Objective structured clinical examinations (OSCEs) are extensively used in many medical schools worldwide with the stated objective to assess students' clinical skills acquired during internships. The objective of the present study was to assess the factors associated with success in university summative OSCEs, especially the impact of previous hospital internships in corresponding disciplines and supervision during internships. This was a cross-sectional study assessing the results in the summative OSCEs of 4th year medical students during the 2021-2022 academic year in a French medical school. The summative OSCEs included five stations for each student. Each student answered a survey at the end of summative OSCEs about previous internships, the supervision they had and perceived difficulty levels for each station. The scores in each station were assessed according to previous hospital internships in the corresponding discipline. Analysis of predictive factors of success in OSCEs, defined by a score ≥ 10/20 at each station, were performed using univariate and multivariate logistic regression. Out of the 220 students participating in the summative OSCEs, 182 (83%) answered the survey. One hundred and forty-four (79%) of these students had carried out hospital internships in at least one of the disciplines evaluated during the OSCEs. Students having completed an internship in the corresponding discipline had significantly higher OSCEs scores for interrogation, communication, therapeutic education and procedure stations compared to those who had not. Previous internship in corresponding disciplines was independently associated with success in OSCEs in interrogation (OR 9.45 [1.34-66.8] p = 0.02), clinical examination (OR 6.93 [1.88-25.57] p = 0.004, and therapeutic education (OR 3.09 [1.22-7.82] p = 0.02) stations. Previous hospital internships in the discipline evaluated by the OSCEs are associated with success in summative OSCEs. This reinforces the importance of student involvement during their hospital internships.