Stressful life events and negative mood have been associated with elevated blood glucose and poor self-care in individuals with diabetes. The purpose of this controlled study was to determine the effect of mood state, specifically depression, anxiety, and daily hassles on the outcome of biofeedback assisted relaxation in insulin dependent diabetes mellitus. Eighteen subjects completed the study, nine in biofeedback assisted relaxation and nine in the control group. There were no significant group differences in blood glucose between those receiving biofeedback assisted relaxation and the subjects continuing usual care. Five of the nine experimental subjects and one of the nine control subjects were identified as succeeders according to an arbitrary criterion. Treatment failures were more depressed, more anxious, and took longer to complete the protocol than succeeders. Statistically significant correlations were found between high scores on inventories measuring depression, anxiety, and hassles intensity and higher blood glucose levels and smaller changes in blood glucose as a result of treatment. It is suggested that mood has an important impact on the response to biofeedback assisted relaxation. Further research is necessary to determine whether assessment of anxiety and depression followed by appropriate treatment where necessary should precede biofeedback assisted relaxation in insulin dependent diabetes.