AHA Scientific Sessions 2026 Abstract · Nov 2, 2026
Estimating Antihypertensive Medication Effect on Midlife Cognitive Function Across Childhood Systolic Blood Pressure Variability Levels: A Target Trial Emulation in the Bogalusa Heart Study
By Kang SJ, Gustat J, Tang W et al.
Introduction: Higher systolic blood pressure variability (SBPV), independent of mean SBP levels has been linked to cognitive decline in adults. Although blood pressure lowering with antihypertensive medication (AHTM) is known to be protective against target organ damage, associations between SBPV, AHTM use, and cognitive outcomes have been inconsistent in older adults. Whether the effect of AHTM on midlife cognition differs by SBPV from childhood is unclear. We conducted a target trial emulation to estimate the effect of AHTM initiation on midlife cognitive function and to assess whether childhood SBPV modifies this effect. Methods: Eligibility criteria were participants of the Bogalusa Heart Study who developed hypertension at ≥20 years. Childhood SBPV was measured by the coefficient of variation using repeated SBP measures from childhood up to age 20. High variability was defined as the top tertile. Time at start of follow up (t0) was the age at hypertension onset. We compared two treatment strategies: (1) initiate AHTM within 5 years of hypertension onset, and (2) never initiate AHTM. Participants with prevalent hypertension or AHTM use ≤ t0 were excluded. Each participant was cloned and assigned to both strategies to align strategy assignment to t0 (Figure 1). Clones were censored when observed treatment deviated from the assigned strategy. The outcome was global cognitive score (GCS) measured at midlife, defined as the average of 11 age-, race-, and sex-standardized neuropsychological tests, and four cognitive domains. Per-protocol effects of AHTM on midlife cognition were estimated using inverse-probability-of-censoring weighted pooled linear regression. Censoring weights were obtained from models for informative censoring under each strategy that included confounders measured at hypertension onset (glucose, BMI, and non-HDL cholesterol) and time-varying SBP. Effect modification by childhood SBPV was evaluated using stratified analyses. Results: Of 549 eligible participants (age at t0: 38 ± 9 years), 118 initiated AHTM at a mean age of 46 ± 5 years. A total of 359 participants completed cognitive assessment at a mean of 10 ± 9 years after hypertension onset (t0). Across all childhood SBPV tertiles, initiating AHTM within 5 years of t0 showed no effect on midlife GCS or cognitive domain performance compared with never initiating AHTM (Figure 2). Conclusions: Across all childhood SBPV tertiles, no effect of AHTM initiation within 5 years of hypertension onset on midlife cognition was observed. This null effect may reflect persistence of SBPV even when mean SBP levels are lowered by AHTM, which may adversely affect midlife cognition. Alternatively, unmeasured confounders such as poor medication adherence, which can aggravate SBPV, as well as small sample size, may limit the ability to detect heterogeneity in the effect of AHTM on midlife cognition across childhood SBPV tertiles.