Background: Observational studies suggesting that immunizations strongly decrease the risk of dementia had several methodological limitations. We assessed whether common vaccines are associated with the risk of dementia.
Methods: We assembled a population-based cohort of dementia-free individuals aged ≥50 years in the United Kingdom's Clinical Practice Research Datalink between 1988 and 2018. Using a nested case-control approach, we matched each patient with dementia with 4 controls. Conditional logistic regression yielded confounder-adjusted odds ratios (ORs) with 95% confidence intervals (CIs) of dementia associated with common vaccines >2 years before the index date compared with no exposure during the study period. Moreover, we applied a 10-year lag period and used active comparators (participation in breast or prostate cancer screening) to account for detection bias.
Results: Common vaccines were associated with an increased risk of dementia (OR, 1.38 [95% CI, 1.36-1.40]), compared with no exposure. Applying a 10-year lag period (OR, 1.20 [95% CI, 1.18-1.23]) and comparing versus prostate cancer screening (1.19 [ 1.11-1.27]) but not breast cancer screening (1.37 [1.30-1.45]) attenuated the risk increase.
Conclusions: Common vaccines were not associated with a decreased risk of dementia. Unmeasured confounding and detection bias likely accounted for the observed increased risk.
Keywords: epidemiology; immunization; public health; real-world evidence.
© The Author(s) 2022. Published by Oxford University Press on behalf of Infectious Diseases Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.