Generalized anxiety disorder (GAD) has been effectively treated with cognitive behavioural therapy (CBT) in face-to face settings. Internet-delivered CBT could be a way to increase the accessibility and affordability of CBT for people suffering from GAD. The aim of this study was to evaluate the efficacy of guided Internet-delivered CBT for GAD in a controlled trial with a wait-list control group. A total of 89 participants were included following online screening and a structured psychiatric telephone interview. Participants were randomized to either an 8-week treatment group (n = 44) or a wait-list control group (n = 45). Treatment consisted of a self-help program based on CBT principles and applied relaxation along with therapist guidance. The main outcome measure was the Penn State Worry Questionnaire. Ratings of clinical improvement and symptoms were included as well as secondary outcome measures dealing with anxiety, depression, and quality of life. Among the treatment group participants, 13.6% did not complete the posttreatment measures. The treatment group showed significant improvement compared with the control group on all outcome measures. Large effect sizes (Cohen's d > 0.8) were found both within the treatment group and between the groups in favor of the treatment on all outcome measures except on a measure of quality of life. Results at 1- and 3-year follow-up indicated that treatment results improved or were maintained. The authors conclude that Internet-delivered CBT with therapist support can reduce symptoms and problems related to GAD.