Objective: To assess the objective and subjective risk communication skills of medical students in three universities in Germany METHODS: We developed a risk communication skills Objective Structured Clinical Examination (OSCE) station and implemented it in three medical schools in Germany. 596 students contributed data to a risk communication checklist and a risk communication self-assessment. Multiple linear regression models were used to identify factors associated with the risk communication OSCE performance. Results: Participants in our study achieved on average 73.5% of the total risk communication skills score, which did not differ between locations (F(2-595) = 1.96; p = 0.142). The mean objective performance of students who assessed their skills as poor was significantly worse than the performance of students who assessed their skills as good (t(520) = -5.01, p < 0.001). The risk communication skills score was associated with native language but not with gender nor General Point Average (p < 0.001). Conclusions: Medical students demonstrated acceptable risk communication skills scores and were able to self-assess their performance. However, selected communication techniques should be re-emphasised in the undergraduate medical curriculum. Practice implications: Our research identified shortcomings in particular subgroups that can be addressed through tailored curriculum interventions. Keywords: Communication skills; GPA; Gender; Medical education; Native tongue; Objective structured clinical examination (OSCE); Risk communication.