Twenty-one consecutive patients with severe tinnitus were interviewed using a structured psychiatric interview (the National Institute of Mental Health Diagnostic Interview Schedule) and were asked to complete the Hopkins Symptom Checklist (SCL-90) and the Chronic Illness Problem Inventory. They were compared to a control group of 14 patients attending an otolaryngological clinic with a complaint of hearing loss. The tinnitus patients had a significantly greater lifetime prevalence of major depression (62% vs 21%) than controls and a significantly higher prevalence of current major depression (48% vs 7%). The currently depressed tinnitus patients had significantly higher scores on all subscales of the SCL-90, except the phobia and paranoid subscales, compared to the non-depressed tinnitus group and on all scales compared to the controls. The number of psychosocial problems and thus the resulting disability experienced was significantly greater in the tinnitus group compared to controls and in the currently depressed tinnitus patients when compared to non-depressed tinnitus patients. In view of our results treatment should aim at not only alleviation of tinnitus, but also the frequently co-existing major depression.