Effectiveness of Self-Guided Virtual Reality-based Cognitive Behavioral Therapy for Panic Disorder (Preprint)

BACKGROUND Virtual reality (VR) is an effective technique as a traditional cognitive behavioral therapy (CBT) and a promising tool for treating panic disorder (PD) symptoms because VR exposure can be safer, with higher acceptability, than in vivo exposure; it is more immersive than exposure through imagination. CBT techniques can be delivered more effectively using VR. Thus far, VR has required high quality devices; however, the development of mobile VR technology has improved users' availability. Meanwhile, a well-structured form of VR can be reproduced and used anywhere, which means that VR is appropriate for self–guided treatment to address the high-treatment costs of evidence–based therapy and the lack of professional therapists. This study investigates the potential of self-guided VR as an alternative to high-cost treatment. OBJECTIVE The study’s main goal is to offer data about the efficacy of the mobile app–based self-led VR CBT in the treatment of PD. METHODS 54 subjects with PD were enrolled and randomly assigned to either the VR treatment group or waitlist group. The VR treatment was designed for a total of 12 sessions in 4 weeks. The VR comprises 4 steps in which patients are gradually exposed to phobic stimuli while learning to cope with panic symptoms. The effectiveness of the treatment was assessed through the PD Severity Scale (PDSS), the Hamilton Rating Scale for Depression (HRSD), the body sensation questionnaire (BSQ), the Albany Panic and Phobia Questionnaire (APPQ), the Anxiety Sensitivity Index (ASI), the state-trait anxiety inventory (STAI), the hospital anxiety and depression scale (HADS), the social avoidance and distress scale (K-SAD), the inventory for depressive symptomatology self-report (KIDS-SR), and the perceived stress scale, (PSS). Additionally, physiological changes using heart rate variability (HRV) were evaluated. RESULTS In total, 40 subjects (20 VR treatment and 20 waitlist patients) were included in the final analysis. For the PDSS scores, the main time (F_1,39 = 20.76, P < 0.01, η_p^2 = 0.09) and group-by-time interaction (F_1,39 = 10.59, P < 0.01, η_p^2 = 0.04) effects were significant. The post-hoc tests showed that the PDSS scores decreased significantly in the VR group (t_37 = 2.68; P = 0.01), but not in the waitlist group. The group-by-time interaction effect on the HADS total scores (F_1,39 = 5.51, P = 0.02, η_p^2 = 0.01) was significant, as was the main time effect on the STAI_total (F_1,39 = 4.32; P = 0.04) and STAI_S (F_1,39 = 6.00; P = 0.01) scores; however, there were no statistically significant between-group differences on the other scales. CONCLUSIONS The self-guided, mobile app–based VR was effective in treating panic symptoms and helped restore the autonomic nervous system, demonstrating the validity of VR for self-guided treatment and its cost-effective therapeutic approach.