Association of Antihypertensive Drugs and Cognitive Function

Background: Neuroprotective effects of angiotensin-II stimulating drugs (ASDs) vs angiotensin-II inhibiting drugs (AIDs) have been inconclusive, largely due to a lack of cognitive assessments and consideration of confounders such as tobacco and alcohol use, and health, mental, and physical status in previous studies. This dissertation was aimed at identifying individual-level factors associated with the use of ASDs and mixed use of ASDs and AIDs, compared with AID use only in Paper 1. Paper 2 combines Aims 2 and 3, aimed at examining the association of ASDs use or mixed use vs. AIDs use with composite memory score and predicted dementia probability. Methods: In paper 1, the study population included Health and Retirement Study (HRS) participants aged ≥65 years who had hypertension and consented to link their Medicare data from 2006 to 2020. Modified Poisson regression models were used to analyze factors associated with use of ASDs, AIDs, or a mix of both drugs. In paper 2, a retrospective longitudinal cohort study was conducted among Health and Retirement Study (HRS) participants with linked Medicare data from January 2006 to December 2020. Eligible participants aged 65 years or older with hypertension and without dementia entered the cohort between 2008 and 2012 HRS (index wave) and were followed up until the last observed HRS wave before death or administrative censoring (i.e., Medicare Part D enrollment or drop out from HRS). Key exposure was the use of prescription ASDs, AIDs, or both ASDs and AIDs (mixed) for combination therapy, ascertained using Medicare Part D plan and measured biennially in the follow-up as a time-varying exposure. Cognitive outcome included a composite z memory score (standardized to the mean (SD) of 0.81 (0.51) at the index HRS wave) and dementia probability, derived from cognitive measures in biennial HRS surveys. Linear mixed-effects models were used to estimate the associations of ASDs and mixed use (vs AIDs) with each cognitive outcome, with adjustment for confounders (e.g., history of tobacco or alcohol use, health, mental and functional status, comorbidities, and healthcare and medication utilization) assessed at baseline and each 2-year interval in the follow-up. Results: In paper 1, of 4624 identified HRS-Medicare participants, 23.8% used ASDs, 38.6% used AIDs, and 37.6% had mixed use of ASDs and AIDs. Patients with ASD use (vs AID use) and patients with mixed use (vs AID use) were more likely to be female and Black and were less likely to have heart conditions. Other factors associated with ASD use (vs. AID use) included having a household income >$50 000 USD and being less likely to have any hospitalization or emergency department visit. Other unique factors associated with mixed use (vs AID use) included being obese or overweight, undergoing cholesterol screening, experiencing stroke, receiving ≥5 medications other than the studied CNS drugs, and having no use of anticholinergic drugs. In Paper 2, data from 2922 eligible HRS-Medicare participants (mean [SD] aged, 75.4 [6.8] years; 1952 [66.8 %] female) were assessed. Of the sample, 596 (20.4%) received ASD, 1014 (34.7%) received AID, and 1312 (44.9%) had mixed use of both ASDs and AIDs during the 2 years following the index HRS wave. After adjusting for confounders, no difference in memory decline and dementia probability was observed between participants with ASD use or mixed-use vs AIDs at the 12-year follow-up. Conclusion: This retrospective cohort study of older adults aged ≥65 years found several individual-level factors associated with the use of ASDs and mixed use of ASDs and AIDs, compared with AID use only. After accounting for these factors, we found no detectable difference in cognitive performance and dementia probability among older adults with hypertension taking ASDs or mixed-use (vs AIDs). Our results suggest that the neurocognitive outcomes are not significantly different between ASDs or mixed-use of ASDs and AIDs vs AIDs only. Future research should examine whether ASDs exert neuroprotective effects compared to AIDs, to preferentially be recommended among high-risk population for dementia.