Group Cognitive Behavioral Therapy With Virtual Reality Exposure Versus In-Vivo Exposure for Social Anxiety Disorder and Agoraphobia: Underpowered Results From the SoREAL Pragmatic Randomized Clinical Trial.

Background: Social anxiety disorder (SAD) and agoraphobia are common, impairing conditions often treated with cognitive behavioral therapy (CBT) conducted in groups. In CBT, exposure therapy is a core element. However, in-vivo exposure therapy is logistically challenging and aversive for both patient and therapist, especially in a group context, often leading to exposure being skipped altogether in clinical practice. Virtual reality exposure (VRE), in which phobic stimuli are presented through immersive virtual reality technology, has shown promise as a flexible alternative to in-vivo exposure. We thus hypothesized that using VRE would result in more overall exposure and more individualized exposure, resulting in statistically significant symptom reduction compared with a group using in-vivo exposure. Objective: This trial evaluated the efficacy of group CBT with VRE (VR-CBT) versus CBT with in-vivo exposure for treating SAD and agoraphobia in clinical settings. Methods: In this randomized, parallel-group, assessor-blinded trial, 177 participants with SAD (n=150) or agoraphobia (n=27) as a primary diagnosis were assigned to either VR-CBT (n=81) or traditional CBT (n=96) across 5 Danish mental health outpatient clinics. Both groups received 14 weekly group sessions. The difference between the 2 treatments was that the VR-CBT group received exposure therapy via head-mounted displays (HMDs) displaying 360° videos of anxiogenic situations for individuals with SAD (eg, presenting at work) and agoraphobia (eg, faulty elevator), while the CBT group conducted traditional in-vivo exposure exercises (eg, presenting to the group, using the clinic elevator). Primary outcomes were phobic anxiety reductions, measured by the Liebowitz Social Anxiety Scale and the Mobility Inventory for Agoraphobia at baseline, posttreatment, and 1-year follow-up (from baseline). Secondary outcomes included work and social functioning, depressive symptoms, and quality of life. Results: Both groups showed significant reductions in primary, secondary, and exploratory outcomes, with no significant differences between groups at posttreatment (d=-0.026) and 1-year follow-up (d=0.097). Baseline characteristics and attrition rates were balanced across the groups. Conclusions: Due to insufficient recruitment and substantial missing data, no definitive conclusions can be drawn regarding group differences between VR-CBT and traditional CBT in group settings. The feasibility issues encountered suggest that careful consideration of the benefits and limitations of VR technology is essential before implementation in clinical practice.