There are an increasing number of students with learning difficulties attending university, and currently much debate about the suitability and ability of students with dyslexia at both medical school and once they graduate into clinical practice. In this study we describe the performance of students with dyslexia compared to fellow students in extended matching questions (EMQ), short answer question (SAQ) and observed structured clinical examinations (OSCE) and discuss the implications of differences identified. End of year assessment results for 5 cohorts of medical students were analysed. Students with dyslexia did less well overall in all assessment types in year 1 but this difference was not evident in later years. Dyslexic students who were allowed extra time in written assessments did better than dyslexic students who did not have their assessment concessions in place. When station type within OSCE assessments was analysed students with dyslexia did less well in both examination skills and data interpretation stations in years 1, 2 & 3. In conclusion, differences in performance in written assessments are only evident early in training and may be partly due to delayed adjustment to medical school or implementation of assessment concessions. Performance in individual OSCE stations is dependent on station type. Why students with specific learning difficulties (SpLDs) perform less well in examination skills and data analysis OSCE stations requires further investigation.