
⁵School of Pharmacy, University of Eastern Finland, Kuopio, Finland

Text

Background: Evidence derived from real-world data (“real-world evidence”, RWE) is increasingly being considered in regulatory decision-making processes at national and international levels. Observational data have become an established source of knowledge, particularly for the assessment of drug safety. The possible use of this data throughout the entire life cycle of a drug are as diverse as the data sources themselves.

The objective is to provide an overview of the current use of RWE into regulatory decision-making and to provide insights from two research projects conducted at the German Federal Institute for Drugs and Medical Devices (BfArM): FQrisk and Real4Reg.

Materials and Methods: FQrisk is a cohort study based on nationwide claims data from the German statutory health insurance AOK examining the safety of fluoroquinolone antibiotics using current good pharmacoepidemiological practice. Real4Reg is an EU-funded multi-stakeholder project that leverages various observational data sources from Denmark, Finland, Portugal, and Germany. It focuses on the heterogeneity of observational data and the possibilities for optimising the current standard of analysis using AI-driven approaches such as super learners, large language models and neural networks for propensity scores, conditional average treatment effects or the identification of confounding factors.

Conclusion: The role of RWE in regulatory research and decision-making is contextualised by the presentation of the two research projects, while potential challenges, solutions and use cases for the application of observational data in regulatory research are discussed.