Shingles Vaccine Cuts Dementia Risk by 20%, Stanford Study Reveals

Shingles vaccine may help prevent or slow dementia.

An unusual public health policy in Wales may offer the strongest evidence so far that a vaccine can lower dementia risk. In a new study led by Stanford Medicine, researchers examined medical records from older adults in Wales and found that individuals who received the shingles vaccine were 20% less likely to develop dementia over the following seven years compared with those who were not vaccinated.

The findings, published in Nature, reinforce a growing theory that viruses affecting the nervous system may elevate the likelihood of dementia. If confirmed through additional research, the results point to a potential preventive strategy for dementia that is already available.

A companion study, recently published in Cell, indicates that the vaccine might also help people who have already been diagnosed with dementia by slowing how quickly the condition progresses.

Lifelong infection

Shingles is a viral illness known for causing a painful rash, and it is triggered by the same virus responsible for chicken pox, varicella-zoster. After a person contracts chicken pox in childhood, the virus remains inactive in nerve cells for life. When individuals grow older or experience weakened immune function, the virus can reactivate and lead to shingles.

Dementia affects more than 55 million people across the globe, with about 10 million new diagnoses each year. For decades, research has focused heavily on brain changes such as plaques and tangles in people with Alzheimer’s disease, the most common form of dementia. With limited success in preventing or treating the condition, some scientists are now investigating other possible contributors, including specific viral infections.

Earlier studies using medical records have suggested that receiving the shingles vaccine is associated with lower dementia rates, but these studies faced a major limitation: vaccinated individuals often engage in healthier behaviors overall. Factors like diet and physical activity, which influence dementia risk, typically are not documented in medical records and can obscure study results.

“All these associational studies suffer from the basic problem that people who go get vaccinated have different health behaviors than those who don’t,” said Pascal Geldsetzer, MD, PhD, assistant professor of medicine and senior author of the new study. “In general, they’re seen as not being solid enough evidence to make any recommendations on.”

A natural experiment

But two years ago, Geldsetzer recognized a fortuitous “natural experiment” in the rollout of the shingles vaccine in Wales that seemed to sidestep the bias. The vaccine used at that time contained a live-attenuated, or weakened, form of the virus.

The vaccination program, which began Sept. 1, 2013, specified that anyone who was 79 on that date was eligible for the vaccine for one year. (People who were 78 would become eligible the next year for one year, and so on.) People who were 80 or older on Sept. 1, 2013, were out of luck — they would never become eligible for the vaccine.

These rules, designed to ration the limited supply of the vaccine, also meant that the slight difference in age between 79- and 80-year-olds made all the difference in who had access to the vaccine. By comparing people who turned 80 just before Sept. 1, 2013, with people who turned 80 just after, the researchers could isolate the effect of being eligible for the vaccine.

The circumstances, well-documented in the country’s health records, were about as close to a randomized controlled trial as you could get without conducting one, Geldsetzer said.

The researchers looked at the health records of more than 280,000 older adults who were 71 to 88 years old and did not have dementia at the start of the vaccination program. They focused their analysis on those closest to either side of the eligibility threshold — comparing people who turned 80 in the week before to those who turned 80 in the week after.

“We know that if you take a thousand people at random born in one week and a thousand people at random, born a week later, there shouldn’t be anything different about them on average,” Geldsetzer said. “They are similar to each other apart from this tiny difference in age.”

The same proportion of both groups likely would have wanted to get the vaccine, but only half, those not yet 80, were allowed to by the eligibility rules.

“What makes the study so powerful is that it’s essentially like a randomized trial with a control group — those a little bit too old to be eligible for the vaccine — and an intervention group — those just young enough to be eligible,” Geldsetzer said.

Protection against dementia

Over the next seven years, the researchers compared the health outcomes of people of similar age who were eligible and ineligible to receive the vaccine. By factoring in actual vaccination rates — about half of the population who were eligible received the vaccine, compared with almost none of the people who were ineligible — they could derive the effects of receiving the vaccine.

As expected, the vaccine reduced the occurrence over that seven-year period of shingles by about 37% for people who received the vaccine, similar to what had been found in clinical trials of the vaccine. (The live-attenuated vaccine’s effectiveness wanes over time.)

By 2020, one in eight older adults, who were by then 86 and 87, had been diagnosed with dementia. But those who received the shingles vaccine were 20% less likely to develop dementia than the unvaccinated.

“It was a really striking finding,” Geldsetzer said. “This huge protective signal was there, any which way you looked at the data.”

The scientists searched high and low for other variables that might have influenced dementia risk, but found the two groups to be indistinguishable in all characteristics. There was no difference in the level of education between the people who were eligible and ineligible, for example. Those who were eligible were not more likely to get other vaccinations or preventive treatments, nor were they less likely to be diagnosed with other common health conditions, such as diabetes, heart disease, and cancer.

The only difference was the drop in dementia diagnoses.

“Because of the unique way in which the vaccine was rolled out, bias in the analysis is much less likely than would usually be the case,” Geldsetzer said.

Nevertheless, his team analyzed the data in alternate ways — using different age ranges or looking only at deaths attributed to dementia, for example — but the link between vaccination and lower dementia rates remained.

“The signal in our data was so strong, so clear and so persistent,” he said.

Not too late

When the researchers analyzed the health records further, taking advantage of the same natural experiment, they discovered that the vaccine’s benefits extended from the earliest signs of cognitive decline to the latest stages of dementia.

Many cases of dementia are preceded by a period of mild cognitive impairment — characterized by deficits in memory and cognitive skills that do not interfere with independent living, Geldsetzer said.

They found that people who received the vaccine were less likely to be diagnosed with mild cognitive impairment during a nine-year follow-up than those who were unvaccinated.

Even more dramatic, people who received the vaccine after a dementia diagnosis were significantly less likely to die from dementia in the next nine years (as indicated on their death certificates), suggesting that the vaccine could slow the progress of the disease.

Overall, nearly half of the 7,049 Welsh seniors who had dementia at the start of the vaccination program died from dementia during follow-up, but only about 30% of those who received the vaccine died from dementia.

“The most exciting part is that this really suggests the shingles vaccine doesn’t have only preventive, delaying benefits for dementia, but also therapeutic potential for those who already have dementia,” Geldsetzer said.

Stronger response in women

In a further finding, the study showed that protection against dementia was much more pronounced in women than in men. This could be due to sex differences in immune response or in the way dementia develops, Geldsetzer said. Women, on average, have higher antibody responses to vaccination, for example, and shingles is more common in women than in men.

Whether the vaccine protects against dementia by revving up the immune system overall, by specifically reducing reactivations of the virus, or by some other mechanism is still unknown.

Also unknown is whether a newer version of the vaccine, which contains only certain proteins from the virus and is more effective at preventing shingles, may have a similar or even greater impact on dementia.

Geldsetzer hopes the new findings will inspire more funding for this line of research.

“At least investing a subset of our resources into investigating these pathways could lead to breakthroughs in terms of treatment and prevention,” he said.

In the past two years, his team has replicated the Wales findings in health records from other countries, including England, Australia, New Zealand, and Canada, that had similar rollouts of the vaccine. “We just keep seeing this strong protective signal for dementia in dataset after dataset,” he said.

But Geldsetzer has set his sights on a large, randomized controlled trial, which would provide the strongest proof of cause and effect. Participants would be randomly assigned to receive the live-attenuated vaccine or a placebo shot.

“It would be a very simple, pragmatic trial because we have a one-off intervention that we know is safe,” he said.

Geldsetzer is seeking philanthropic funding for the trial as the live-attenuated vaccine is off-patent, but the vaccine type for which he has generated his compelling evidence from natural experiments.

And such a trial might not take long to see results. He pointed to a graph of the Wales data tracking the dementia rates of those who were eligible and ineligible for the vaccine. The two curves began to separate in about a year and a half.

References:

“The effect of shingles vaccination at different stages of the dementia disease course” by Min Xie, Markus Eyting, Christian Bommer, Haroon Ahmed and Pascal Geldsetzer, 2 December 2025, Cell.
DOI: 10.1016/j.cell.2025.11.007

“A natural experiment on the effect of herpes zoster vaccination on dementia” by Markus Eyting, Min Xie, Felix Michalik, Simon Heß, Seunghun Chung and Pascal Geldsetzer, 2 April 2025, Nature.
DOI: 10.1038/s41586-025-08800-x

This study was funded by grants to P.G. by The Phil & Penny Knight Initiative for Brain Resilience at the Wu Tsai Neurosciences Institute, Stanford University (KPI-003); the National Institute on Aging (R01AG084535); National Institute of Allergy and Infectious Diseases (DP2AI171011); and Chan Zuckerberg Biohub-San Francisco.

Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.