Empathy is a relevant attribute in the context of patient care. However, a decline in empathy throughout medical education has been reported in North-American medical schools, particularly, in the transition to clinical training. The present study aims to longitudinally model empathy during medical school at three time points: at the entrance, final of pre-clinical phase and at the beginning of clinical training. Data collected with the adaptation to Portuguese of the Jefferson Scale of Physician Empathy (student version) were analysed with latent growth modelling, conditioned by gender, openness and agreeableness. Empathy scores at all times were higher for females than for males, but only significantly at the end of the preclinical phase. The model showed a satisfactory fit level and the primary finding was that undergraduate medical student's empathy did not decline over time. Empathy scores were significantly and positively related with Openness to Experience and Agreeableness at admission, but the empathy rate of change across time was not significant. The stability of empathy revealed by a longitudinal methodology applied for the first time to empathy studying, contradicts previous results of decline and contributes to the understanding of the empathy development of medical students.