The question almost always arrives the same way. Someone with advanced dementia who no longer speaks, or someone in their last days. The family asks whether "that goggles thing" might do some good. And the team wants an honest answer.
This is the area where hope is easiest to sell, which is why it demands the most care to write about.
What changes when there are no words
Three things at once, and it is the accumulation that makes this hard:
- There is no verbal consent. The usual conversation — explain, ask, hear yes or no — stops being available.
- There is no account. Nobody can say whether it was good, strange or frightening. The person's experience stays inaccessible to us.
- Observation is fallible. We read an expression, a loosening of the shoulders, a smile. We may be reading it right. We may be reading what we want to see, and the risk is greatest precisely when we badly wanted it to go well.
What the evidence says, and it is little
Frankly: research on virtual reality in advanced dementia and at end of life is thin. The bulk of the dementia literature covers mild to moderate stages, with wellbeing, mood and reminiscence outcomes — and even there with small studies.
For people who no longer communicate verbally, what exists is largely observational and does not support claims of benefit. Anyone telling you there is evidence that this improves end-of-life quality is going well beyond what exists. We gathered the work in the study library, and the palliative and dementia themes are there with their counts on show, including when they are few.
What stands in for consent
With no informed consent in the usual sense, what exists is a combination worth naming:
- Authorisation from whoever represents the person — family, legal representative — informed honestly, including the part about not knowing;
- Assent and dissent: what the person expresses in the moment, without words. Turning the head away, reaching for the headset, frowning, tensing. This counts as refusal and is not negotiated;
- What was known about the person before. Did they love the sea? Were they claustrophobic? Did they refuse devices? Biography is the best available source when the person can no longer answer, and it usually sits with the family.
We have written about consent generally; the difference here is that dissent becomes the primary signal rather than an exception.
The question that decides
There is a question that separates caring from staging, and it is uncomfortable:
Whose benefit are we looking for?
Sometimes the honest answer is: the family's. Seeing your mother smile at the sight of her home town is a good and legitimate thing for the family, and may be the only good thing that week. But it is different from saying it did your mother good, and the difference should be said out loud when deciding.
We are not saying the family's benefit does not count. It does, and in palliative care the unit of care includes them. We are saying it should be called by its name, rather than dressed up as clinical benefit.
And there is a clear limit on the other side: if the person shows discomfort and the session continues because the family wants to watch, it has stopped being care.
If you go ahead, what makes a difference
- Short. Much shorter than in other contexts. Minutes.
- Content from the biography, not pretty content. Their place, not a generic beach.
- Someone trusted alongside, talking, holding a hand. The familiar voice is part of the session.
- Position: at end of life people are rarely sitting up. Lying-down mode exists so the scene does not appear tilted to someone reclined or in bed.
- Stop at the first sign, and the threshold here is lower than in any other context.
- Log what was observed in the words of what was seen — "frowned", "breathing settled" — not in conclusions like "enjoyed it".
Where we stop
Let us be explicit, because this is an area where the temptation to say more is strong:
- RVer does not assess comfort, pain or emotion. There is no measurement of emotional state. What exists is the record of what the team observed and wrote.
- We have no evidence of benefit in this population, and we will not claim any.
- We do not know what the person is experiencing. Nobody does.
What we can state is what the system allows: a scenario chosen by the team, short, stoppable remotely, with family present if that makes sense. Whether it makes sense is a clinical decision, belonging to the team and to whoever represents the person, taken case by case.
In short
- Without words, you lose consent, account and certainty — all three at once.
- The evidence in this population is thin and supports no promises.
- Dissent is refusal, and it is the primary signal.
- Ask whose benefit it is, and call it by its name.
- If the person shows discomfort and it continues, it has stopped being care.