Metaskills
RESEARCH · HEALTHCARE · PILOT EVIDENCE

What 117 participants across five Spanish pilot sites told us

Five healthcare and academic institutions, in a project that ran from October 2025 to March 2026. The ratings were good. The more useful half of the report is the list of things that were not, and this note publishes both.

VR Health Champions pilot evaluation, Spain, 2026

StudyConsortiumOngoing researchInternal pilot evaluation - not peer-reviewed

The campaign in numbers

117
Participants

Across five healthcare and academic institutions in Spain, in a project that ran from October 2025 to March 2026.

VR Health Champions pilot evaluation, Spain, 2026 (internal report).

5
Sites

Universities of Malaga and Granada, Hospital de Antequera, Hospital Materno-Infantil and San Juan de Dios Comillas.

VR Health Champions pilot evaluation, Spain, 2026 (internal report).

4,28/5
Overall rating

Average across the post-session surveys that were completed, not across all 117 participants. Coverage differed by site - 22 at Antequera, all 7 at Materno-Infantil.

VR Health Champions pilot evaluation, Spain, 2026 (internal report).

4,14/5
Perceived usefulness

Same post-test group. Pre-test expectations were already high at 4,55/5, so this is expectations met rather than exceeded.

VR Health Champions pilot evaluation, Spain, 2026 (internal report).

4,21/5
Would recommend it

Same post-test group, self-reported intention rather than observed behaviour.

VR Health Champions pilot evaluation, Spain, 2026 (internal report).

How did it go? campaign

The study was conducted in two phases. In the first stage, conducted in the fall of 2025 at the universities of Málaga and Granada, access via a web browser was tested with 51 medical students, most of whom had never used virtual reality before; at both centers, 19 and 28 participants, respectively, completed the pre-test questionnaire. The second phase was moved to clinical and teaching facilities in early 2026: in January-doctors from Hospital de Antequera; in February-gynecology and pediatrics residents from Hospital Materno-Infantil; and the following day-nurses from San Juan de Dios Comillas Hospital. In parallel with the pilot projects, 25 semi-structured interviews and 5 co-creation workshops were conducted with the participation of clinicians, educators, simulation coordinators, and training managers. The evaluation combined pre- and post-session surveys with qualitative feedback sessions and observations by the facilitators. At San Juan de Dios Comillas Hospital, 29 participants completed a pre-test questionnaire, and the evaluation following the pilot program was qualitative rather than based on a survey-therefore, the averages reported above for the campaign are not simply the average of 117 people.

The most important conclusion

Stability was non-negotiable. Browser instability in the initial phase-session crashes, data loss, incomplete logs-was the biggest obstacle to purchase intent throughout the entire campaign. This had nothing to do with scenario design, artificial intelligence, or price. A training tool that crashes in front of a group of doctors won’t get a second invitation, and the report clearly states that this problem needed to be resolved before any marketing activities began. The less pleasant side of this same conclusion: the above assessments were gathered from people who liked the concept. The obstacle was never whether people wanted it.

What clinicians told us to fix

Put the patient in the room

The paediatric scenario had no visible child avatar. Clinicians said it broke the clinical authenticity of the whole exercise - and that the mother talking about her own wellbeing rather than the child was not how an emergency department behaves.

Translation is not localisation

The AI layer needed Castilian phrasing, not Spanish in general, plus domain-specific tuning. Clinicians reported abrupt, repetitive replies and misread clinical language.

A flat voice breaks the illusion

The voice system came across as mechanical and emotionally flat. The report rates replacing voices and adding emotional modulation as high impact for relatively low effort.

Feedback belongs after the conversation

Real-time feedback during the interaction broke conversational flow. Participants clearly preferred structured feedback afterwards - which is how the assessment is designed to work anyway.

More scenarios, harder ones

Clinicians asked specifically for psychiatric, mental health and difficult-conversation scenarios. The requests were about scope, not usability - which tells you where the perceived value sits.

Web first, VR where it earns its place

The hybrid model held up: browser access is the right entry point, and VR adds value in supervised settings with smaller groups. Institutional Wi-Fi was troublesome often enough that people fell back on phone hotspots.

Source

MetaMedicsVR, Metaskills

2026-03-31

Pilot evaluation carried out by MetaMedicsVR with partner institutions in Spain, within the VR Health Champions project. Internal report, not a peer-reviewed publication. The pilots at Hospital de Antequera and Hospital Materno-Infantil also drew regional press coverage.

Related evidence

Case study

VR Health Champions in Spain

The project behind these pilots: who took part, what was deployed and what came out of it.

VR Health Champions in Spain
Study

Usability of avatar-based VR training

A separate study, 47 participants, on how usable the immersive VR training felt.

Usability of avatar-based VR training
Industry

Metaskills for healthcare

How the platform is used in clinical and academic settings.

Metaskills for healthcare