As a contribution to assessment of quality of survival in malignant disease a rating scale for emotional distress (EDS) has been developed. The scale consists of eight items with explicit definitions, and the ratings are based on short interviews. The scale was applied on 34 patients with malignant blood diseases, and ratings were performed at regular intervals both prior to and during cytostatic treatment courses. The patients accepted the interview procedure well and the scale proved to be easy to apply in a clinical setting. Raters with little previous experience (nurses and physicians) had good interrater reliability (r = 0.98). The EDS scores of patients in good somatic condition increased significantly (P less than 0.01) during cytostatic treatment courses while no change was found among the more severely ill. There was a tendency for pure cytostatic schedules to inflict more discomfort than those also containing high doses of steroids, indicating that the scale may discriminate between various treatments. There was a significant correlation (r = 0.73, P less than 0.01) between the EDS scores and the physical condition of the patients. The results indicate that the EDS may be a reliable and sensitive instrument which adds an important aspect to the evaluation of the quality of survival in the treatment of malignant disease. It may also be of value when comparing different treatment regimens.