Fourteen dementia risk factors: where technology fits, and where it does not

In July 2024, the Lancet Commission on dementia prevention, intervention and care published its updated report. There are now 14 modifiable risk factors across the life course, and the figure that went around is this: up to 45% of future dementia cases could be prevented if all 14 were eliminated.

It is a number that appears in many decks, technology suppliers' included. It is worth understanding what it says.

The fourteen

To the previous twelve (low education, hearing loss, hypertension, smoking, obesity, depression, physical inactivity, diabetes, excessive alcohol, traumatic brain injury, air pollution and social isolation), the 2024 update added two:

• high LDL cholesterol in midlife, contributing around 7% of attributable risk;
• untreated vision loss in later life, around 2%.

What “45% preventable” means, and what it does not

This is the part lost in the citation, and the part that matters.

It is a population estimate, not an individual promise. It says what would happen to incidence in a country if those factors all disappeared. It does not say that a specific person correcting one factor reduces their risk by any known percentage.

It is a limit case. “If all 14 were eliminated” is not a plan; no health system on earth eliminates air pollution and low education.

Attributable risk is not demonstrated causality for each factor, as the Commission itself notes. It is the best available synthesis, and it remains a synthesis.

And most factors are midlife or earlier. Education, hypertension, cholesterol, tobacco: by the time someone enters a care home, the window on several of them closed decades ago. Serious prevention happens long before that door.

What a care home can actually act on

Six things, none of them technological:

• Hearing loss: check the aid exists, has a battery, and is worn. Probably the best effort-to-effect ratio on the whole list.
• Vision loss: the 2024 addition. Un-operated cataracts and wrong prescriptions are common and treatable.
• Hypertension and diabetes: already being managed; the question is whether they are controlled.
• Physical inactivity: what exists, how often, and for the people who never attend group classes.
• Depression: under-diagnosed, and mistaken for “he's quieter lately”.
• Social isolation: the only one where the conversation below applies.

If a care home wants to act on dementia risk from this list, this is where it starts, not with a supplier catalogue. Saying so runs against our immediate commercial interest and is the only honest reading of the report.

Social isolation, and the boundary

Social isolation has been on the list since before 2024, and it is real: the Portuguese data we cited in undiagnosed cognitive decline in care homes describe an institutionalised population with very little outside contact: three in four people with dementia in care homes never go outside, and more than a quarter never receive visitors.

Precision matters here, because this is where everyone stretches:

Reducing isolation with technology is not the same as reducing dementia risk. The factor the Commission identifies is isolation over years, not loneliness on one afternoon. No study shows that virtual reality sessions change dementia incidence, and we know of none that has even tried to measure it.

Human contact is not substitutable. What the evidence associates with lower risk is real, sustained company. A video call with family is human contact with a medium in between; an immersive scene, however good, is not.

What remains is modest and true: a session can give someone something to talk about, pleasure on an afternoon, company beside them, and a reason to leave their room. That is worth having on its own: it does not need to be sold as dementia prevention, and should not be. That is how RVer fits into the day of a care home.

The sentence not to let pass

If a supplier (us or anyone else) links a product to the Commission's 45%, the question is simple: in which trial did this product reduce dementia incidence?

The honest answer today, for every virtual reality product on the market, is none. What exists is literature on wellbeing, reminiscence and stimulation, which we describe in dementia and Alzheimer's and cognitive stimulation, and it measures something else (the library gathers it under quality of life).

The RVer platform's base product is a Class I medical device registered in EUDAMED and with INFARMED (Portugal). Sessions are run by the institution's own professionals, complement existing care, and do not prevent, treat or delay dementia.