# Inconsistency in the Association Between Proton Pump Inhibitor Use and Dementia Risk: An Updated Meta-Analysis

## TL;DR

This study finds mixed evidence linking proton pump inhibitors to dementia risk, with significant differences in results across age groups and regions.

## Contribution

An updated meta-analysis of PPI use and dementia risk with over 6 million participants, highlighting subgroup-specific associations.

## Key findings

- Overall PPI use not significantly associated with dementia risk (RR = 1.14, 95% CI 0.98–1.33).
- Significant increased risk found in individuals aged ≥65 years (RR = 1.21, 95% CI 1.01–1.46).
- Higher risk observed in studies from Asia and Europe (RR = 1.31 and 1.32 respectively).

## Abstract

Background: Proton pump inhibitors (PPIs) are commonly used to manage acid-related gastrointestinal conditions. Nevertheless, growing attention has been paid to their long-term safety, especially their possible link to dementia and Alzheimer’s disease (AD). Prior research has yielded inconsistent findings, underscoring the need for a comprehensive and current evaluation. Methods: A systematic search was conducted across PubMed, Embase (Ovid), and the Cochrane Library to identify relevant publications up to May 28, 2025, without language restrictions. Two investigators independently extracted study information and evaluated methodological quality as well as potential sources of bias. Eligible studies were observational in design and investigated the association between proton pump inhibitor (PPI) exposure and the risk of developing dementia compared with non-use. For the quantitative synthesis, pooled risk ratios (RRs) and corresponding confidence intervals were generated using a random-effects approach. Study Results: Eighteen studies, encompassing more than 6.3 million participants, met the inclusion criteria. The pooled estimate showed no statistically significant association between PPI use and overall dementia risk (RR = 1.14, 95% CI 0.98–1.33; I2 = 99%). However, significant heterogeneity and variable risk of bias—particularly due to confounding, exposure misclassification, and immortal time bias—limit certainty in these findings. Subgroup analyses revealed significantly elevated risks among individuals aged ≥65 years (RR = 1.21, 95% CI 1.01–1.46) and in studies from Asia (RR = 1.31, 95% CI 1.12–1.52) and Europe (RR = 1.32, 95% CI 1.10–1.59), suggesting possible population- or context-specific vulnerability. Conclusions: Our findings reveal a lack of consistent evidence supporting a link between PPI use and dementia risk, primarily due to significant heterogeneity among existing studies. While no robust overall association was demonstrated, significant subgroup signals in older adults and specific regions suggest that clinical uncertainty remains. Rather than indicating a direct causal risk, these results underscore the importance of prescribing stewardship. Clinicians should focus on appropriate prescribing, ensuring long-term PPI therapy is reserved for patients with a clear therapeutic justification and utilized for the shortest effective duration.