JMIR Publications

Background:

Supervised fresh-air breaks (FABs) are widely used in inpatient psychiatric units to support patient well-being, but staffing, environmental, and safety constraints limit their availability. Virtual reality (VR) may offer an indoor, immersive alternative that replicates natural environments and supports emotional regulation in confined settings.

Objective:

To evaluate immediate and short-term changes in mood, anxiety, and subjective distress following a brief, nature-based virtual fresh-air break (VFAB), and to assess feasibility, acceptability, usability, presence, cybersickness, and safety among adult psychiatric inpatients.

Methods:

A prospective, single-arm pilot study was conducted on a 20-bed locked medical–surgical psychiatric unit at Tufts Medical Center between July and August 2025. Eligible adults were English-speaking, able to provide informed consent, and deemed by the treating team to be suitable for participation without any acute behavioral or safety concerns. Participants completed one 15-minute VFAB session using a VR headset while seated in a monitored private room. Mood, anxiety, and subjective distress were rated using 0–10 numeric scales before, immediately after, and 48 hours after the session. Post-session questionnaires assessed presence (modified Igroup Presence Questionnaire), usability (System Usability Scale [SUS]), and cybersickness (Virtual Reality Sickness Questionnaire [VRSQ]). Acceptability and feedback were collected post-session, and medical records were reviewed for safety outcomes and medication administrations within 48 hours. Data were analyzed using Friedman tests for repeated measures and Wilcoxon signed-rank tests with Bonferroni correction.

Results:

Of 64 patients screened, 39 (61%) were eligible and 26 (67% of eligible) participated. Of 26 enrolled participants, all completed baseline and immediate post-session assessments, and 25 completed 48-hour follow-up. Most participants (19/26, 73%) were restricted from physical FABs. Median anxiety scores were lower immediately following VR compared to pre-VR (5.0 pre-VR vs 2.0 post-VR; P = .018) and median distress scores were similarly lower (5.0 vs 1.0; P = .001). At 48 hours, scores had returned toward baseline. Mood scores were higher immediately post-VR (5.0 vs 7.0), although this difference did not reach statistical significance (P = .052; overall P = .066). Usability was high (median SUS 80.0, IQR 72.5–86.9) and cybersickness was low (median VRSQ 2.0, IQR 0–6). General presence was high (median 2.0). Most VR sessions were completed (23/26, 88%), and 69% of participants strongly recommended the experience; 77% wished to repeat it. seizures, agitation, or use of restraints observed within 48 hours.

Conclusions:

In this uncontrolled single-arm pilot study, VFAB was feasible, acceptable, and well tolerated in an acute psychiatric unit. Within-person reductions in anxiety and distress were observed immediately after the session, but not at 48 hours. Controlled studies are needed to determine causal efficacy and comparative value.

Clinicaltrial: