Proton pump inhibitors (PPIs), commonly prescribed for gastrointestinal disorders, have been investigated for their association with the risk of dementia. We aimed to assess the association between PPI use and dementia risk in older adults and examine the influence of methodological choices on this association. We conducted a meta‐analysis of longitudinal observational studies, including subgroup analyses to address time‐related biases and comparator selection. Relevant studies were systematically searched in the MEDLINE, Embase, and CENTRAL databases from their inception to January 25, 2026. Twenty‐three studies of older adults without dementia at baseline, reporting adjusted estimates for PPI users versus a comparator group, met the inclusion criteria. The pooled hazard ratio (HR) indicated a modest association between PPI use and dementia (HR = 1.18, 95% confidence interval [CI] 1.02–1.37). Dementia risk did not differ significantly between PPI users and users of histamine‐2 receptor antagonists (H2RAs) (HR = 1.04, 95% CI 0.89–1.20). In a post hoc analysis, H2RA use itself was associated with dementia (HR = 1.44, 95% CI 1.07–1.93), whereas excluding H2RA‐treated patients strengthened the PPI estimate (HR = 1.75, 95% CI 1.38–2.22). Estimates also differed according to how time‐related bias was handled: lag time, HR = 1.25, 95% CI 1.07–1.46; time‐varying exposure, HR = 1.38, 95% CI 1.16–1.64. PPI use was modestly associated with incident dementia. However, the strength and specificity depended substantially on comparator selection and handling of time‐related bias. Well‐designed, long‐term studies using active comparators are needed to clarify whether this effect is PPI‐specific.
