JMIR Publications

ABSTRACT

Background:

Anxiety is the most common psychiatric comorbidity among people with epilepsy, yet epilepsy-specific (ES) interictal anxiety remains underrecognized and undertreated, highlighting the need for accessible non-pharmacological approaches. Virtual reality exposure therapy (VR-ET) can deliver controlled exposure to seizure-related fears that are difficult to reproduce in vivo, but its effects and delivery requirements need evaluation in the context of seizure unpredictability and heterogeneous seizure experiences. The epilepsy monitoring unit (EMU) allows repeated engagement under medi

Objective:

This mixed-methods randomized controlled trial evaluated the preliminary effectiveness of VR-ET for ES-anxiety compared with neutral VR, together with feasibility, acceptability, usability, safety, tolerability, and scenario fidelity among people with epilepsy admitted to an EMU.

Methods:

Fourteen people with self-reported epilepsy- or seizure-related anxiety were randomized to VR-ET (n=7) or neutral VR (n=7). Participants completed self-administered 5-minute 360-degree video sessions twice daily for up to 10 days or until discharge. VR-ET participants progressed through personalized hierarchies of graded seizure-related exposure scenes selected at baseline using a diagnostic interview; control participants viewed nature-based scenes in a standardized order. Quantitative outcomes included anxiety, fear and situational avoidance, quality of life, simulator sickness, usability, and presence, and were summarized descriptively. Post-intervention and 1-month interviews were analyzed using framework analysis with inductive coding within predefined domains. Findings were integrated through joint interpretation and cross-case comparison.

Results:

Quantitative outcomes were heterogeneous and showed no clear between-group pattern. Adherence was high, with all VR-ET participants and 5 of 7 (71.4%) neutral VR participants meeting the predefined threshold. Five of 7 (71.4%) VR-ET participants described the intervention as helpful, and 4 of 7 (57.1%) reported experiences consistent with exposure-related learning, including increased insight, self-efficacy, preparedness, cognitive reappraisal, and habituation. Fear and anxiety changes were variable, and incomplete 1-month follow-up limited interpretation of longer-term outcomes. Scenarios were generally realistic and representative of epilepsy-related fears, but emotional engagement depended on personal relevance. Seizure simulations were identified as most impactful, and participants recommended greater intensity, variety, interactivity, and treatment duration. VR-ET participants more often expressed interest in continued use outside the EMU, whereas neutral VR participants primarily described short-term calming and distraction during admission with some reporting proactive use during periods of heightened anxiety. Across groups, usability was favorable, simulator sickness was minimal, and no serious adverse events were attributable to VR. Engagement was affected by EMU-related fatigue, seizure activity, environmental interruptions, and technical problems.

Conclusions:

The VR interventions were feasible, acceptable, and safe in this sample. VR-ET showed no clear quantitative advantage over neutral VR, but qualitative findings were consistent with exposure-related learning. Neutral VR may provide low-intensity anxiety support during EMU admission. Future research should evaluate more personalized, varied, and interactive VR-ET and examine integration with therapist-supported care. Clinical Trial: ClinicalTrials.gov NCT06028945