Neurotrophic herpesviruses are suspected to contribute to dementia development. Vaccination is a potential strategy to mitigate these effects. Live-attenuated vaccines can trigger broader immune responses with sex-specific effects. Studies on vaccination and dementia often assume no confounding bias. However, unmeasured factors such as motivation and health literacy challenge this assumption, necessitating further verification of findings.
The recent study was published in Nature and aimed to assess the impact of the live-attenuated herpes zoster vaccine (Zostavax) on the incidence of dementia disorders. Zostavax was introduced in Wales on September 1, 2013, using a staggered rollout. Individuals aged 71-78 became eligible over time; those aged 79 formed a catch-up cohort, and those 80+ were ineligible. This analysis compared the vaccinated individuals born after September 2, 1933, with those who were never eligible for vaccination.
The Secure Anonymised Information Linkage (SAIL) databank links primary and secondary care records for approximately 80% of healthcare providers in Wales and 83% of the Welsh population. This study included dementia-free adults born from 1952 to 1942 (n = 296,603) registered in Wales before the 2013 zoster vaccine program. Individuals with preexisting dementia (n = 13,783) and unknown birthdates (n = 279) were excluded. Over the seven-year follow-up period, 37.8% of participants passed away.
This study found that vaccination rates rose from 0.01% in adults who were just one week too old for eligibility to 47.2% in those one week younger (P < 0.001) through SAIL health records. Except for this significant difference in zoster vaccination likelihood, individuals born one week before September 2, 1933, showed no significant differences from those born a week later.
Over the 7-year follow-up period, 14,465/296,324 adults had at least one shingles diagnosis. It was also demonstrated that vaccine eligibility reduced this probability by 1% with a 95% confidence interval of (CI) of 0.2–1.7 (P = 0.010) and an 18.8% relative reduction with a 95% CI of 8.8–28.9. According to Zostavax clinical studies, receiving the vaccine lowered the relative probability by 2.3% with a 95% CI of 0.5–3.9 (P = 0.011) and a 37.2% relative reduction with a 95% CI of 19.7–54.7.
In regression discontinuity analysis, it was estimated that vaccine eligibility led to a 1.3% absolute (95% CI: 0.2–2.7; P = 0.022) and 8.5% relative decline (95% CI: 1.9–15.1) in new dementia diagnoses during the 7-year follow-up period. Adjusting for actual vaccination rates, receiving the zoster vaccine reduced this probability by 3.5% (95% CI: 0.6–7.1; P = 0.019) and 20% relative reduction (95% CI: 6.5–33.4). Vaccination also reduced the dementia diagnosis based on new Alzheimer’s drugs such as rivastigmine, donepezil hydrochloride, memantine hydrochloride, and galantamine. It remained consistent after adjusting for dementia risk score variables.
Women experienced this protective effect more strongly compared to men. The impact of receiving the zoster vaccine on the diagnosis of dementia over 7-years was similar between difference-in-differences instrumental variable analysis (DID-IV) and regression discontinuity analysis (−3.1% [95% CI: −5.8 to −0.4, P = 0.024] versus −3.5% [95% CI: −7.1 to −0.6, P = 0.019]). This DID-IV approach showed a significant effect only on outcomes of postherpetic neuralgia, dementia, and shingles.
Furthermore, all these findings were validated in different populations (England and Wales) through death certificate data and outcomes (death primarily due to dementia) that are closely related to dementia but less dependent on healthcare systems’ prompt dementia diagnosis.
This study’s limitations included those delays in dementia diagnosis, inability to assess vaccine effects in remaining age groups, pandemic related diagnosis delays, 8-year follow-up limit and applicable only to live-attenuated zoster vaccine.
In conclusion, this study’s findings highlight robust evidence that zoster vaccination may help prevent or delay dementia with reduced susceptibility to confounding and bias compared to previous clinical studies.
Reference: Eyting M, Xie M, Michalik F, Heß S, Chung S, Geldsetzer P. A natural experiment on the effect of herpes zoster vaccination on dementia. Nature. 2025. doi:10.1038/s41586-025-08800-x
