This study evaluates the efficacy of a cognitive-behavioral treatment for generalized anxiety disorder (GAD) that addresses two types of worries: (a) those about situations that are amenable to problem solving, and (b) those about situations that are not. The treatment's goal is to help patients become more tolerant of uncertainty by discriminating between both types of worry and applying the correct strategy to each type. A multiple baseline design was used and subjects were 4 adults with a primary diagnosis of GAD. Treatment outcome was assessed with daily self-monitoring, self-report questionnaires, and standardized clinician ratings. At posttest and 6-month follow-up, 3 of 4 subjects no longer met diagnostic criteria for GAD and had attained high end-state functioning. At 12-month follow-up, none of the subjects met GAD diagnostic criteria but end-state functioning was variable. The results also show that treatment outcome was highly related to change in intolerance of uncertainty.