A systematic review of 14 studies identified key medications linked to increased and decreased dementia risk.
Researchers evaluated associations between prescribed medications and dementia risk using a data-driven approach. The analysis was comprised of administrative and electronic health records data from over 130 million participants and more than 1 million dementia cases. The review aimed to identify medication classes with potential protective or adverse effects on dementia incidence, supporting drug repurposing efforts.
The review, published in Alzheimer’s & Dementia: Translational Research & Clinical Interventions, followed Preferred Reporting Items for Systematic reviews and Meta-Analyses guidelines, with literature searches conducted in MEDLINE, Embase, and PsycINFO through August 2023. Studies included in the review used data-driven methods rather than hypothesis-driven approaches to evaluate medication-dementia associations. Eligible studies included retrospective cohort and case-control designs analyzing large-scale prescription records linked to dementia diagnoses based on International Classification of Diseases codes. Studies using predefined hypotheses, reviews, and conference abstracts were excluded. Medication exposure was determined through prescription claims, though adherence was not directly assessed.
Results
The findings indicated that antimicrobials, anti-inflammatories, and vaccines were associated with reduced dementia risk. Specifically, antibiotics such as amoxicillin, azithromycin, and doxycycline were linked to lower dementia incidence across multiple studies. Four vaccines—hepatitis A, typhoid, hepatitis A and typhoid combination, and diphtheria—were associated with a reduced risk of dementia, with hazard ratios ranging from 0.68 to 0.92. Antihypertensives and lipid-lowering agents, including angiotensin II receptor blockers and statins, were associated with lower risk in some studies, though results varied.
Conversely, antipsychotics, antidepressants, and diabetes medications were frequently associated with increased dementia risk. In one medication-wide association study, 217 of 744 drugs analyzed were linked to higher risk, including antipsychotics such as quetiapine (hazard ratio range = 3.11–4.03) and antidepressants including mirtazapine, sertraline, and trazodone. Reverse causation remains a concern, as these medications may be prescribed for early dementia symptoms. A sensitivity analysis found that excluding prescriptions within 5 to 10 years before diagnosis did not eliminate associations with antidepressants and antipsychotics, warranting further investigation.
Led by Benjamin R. Underwood, PhD, of the Department of Psychiatry and Cambridgeshire and Peterborough NHS Foundation Trust, Windsor Unit, Fulbourn Hospital Cambridge, University of Cambridge, the research team found conflicting evidence regarding the effects of antihypertensives and lipid-lowering agents on dementia risk, with some studies suggesting a protective effect while others showed no significant association.
Machine-learning models across multiple studies identified antidepressants, psychotropics, and cardiovascular drugs as potential dementia risk predictors, though external validation remains limited. Model performance, measured by the area under the receiver operating characteristic curve, ranged from 0.64 to 0.94.
Future research integrating genetic and biomarker data may help refine causal inference and assess the potential for drug repurposing strategies.
Full disclosures can be found in the published study.
