<p>Abstract</p> <p>Background</p> <p>There is general consent that empathy is crucial for the physician-patient relationship and thus an important issue in medical education. This comparative study was designed to examine the differences in empathy between first year and final year medical students in Jimma University, Ethiopia.</p> <p>Methods</p> <p> A comparative cross-sectional study among 131 first year and 106 final year medical students was conducted in Jimma University, Ethiopia on academic year 2010/11. The study subjects were selected using simple random sampling technique from the list of the students. Study participation was voluntary. The Balanced Emotional Empathy Scale (BEES) was used for the detection of “heart-reading”, i.e. emotional empathy and the Reading the Mind in the Eyes test (RME-R test) to evaluate “mind-reading”, i.e. cognitive empathy. We performed <it>t</it>-test to compare the mean difference in empathy and RME-R scores between the two groups of students. A linear regression was computed to identify potential factors influencing the BEES and RME-R.</p> <p>Results</p> <p>Out of the total 237 students, 207 (87.3%) were males. The mean age of first year and final year students was 19.3 ± 1.1 and 24.0 ± 1.4 years respectively. First year students have scored 40.6 ± 23.8 while final year students scored 41.5 ± 20.8 mean in the BEES measuring emotional empathy score. However, this difference was not statistically significant (t = −0.30, df = 231, P-value >0.05). Final year students had significantly higher mean cognitive empathy score (17.8 ± 4.5) than first year students (14.4 ± 4.8) [β = 2.7, 95%CI (1.20, 4.13)]. Males had scored lower cognitive [β = −2.5, 95%CI (−4.37, −0.66)] and emotional empathy [β = −12.0, 95%CI (−21.66, −5.46)].</p> <p>Conclusions</p> <p>Low emotional (BEES) and cognitive empathy sores were found in first year and final year students of Jimma University could have implications on the medical education curricula. Medical education targeted at enhancing emotional empathy and increasing cognitive empathy is required by segmenting with gender for effective physician-patient interaction. The influence of empathy on clinical competence should be studied using more rigorous design.</p>