Abstract 54: Newly Initiation of Antihypertensive Medication on Reduction of Incident Dementia: Emulation of Target Trials Based on the Health and Retirement Study

Abstract

Background: Although antihypertensive medication use is associated with a lower risk of dementia in observational studies, such an effect is rarely found in randomized controlled trials (RCTs). As midlife high blood pressure is more predictive of incident dementia than hypertension in late life, so midlife hypertension may be the target for intervention. However, no such an RCT has been conducted to prevent incident dementia through reducing high blood pressure from midlife. Therefore, we aimed to emulate a target trial with an observational study design to estimate the effect of initiating antihypertensive medication from midlife on reducing incident dementia. Hypothesis: We assessed the hypothesis that newly initiating antihypertensive medication among midlife adults with hypertension reduces incident dementia. Methods: New-user design was used to emulate a target trial using data from the Health and Retirement Study from 1996 to 2018 with non-institutional dementia-free participants aged 45 to 65 years without taking antihypertensive medication in past two years. Participants with information of self-reported hypertension diagnosis and new use of antihypertensive medication, as well as no dementia based on algorithm were included in the analysis and represented group with/without initiating antihypertensive medication. An intention-to-treat approach was applied for analysis, and sequential emulation was used to increase the efficiency of intention-to-treat analysis. Proportional hazards models were applied to estimate the effect of initiating antihypertensive medication on incident dementia. Inverse probability of treatment and censoring weights were created to adjust for confounding and selection bias due to loss to follow-up, and death was considered as a competing event. Another emulated target trial with the same specifications was conducted among late-life adults aged 66 years or older for comparison. Results: A total of 3,151 non-unique participants (baseline mean age: 59.0 ± 4.0 years, 54.6% (n=1,719) women, 78.3% (n=2468) Whites, 71.4% (n=2,249) initiating antihypertensive medication) were included in the analysis. Initiating antihypertensive medication showed benefit in reducing incident dementia over a total of 22-year follow-up, with a reduction of risk by 19% (hazards ratio=0.81, 95% confidence interval: 0.66, 0.99). No significant reduction was observed for initiators of antihypertensive medication from late life. Conclusions: Initiating antihypertensive medication from midlife, but not from late life may reduce incident dementia. Future studies are warranted to estimate the effect of blood pressure lowering in midlife through other means of intervention in reducing incident dementia.