The present study investigated the efficacy of a coping-technique, applied relaxation (AR) and cognitive therapy (CT), in the treatment of generalized anxiety disorder. Thirty-six outpatients fulfilling the DSM-III-R criteria for generalized anxiety were assessed with independent assessor ratings and self-report scales before and after treatment and at a 1 yr follow-up. The patients were randomized and treated individually for 12 weekly sessions. The results showed that both treatments yielded large improvements, which were maintained, or furthered at follow-up. There was no difference between AR and CT on any measure. The drop-out rate was 12% for AR and 5% for CT. The proportions of clinically significantly improved patients were 53 and 62% at post-treatment and 67 and 56% at follow-up for AR and CT, respectively. Besides affecting generalized anxiety the treatments also yielded marked and lasting changes on ratings of worry, cognitive and somatic anxiety and depression. The conclusion that can be drawn is that both AR and CT have potential as treatments for generalized anxiety disorder but they have to be developed further in order to increase the efficacy to the level usually seen in panic disorder, 80-85% clinically improved.