Designing accessible VR for hospitals

Most VR design assumes the user has hands that work, eyes that focus quickly, and patience for tutorial menus. Therapeutic VR in a clinical setting often has none of those.

The headset is half the design

The patient sees what the headset shows them. They don't see the menu we built three months ago, the controller mapping, the developer's mental model. They see whatever appears, and they have to make sense of it inside 10 seconds — because if it confuses them in the first 10, they take the headset off.

We design every scene to be intelligible without instruction. No tutorials. No controller diagrams. The first thing the patient sees is the destination — a calm beach, a slow forest, a quiet snowy landscape — and the only interaction they need to do is be there.

The clinician is the operator

The patient doesn't control the app. The nurse, the physiotherapist, the oncology technician does — from a tablet, in real time. That changes the whole UI surface area. The clinician's panel needs glanceable status (signal, battery, session timer, abort) in 1-second comprehension. No menus, no settings, no choices that aren't immediately reversible.

Accessibility as core, not bolt-on

Aniridia, low contrast vision, motor tremor, hearing loss — every one of these is over-represented in the patient population we serve. So: every voice line subtitled. Every soundscape level-locked under 70dB. Every interaction either a long-press or a head-tilt, never a finger-precision input. Every motion bounded to avoid simulator sickness on first exposure.

None of this slows the user who doesn't need it. All of it saves the session for the user who does.

The thing nobody says

The most accessible VR experience is one a stressed clinician can run without thinking. The patient experience is mostly the byproduct of getting the operator experience right.