Benzodiazepine use shows only a modest link to dementia
Dementia incidence was 382.4 per 100,000 person-years in benzodiazepine users versus 280.7 in controls, but the link weakened after adjusting for underlying illnesses.
In an observational matched cohort study, researchers analysed health claims from 132,805 benzodiazepine users and 132,805 age- and sex-matched controls in South Korea. They tracked new cases of dementia, including Alzheimer's disease, defined by three or more diagnostic claims within one year.
Dementia incidence was 382.4 per 100,000 person-years among benzodiazepine users compared with 280.7 in controls. Benzodiazepine users had higher rates of other illnesses, including anxiety disorders. After full adjustment for comorbidities, the hazard ratio for overall dementia attenuated from 1.43 to 1.15. Benzodiazepine use was not associated with progression from dementia to severe dementia, with no dose- or duration-response observed.
Why it matters
Dementia is a major disease linked to aging, but whether benzodiazepines contribute to its development has remained controversial. The findings bear on how co-occurring illnesses may shape apparent links between medication use and dementia risk.
Caveats
Because this was an observational study using health claims data, dementia diagnoses and severity were identified through billing codes rather than clinical examinations. Residual confounding from unmeasured factors may also remain.
The paper
College of Medicine
Journal of Alzheimer's Disease, 9 Oct 2026
