Full skeletal survey, localised radiographs and computed tomography (CT) examinations were compared with clinical assessment in the evaluation of treatment response of bony metastases in 20 patients with carcinoma of the breast. Conventional radiology, skeletal survey and localised views compared poorly with clinical assessment agreeing in only 35% and 50% respectively. CT concurred with the clinical assessment in 65% of patients, particularly with respect to healing (86%). CT predicted the effect of treatment in six additional patients and this was confirmed on follow-up assessment. It is suggested that the use of skeletal surveys in monitoring treatment response is limited and that for critical evaluation of treatment CT should be the method of choice.