We suspected that safety of medication in clinical studies is not sufficiently guaranteed, despite a variety of safety nets. In order to check whether this suspicion is correct, we conducted a retrospective analysis of 75 patients enrolled in 11 clinical studies conducted at our hospital in the last couple of years. We focused specifically on the number of clinically relevant interactions between study medication and concomitant medication and on the information in the study protocol to prevent such interactions. Our analysis showed that approximately 45% of study participants experienced at least one clinically relevant interaction and that not a single study protocol provided sufficient information to prevent interactions with study medication. This means that study participants are at risk to experience unnecessary adverse effects caused by interaction of medicines. In this article, we give recommendations to improve medication safety in clinical studies.