BACKGROUND: Parental education is an important determinant of late-life cognition, but the extent to which intervening on midlife risk factors, such as hypertension, mitigates the impact of early-life factors is unclear. Novel methodological approaches, such as causal decomposition, facilitate the assessment of contributors to health inequities through hypothetical interventions on mediating risk factors. METHODS: Using harmonized cohorts (Kaiser Healthy Aging and Diverse Life Experiences Study
Study of Healthy Aging in African Americans) and a ratio of mediator probability weights decomposition approach, we quantified disparities in late-life cognitive test scores (semantic memory, executive function, and verbal memory z-scores) across high versus low parental education, and evaluated whether socioeconomic disparities in late-life cognitive test scores would change if the corresponding disparity in midlife hypertension were eliminated. RESULTS: We observed substantial disparities across levels of parental education in late-life cognitive test scores (eg, =-0.72 95% CI: -0.84 to -0.60 for semantic memory). Hypothetical intervention on midlife hypertension did not substantially reduce disparities in any cognitive domain. Patterns were similar when stratified by race. CONCLUSIONS: Future work should evaluate other points of intervention across the lifecourse (eg, participant education) to reduce late-life cognitive disparities across levels of parental education.