Does the Shingles Vaccine Reduce Dementia Risk? What the Research Shows
August 14, 2026
It’s the question we’ve been asked most often this year, usually after someone has seen a headline: does the shingles vaccine protect against dementia? The honest answer is that something real appears to be going on, several large studies point the same way, and nobody has yet proved cause and effect. This page sets out what the research has actually found, and what it hasn’t, without the headline gloss in either direction.
We should say plainly at the top: the shingles vaccine is not a dementia treatment or a dementia preventative. It isn’t licensed for that, no health authority recommends it for that, and anyone offering it to you on that basis is ahead of the evidence. What follows is a summary of an interesting and unusually strong body of research, offered as information rather than as a reason to do anything.
The short version
- Several large studies have found lower rates of dementia among people who received a shingles vaccine.
- The strongest, a natural experiment in Wales, found a 20% relative reduction over seven years.
- That study looked at the old live vaccine. Later work on the current vaccine points the same way.
- One recent finding suggests the effect may come from the vaccine’s adjuvant rather than from preventing shingles.
- No randomised trial has been done. Cause and effect is not established, and the vaccine is not licensed for this.
Where the question came from
Researchers have wondered for years whether viruses that live permanently in the nervous system might contribute to dementia. Varicella-zoster, the virus behind chickenpox and shingles, is an obvious candidate, because it sits dormant in nerve tissue for decades and causes inflammation when it reactivates. Testing that idea rigorously is difficult, because people who choose to be vaccinated tend to be healthier, better off and more engaged with healthcare than people who don’t, and all of those things independently affect dementia risk.
In 2025, a team found a way around the problem. When Wales introduced its shingles vaccination programme in September 2013, eligibility was drawn by date of birth: anyone born on or after 2 September 1933 could have the vaccine, and anyone born a week earlier never became eligible. Vaccination rates jumped from almost nothing to 47.2% across that single week. Two groups of people, effectively identical in every other respect, separated only by a birthday.
Over the following seven years, the vaccinated group had 3.5 percentage points fewer new dementia diagnoses, a relative reduction of about 20%. The study covered 282,541 adults and was published in Nature. Because the dividing line was arbitrary, this comes closer to a randomised trial than observational research usually manages.
The Wales study is unusual because the birthday cut-off did the work randomisation normally does. It removes most of the “healthier people get vaccinated” objection that weakens other studies on this question.
What came next
The obvious limitation of the Wales work is that it studied Zostavax, the older live vaccine, which the UK replaced in September 2023. The vaccine in use today is a different product built on different technology, so the finding didn’t automatically carry across. Two later studies looked at that question.
A 2024 study in Nature Medicine compared people who received the current recombinant vaccine against people who received the older live one, using the switchover in the United States as its dividing line. Across 207,674 matched patients, the recombinant group spent 17% longer free of a dementia diagnosis, which worked out at about 164 additional days among those eventually diagnosed.
A 2026 study in Nature Communications, following 65,800 vaccinated adults at Kaiser Permanente Southern California, found a 51% lower rate of dementia against unvaccinated people. That figure needs treating with care, and the researchers did treat it with care: when they compared against people who had received a different vaccine instead, which controls for the healthy-vaccinee problem, the reduction fell to 27%. The second number is the more trustworthy one.
Across all three, the effect was consistently larger in women than men, which nobody has yet explained.
The twist nobody expected
Here’s where it gets interesting, and where the simple story falls apart. A 2025 study in npj Vaccines looked at 436,788 adults and compared two quite different vaccines that happen to share the same adjuvant, the ingredient added to provoke a stronger immune response. One was the shingles vaccine. The other was an RSV vaccine, which has nothing to do with varicella-zoster.
Both were associated with less dementia. Stranger still, people who had received both showed no additional benefit over people who had received just one. If preventing shingles were the mechanism, that pattern makes little sense. If the adjuvant is doing something to the immune system more broadly, it makes rather more.
That finding is early and could yet be explained another way. But it raises the possibility that this story isn’t really about shingles at all, which would be a genuine surprise and would change what any of it means in practice.
We don’t yet know whether the association comes from preventing shingles, from the adjuvant’s effect on the immune system, or from something else entirely. Until that’s settled, “the shingles vaccine protects your brain” is a headline rather than a finding.
What this does not mean
Four things are worth stating clearly, because the headlines rarely include them.
It doesn’t mean the vaccine prevents dementia. Every study above is observational. Observational research shows that two things occur together; it cannot show that one causes the other. The Wales design gets closer than most, but its own authors called for confirmation in other populations before drawing conclusions.
It isn’t licensed for this. The shingles vaccine is authorised to prevent shingles and post-herpetic neuralgia. Dementia is not among its approved uses, and no UK health authority recommends vaccination for that purpose.
The size of any effect is uncertain. Published figures range from 17% to 51% depending on design and comparison group, and the more carefully controlled the comparison, the smaller the number tends to get. That pattern usually means the true effect, if it exists, sits at the lower end.
Dementia has many causes. Nothing here changes what’s already known about blood pressure, hearing loss, physical activity, smoking, alcohol, social contact and head injury, all of which have far better evidence behind them than any vaccine.
What happens next
The field’s own view is that the observational evidence has gone as far as it usefully can. Writing in Nature Medicine in 2026, one of the Wales researchers argued that “we have never had as robust a body of evidence from observational data on an intervention for dementia” as exists here, and that the necessary next step is a large randomised trial. A US National Institutes of Health workshop and an international expert consensus on repurposing existing drugs for Alzheimer’s have both reached the same conclusion.
That trial has not yet been run. Until it is, this remains an open and promising question rather than a settled one, and we’d rather tell you that than let a headline do the talking.
What we say when patients ask
People do ask us about this, usually having seen a piece in the papers, and often about a parent rather than themselves. What we tell them is roughly what’s on this page.
If you’re thinking about the shingles vaccine, think about it for what it’s actually proven to do. Shingles affects around one person in four in their lifetime, has no cure, and can leave nerve pain lasting months. That case stands entirely on its own, and it’s set out in our guide to the shingles vaccine, along with who it suits and who should avoid it. If the dementia research eventually holds up, it will be a welcome bonus on top of a decision that already made sense.
And if dementia is what’s really on your mind, the conversation worth having is with your GP, about the risk factors that have solid evidence behind them today.
Questions we’re asked about this research
Does the shingles vaccine prevent dementia?
Not as far as anyone can currently say. Several large studies have found lower dementia rates among vaccinated people, but none can show cause and effect, and the vaccine is not licensed or recommended for this purpose.
How strong is the evidence?
Unusually strong for observational research, and consistent across different countries, datasets and study designs. That still isn’t the same as proof, which is why researchers are calling for a randomised trial.
Was the Wales study about the vaccine used today?
No. It looked at Zostavax, the live vaccine the UK used until September 2023. Later studies on the current recombinant vaccine point in a similar direction, but they’re observational and shorter.
Why would a shingles vaccine affect the brain at all?
Two theories. One is that preventing viral reactivation reduces inflammation in nerve tissue. The other, raised by the adjuvant research, is that the vaccine’s immune-boosting ingredient has broader effects. Neither is established.
Should I have the vaccine because of this?
We wouldn’t suggest so. Decide based on shingles, which is what the vaccine is proven to prevent. Anything else is speculative at this stage.
My parent already has dementia. Would it help?
There’s no evidence that vaccination helps once dementia has developed, and it should not be considered a treatment. Their care team is the right place for that conversation.
Why do the studies show bigger effects in women?
Nobody knows. It’s a consistent finding across several studies and remains unexplained, which is one more reason to treat the whole picture as unfinished.
This article is general information, not personal medical advice, and it is not a recommendation to be vaccinated for any purpose other than protection against shingles. The shingles vaccine is not licensed or recommended for the prevention of dementia. If you have concerns about dementia or your memory, speak to your GP.
