Although medical curricula now adopt an integrated teaching approach, this is not adequately reflected in assessment of anatomy knowledge and skills. In this study, we aimed to explore the impact of the addition of clinical vignette to item stems on students' performance in anatomy practical examinations. In this study, 129 undergraduate medical students of Durham University took part in a 30-item anatomy practical test consisting of those with and without clinical stem, in a crossover design. Classical test theory was used to analyze item difficulty, discrimination index, point biserial, and reliability. Student performance on items with clinical stem and the percentage of students who correctly answered each item was significantly improved by the addition of a clinical stem in the Year 2 cohort. Also, items with a clinical stem showed much better discrimination index than non-clinical items in the Year 2 cohort. In contrast, there was no significant difference in item performance, student performance and discrimination index between items with a clinical and non-clinical stem in Year 1 cohort. Over 65% of test items in both year groups were of good quality with point biserial exceeding 0.2. However, Year 1 test reliability for non-clinical items was better than clinical items. The results raise question as to what level to apply this method of assessment in undergraduate education. Although interpreted on the basis of a relatively small item sample, the findings support the need for improving anatomy practical examinations in line with overriding curricula changes.