Concurrent Gabapentin and Dihydropyridine Calcium Channel Blocker Use and Risk of Incident Dementia: An Active-Comparator Cohort Study

Abstract Importance Gabapentin and dihydropyridine calcium channel blockers (DHP CCBs) are commonly prescribed together in older adults: gabapentin reaches more than 15.5 million U.S. patients annually, and DHP CCBs such as amlodipine are a first-line antihypertensive class. Both classes act on neuronal calcium signaling, raising the question of whether their combined use is associated with greater cognitive risk than gabapentin alone. Objective To examine whether concurrent CCB use, and DHP CCB use in particular, modifies the association between gabapentin initiation and incident dementia in adults with hypertension. Design, Setting, and Participants Active-comparator new-user cohort study within the Rutgers Clinical Research Data Warehouse (2018–2024), encompassing 405 clinical sites in northern New Jersey. Participants were adults aged 40 years or older with hypertension who initiated gabapentin or pregabalin. Data were analyzed from January through December 2025. Exposures Gabapentin vs pregabalin initiation, with pre-specified stratification by baseline CCB exposure and CCB subtype (DHP vs non-DHP). Main Outcomes and Measures Incident dementia (ICD-10 F00–F03, G30). The pre-specified primary analysis used inverse probability of treatment weighting with time-varying CCB exposure; secondary analyses examined CCB subtype specificity, dementia subtypes, falsification outcomes, and competing risk of death. Results Among 33,791 patients (mean [SD] age, 69.9 [11.5] years; 57.1% female; 502 dementia events; median follow-up, 1.22 years), gabapentin initiation was associated with higher dementia incidence among patients also taking a CCB than among those not taking a CCB (time-varying HR, 1.52 [95% CI, 1.23–1.89]; baseline-stratified HR, 2.22 [95% CI, 1.42–3.47]). The signal was directionally larger with DHP CCBs. Median time to dementia diagnosis was 240 days; subtype estimates were hypothesis-generating. Conclusions and Relevance In this active-comparator cohort of older adults with hypertension, concurrent CCB use — particularly with DHP agents — was associated with substantially higher gabapentin-related dementia incidence than gabapentin without CCB exposure. The short event latency suggests the signal may partly reflect drug-induced cognitive impairment, which may improve with deprescribing — a clinically actionable possibility. CCB-subtype and dementiasubtype estimates are hypothesis-generating; confirmatory replication in Medicare fee-forservice claims and the NIH All of Us Research Program is underway. Key Points Question In older adults with hypertension, is concurrent calcium channel blocker (CCB) use associated with higher gabapentin-related dementia incidence than gabapentin alone? Findings In an active-comparator cohort of 33,791 adults, gabapentin initiation (vs pregabalin) was associated with higher dementia incidence among patients also taking a CCB than among those not taking a CCB (time-varying HR, 1.52 [95% CI, 1.23–1.89; P = .0003]; baselinestratified HR, 2.22 [95% CI, 1.42–3.47]; interaction P = .004). The signal was directionally larger among DHP CCB users (HR, 3.20) and concentrated in Alzheimer disease (HR, 4.53) and unspecified dementia (HR, 2.29), with a null finding for vascular dementia (HR, 0.81). Meaning Concurrent gabapentin and DHP CCB use identifies a potentially modifiable cognitive safety signal in older adults; the short event latency raises the possibility that part of the affected cognition may be at least partially reversible after deprescribing. CCB-subtype and dementiasubtype findings are hypothesis-generating and require replication in larger cohorts; independent replication efforts in broader real-world populations are ongoing.