This study assessed the effectiveness of education reforms on student-reported learning outcomes at the end of the 5-year medical school (M5) and 1-year internship (HO) in 2006, 2007 and 2008. A self-administered anonymous survey with 17 learning outcomes assessed, derived from Harden's Three-Circle Outcomes Model for outcomes-based education, was administered to 683 students at the end of medical school (M5) and internship (HO) from 2006, 2007 and 2008. We identified learning outcomes which changed significantly for internship (Cohorts A, B and C) and medical school (Cohorts B, C and D) between cohorts from 2006 to 2008, and compared learning outcomes between medical school and internship within cohorts (i.e. Cohort B which was M5 in 2006 and HO in 2007; Cohort C which was M5 in 2007 and HO in 2008). The proportion of students who agreed that medical school helped them achieve learning outcomes increased significantly from 2006 to 2008 for 15 out of 17 learning outcomes assessed. The proportion of students who agreed that internship helped them achieve learning outcomes increased significantly from 2006 to 2008 for 6 learning outcomes assessed. For Cohorts B and C, internship was more effective than medical school in achieving 8 learning outcomes. Cohort C reported that internship was more effective than medical school in 3 additional learning outcomes than Cohort B: patient management, humility and dedication. We conclude that a successful journey of education reform is an ongoing process that needs to comprehensively address multifaceted components such as faculty, administration and curriculum.