Problem-based learning (PBL) allows students to learn medical statistics through problem solving experience. The aim of this study was to assess the efficiency of PBL modules implemented in the blended learning courses in medical statistics through knowledge outcomes and student satisfaction. The pilot study was designed as a randomized controlled trial that included 53 medical students who had completed all course activities. The students were randomized in two groups: the group with access to PBL modules within the blended learning course (hPBL group) and the group without access to PBL modules–only blended learning course (BL group). There were no significant differences between the groups concerning socio-demographic characteristics, previous academic success and modality of access to course materials. Students from hPBL group had a significantly higher problem solving score (p = 0.012; effect size 0.69) and the total medical statistics score (p = 0,046; effect size 0.57). Multivariate regression analysis with problem solving as an outcome variable showed that problem solving was associated with being in hPBL group (p = 0.010) and having higher grade point average (p = 0.037). Multivariate regression analysis with the medical statistics score as an outcome variable showed the association between a higher score on medical statistics with access to PBL modules (p = 0.045) and a higher grade point average (p = 0.021). All students in hPBL group (100.0%) considered PBL modules useful for learning medical statistics. PBL modules can be easily implemented in the existing courses within medical statistics using the Moodle platform, they have high applicability and can complement, but not replace other forms of teaching. These modules were shown to be efficient in learning, to be well accepted among students and to be a potential missing link between teaching and learning medical statistics. The authors of this study are planning to create PBL modules for advanced courses in medical statistics and to conduct this study on other universities with a more representative study sample, with the aim to overcome the limitations of the existing study and confirm its results.

Background: Randomization allows study groups to be formed so that they are similar in all characteristics except outcomes. Aim: The aim of this study is to examine the frequency of randomization methods and their effect on achieving baseline balance in cluster randomization studies conducted in schools. Methods: A literature search of MEDLINE bibliographic database shows that the total number of collected articles in full text was 343, out of which 81 were eligible for inclusion. Each publication was reviewed by two independent reviewers, and data was extracted and analyzed. Results: Stratification was the most commonly applied randomization method, reported in 28 trials (34.6 %). There was no statistically significant difference in the number of subjects and clusters, as well as in cluster size between trial?s groups in studies in which simple randomization was used. However, there was a statistically significant difference in number of subjects and clusters between groups in trials in which restricted randomization methods were used. Yet, there was no difference in the cluster size. Conclusion: Although there is no difference in the size of clusters between trial arms, either at the level of the entire sample or in relation to randomization methods applied, additional research should be conducted on larger sample in order to establish the effect of randomization methods on baseline balance, when the size of clusters is in question.