Body image is an important endpoint in quality of life evaluation since cancer treatment may result in major changes to patients' appearance from disfiguring surgery, late effects of radiotherapy or adverse effects of systemic treatment. A need was identified to develop a short body image scale (BIS) for use in clinical trials. A 10-item scale was constructed in collaboration with the European Organization for Research and Treatment of Cancer (EORTC) Quality of Life Study Group and tested in a heterogeneous sample of 276 British cancer patients. Following revisions, the scale underwent psychometric testing in 682 patients with breast cancer, using datasets from seven UK treatment trials/clinical studies. The scale showed high reliability (Cronbach's alpha 0.93) and good clinical validity based on response prevalence, discriminant validity (P<0.0001, Mann-Whitney test), sensitivity to change (P<0.001, Wilcoxon signed ranks test) and consistency of scores from different breast cancer treatment centres. Factor analysis resulted in a single factor solution in three out of four analyses, accounting for >50% variance. These results support the clinical validity of the BIS as a brief questionnaire for assessing body image changes in patients with cancer, suitable for use in clinical trials.