Heterogeneity in the association between antihypertensive medications and Alzheimer's disease and related dementias risk: A multidimensional machine learning analysis to inform drug repurposing.

BackgroundAlzheimer's disease and related dementias (ADRD) place an immense burden on patients, families, and health systems in the United States. Hypertension is highly prevalent in ADRD patients, and antihypertensive medications (AHMs) may confer neuroprotective benefits. However, evidence of association between AHM use and ADRD risk remains limited, especially by drug classes and across population subgroups.ObjectiveThis study aimed to evaluate the association between antihypertensive medications (AHMs) use and ADRD risk, and assess heterogeneity by drug classes and demographics.MethodsWe analyzed 2015-2019 Medicare Current Beneficiary Survey data linked with Medicare claims. ADRD was identified by International Classification of Diseases (ICD) codes, anti-dementia drug use, and survey responses. This study included AHMs from 8 classes. Double/debiased machine learning was employed, adjusting for factors from environmental, sociocultural, behavioral, and biological domains.ResultsAmong 144,788,079 weighted Medicare beneficiaries, 75.44% used AHMs. Overall, AHM use was associated with a 5% reduction in ADRD risk (Effect: 0.95; 95% CI: 0.94-0.95). Heterogeneity in protective effects was observed for most AHM classes, except vasodilators and centrally acting alpha-adrenergic agonists, which were linked to increased ADRD risk (p < 0.001). Disparities were found across age and sex groups, with greater effects in older adults, and were observed for all racial and ethnic groups (p < 0.05) except Hispanics.ConclusionsThis study found that AHM use was associated with reduced ADRD risk, with notable heterogeneity across subgroups, highlighting the importance of integrating hypertension care into cognitive health to develop equitable and personalized strategies.