OBJECTIVE: Cognitive-behavioral therapy (CBT) is a first-line treatment for panic disorder with agoraphobia (PD/AG). Nevertheless, an understanding of its mechanisms and particularly the role of therapist-guided exposure is lacking. This study was aimed to evaluate whether therapist-guided exposure in situ is associated with more pervasive and long-lasting effects than therapist-prescribed exposure in situ.

METHOD: A multicenter randomized controlled trial, in which 369 PD/AG patients were treated and followed up for 6 months. Patients were randomized to 2 manual-based variants of CBT (T+/T-) or a wait-list control group (WL; n = 68) and were treated twice weekly for 12 sessions. CBT variants were identical in content, structure, and length, except for implementation of exposure in situ: In the T+ variant (n = 163), therapists planned and supervised exposure in situ exercises outside the therapy room; in the T- group (n = 138), therapists planned and discussed patients' in situ exposure exercises but did not accompany them. Primary outcome measures were (a) Hamilton Anxiety Scale, (b) Clinical Global Impression, (c) number of panic attacks, and (d) agoraphobic avoidance (Mobility Inventory).

RESULTS: For T+ and T- compared with WL, all outcome measures improved significantly with large effect sizes from baseline to post (range = -0.5 to -2.5) and from post to follow-up (range = -0.02 to -1.0). T+ improved more than T- on the Clinical Global Impression and Mobility Inventory at post and follow-up and had greater reduction in panic attacks during the follow-up period. Reduction in agoraphobic avoidance accelerated after exposure was introduced. A dose-response relation was found for Time × Frequency of Exposure and reduction in agoraphobic avoidance.

CONCLUSIONS: Therapist-guided exposure is more effective for agoraphobic avoidance, overall functioning, and panic attacks in the follow-up period than is CBT without therapist-guided exposure. Therapist-guided exposure promotes additional therapeutic improvement--possibly mediated by increased physical engagement in feared situations--beyond the effects of a CBT treatment in which exposure is simply instructed.