Abstract Background We introduced video-based teaching in pediatrics. We evaluated the impact of a pediatric video program on student performance in assessing pediatric patients presented as video cases. The program consisted of a library of pediatric videos, and inclusion of these in the teaching and examination for pediatric medicine. Methods Medical students on a pediatric clerkship at the University of Copenhagen assessed eight short pediatric video cases during autumn 2011 and spring 2012. Two independent observers evaluated a subset of records in a pilot study. A blind evaluation was made of the written records of 37 students before, and 58 students after, the introduction of the program using a Rubric score with four domains. Results The intraobserver interclass correlation coefficient was 0.94 and the interobserver interclass correlation was 0.71(n=25). The students’ mean total Rubric score in spring 2012 (7.0) was significantly higher (p