JMIR Publications

ABSTRACT

Background:

Feeding difficulties and restrictive eating patterns are common among autistic children and adolescents, but evidence on immersive virtual reality (VR) for nutrition education in this population remains limited. NutriVRTEA was cocreated as an immersive VR intervention to support nutrition education within routine clinical care, with attention to sensory tolerability, interaction design and professional supervision.

Objective:

This study evaluated participant-perceived usability and acceptability of NutriVRTEA among autistic children and adolescents, together with professional-perceived usability. Intervention completion and tolerability were assessed as feasibility outcomes, while changes in attitudes toward healthy eating, Mediterranean diet adherence and feeding behaviors were examined as exploratory outcomes.

Methods:

We conducted a prospective, single-center, uncontrolled, single-arm pilot study at a child and adolescent mental health center in Badalona, Spain. Participants completed 4 individually supervised VR sessions of approximately 30 minutes, delivered about 2 weeks apart. The intervention used hand tracking, guided audiovisual instructions and 4 progressively structured nutrition activities, with professional supervision through a companion tablet. Participant-perceived usability, acceptability and tolerability were assessed after the intervention using study-specific questionnaires normalized to 0-100 scales. Professional-perceived usability, intervention completion and adverse events were recorded. Exploratory eating-related outcomes were assessed before the intervention, after session 4 and from 6 months after completion of the intervention. Longitudinal changes were analyzed using Friedman tests with Kendall W, followed by Wilcoxon signed-rank comparisons with Holm correction.

Results:

Twenty participants were included, with a mean age of 11.20 years (SD 2.76); 15 (75%) were male. Nineteen participants completed all 4 sessions (95%). Mean participant-perceived usability was 67.27 (SD 13.19), tolerability was 82.73 (SD 19.53) and acceptability was 89.47 (SD 15.36). Mean professional-perceived usability was 92.71 (SD 10.05) among the 3 professionals who completed the assessment. No adverse events required interruption of the intervention or specific clinical management. No statistically significant overall changes were identified in attitudes toward healthy eating, Mediterranean diet adherence or feeding-related behaviors across assessment points (all P≥.06). None of the pairwise comparisons remained statistically significant after Holm correction (all adjusted P≥.17).

Conclusions:

The findings suggest that NutriVRTEA can be delivered as a professionally supervised, 4-session immersive nutrition education intervention for autistic children and adolescents in routine outpatient care. Completion was high, and acceptability and tolerability were rated favorably on study-specific scales. No adverse events required interruption of the intervention or specific clinical management. Participant-perceived usability scores covered a wide range, indicating that interaction design may still require refinement. Exploratory analyses did not provide evidence of changes in eating-related outcomes. Future controlled studies should include prespecified measures of educational learning and real-world dietary behavior and examine whether virtual food familiarization is associated with changes beyond the immersive environment.