Background. Over the last decades, the number of health professionals who participate in Continuing Medical Education (CME) through e-learning, has rapidly grown, due to its flexibility. In Italy, healthcare professionals in active clinical practice are required to keep their knowledge up to date, through this scheme.One of the most relevant challenges is represented by dropouts. Recent CME studies emphasize the dropout rates between 30% and 60 %. Since 2004, the Training Office of the Italian National Institute of Health (ISS) provides CME e-learning courses in public health, using Active and Problem Based Learning (PBL) methodology. The aim of this study was to identify characteristic of dropout participants in e-learning courses delivered during the 2017-2019 CME triennium. To understand factors that underline the dropout rate and consequently, hypothesize actions to improve retention and completion rates. Methods. First we analyzed 21 courses (n= 44.630 participants) within the triennium 2017/19. Henceforth, we chose a single course for retrospective analysis. We ran a linear probability model through Stata 14.2, on age, gender, region, type of profession and occupation. Finally, we developed a matrix, based on the 7 steps of PBL, to identify the critical points of participants withdrawal. Results. On the triennium 2017/19, we found the mean dropout rate of 34% and completion rate of 66%, with a great variability according to the level of interaction, enrolment modality, and accreditation. Based on five points LIKERT scale questionnaire on participants satisfaction into contents of the course, teaching methodology and platform function, out of 44.630, around 26635 (61.6%) who filled in the questionnaire, expressed their satisfaction (4.57 points out of 5). In the retrospective analysis, out of 11.198 participants, the retention rate was 6356 (56,8%) and dropout rate of 43,2% (n=4842). According to linear probability model, significance is noted on gender and age. Occupation type seems not to have significant differences in terms of retention between freelancers and unemployed compared to employed. Between regions and professions, there is also a variation in retention. Conclusions. The results shade a light on key variables may be contributing to retention and dropouts, including critical points at the courses offered by ISS. Levels of interaction, gender and age seems to play a key role on participants to withdraw either remain in the course.