Dementia remains a major public health concern, affecting millions of people worldwide and increasing with population ageing. Because effective treatments remain limited, researchers continue to explore factors that may influence dementia risk over long periods of life. One area receiving growing attention is the potential relationship between common adult vaccinations and cognitive outcomes. A new systematic review and meta-analysis published in Age and Ageing examines this question by compiling observational evidence from very large population cohorts.
The review included 21 studies involving 104,031,186 adults aged 50 or older. These studies came from North America, Europe, and Asia and evaluated routine vaccines such as shingles (herpes zoster), influenza, pneumococcal, tetanus-diphtheria-pertussis (Tdap), as well as historical exposure to vaccines like poliomyelitis. Each study compared dementia incidence between adults who had received one or more vaccinations and those who had not, monitoring participants over varying follow-up periods, often several years. The studies were all observational, meaning they reported associations rather than cause-and-effect relationships.
The strongest statistical associations in the review appeared for shingles vaccination. Across eight cohorts including nearly one million vaccinated individuals, shingles vaccination was associated with a lower likelihood of developing dementia overall (risk ratio 0.76) and Alzheimer's disease specifically (risk ratio 0.53). These values reflect relative risk, meaning vaccinated individuals in these datasets developed dementia at lower rates compared with unvaccinated participants during the follow-up period. The forest plots shown on page 5 of the report illustrate these data visually, with most cohorts trending below the relative risk value of 1.0.
Influenza vaccination was also associated with modestly lower dementia incidence in ten cohorts totaling over 2.5 million vaccinated adults. In these datasets, individuals who received influenza vaccination showed a 13% lower risk of dementia compared to non-vaccinated peers. Associations were also noted for certain dementia subtypes, such as vascular dementia and other non-Alzheimer's forms. However, the authors reported substantial statistical heterogeneity across influenza studies, meaning results varied considerably between cohorts and should be interpreted with caution. Prediction intervals - which estimate the range of results expected in future studies - crossed the null value, indicating variability in possible outcomes.
Pneumococcal vaccination appeared in a smaller number of studies but suggested an association with lower Alzheimer's disease risk. In three cohorts including 276,413 vaccinated individuals, pneumococcal vaccination correlated with a 36% lower risk of Alzheimer's disease. Associations with "any dementia" were weaker and reached only marginal statistical significance. Tdap vaccination showed a pattern of similar associations, with several cohorts showing reduced dementia rates among vaccinated individuals. These findings were consistent across data sources but also showed high heterogeneity, suggesting population differences, diagnostic variation, and confounding factors may contribute to the observed patterns.
The review also included early research on COVID-19 vaccination and dementia outcomes. However, these studies had extremely short follow-up periods - sometimes only three months - due to the vaccine's recent introduction. Because dementia develops over many years, the authors emphasized that no meaningful conclusions can be drawn from such limited timelines. The report notes that some correlations observed in early COVID-19 vaccine studies likely reflect reverse causation or residual confounding, rather than effects of vaccination itself. In addition, underlying COVID-19 infection severity has been shown in other studies to influence cognitive symptoms, complicating interpretation.
Across all vaccine types, the review proposes several possible explanations for why vaccinated adults might show lower dementia rates, while acknowledging these remain hypotheses rather than demonstrated mechanisms. One possibility is that preventing infections reduces episodes of systemic or neuroinflammation, which are known contributors to cognitive decline. Several pages of the review (pages 2 and 6) outline links between viral and bacterial infections - including influenza, herpesviruses, and pneumonia - and inflammation-related pathways involved in neurodegeneration. Another possibility is that adults who keep up with vaccinations may be more engaged with health services or overall preventive care. This "healthy vaccinee effect" could make vaccinated individuals appear at lower risk even if the vaccines themselves are not directly responsible for the observed differences.
The authors emphasize several limitations. First, because all included studies were observational and many were retrospective, they cannot prove that vaccines reduce dementia risk. Differences in medical record systems, diagnostic codes, demographic factors, and confounder adjustment across studies introduce substantial variability. Study heterogeneity was especially high for influenza and shingles analyses. Furthermore, vaccination status may correlate with socioeconomic factors, healthcare access, general health behaviors, or cognitive reserve - differences that could influence dementia risk independently of vaccination. Although most studies adjusted for many confounders and some used propensity score matching, unmeasured factors cannot be fully ruled out.
From the perspective of Seven Reflections' Dimensional Systems Architecture (DSA), the findings can be interpreted as a systems-level interaction rather than a direct causal effect. DSA conceptualizes cognitive function as emerging from multiple interdependent fields - biological, environmental, behavioral, and systemic. In this framework, vaccination could be understood as one of many factors that modulate the biological field by reducing exposure to destabilizing events, such as severe infections or prolonged inflammatory states. Lower volatility in one field may support stability in adjacent cognitive fields over time. However, DSA would treat vaccination as a correlating structural factor, not a deterministic driver. The associations observed in the meta-analysis are therefore consistent with a broader systems perspective in which long-term cognitive outcomes reflect the cumulative balance of multiple interacting influences.
Overall, the review suggests that vaccination status and dementia risk are statistically associated in multiple large population datasets. These associations do not establish that vaccines prevent dementia, but they highlight an area of research that may inform future studies on ageing, infection, inflammation, and cognitive health. The authors call for prospective trials, standardized diagnostic approaches, and longer follow-up periods to better understand the nature of these relationships.
