Abstract
ABSTRACTA substantial literature evaluates the relationship between lifecourse socio-economic status (SES) and cognitive aging, although most studies primarily focus on non-Hispanic white adults with relatively high SES. Studies that are sufficiently inclusive of minoritized and/or low SES adults typically employ traditional regression approaches, which may not be adequately suited for modeling time-varying lifecourse exposures. We used data from the CHAMACOS Maternal Cognition Study (2022-2024), which included middle-aged, primarily immigrant and Latina women who experienced relatively low lifecourse SES (n = 511). Participants provided information on parental (parental education), childhood (respondent education), and midlife SES (current poverty level), and completed the SOL-INCA neurocognitive assessment, yielded global and domain-specific cognitive performance z-scores. We estimated marginal structural models and evaluated evidence against common lifecourse models (accumulation, critical/sensitive periods, pathways). The overall association between parental education and midlife cognitive scores was attenuated to the null after accounting for respondent education and midlife poverty, in support of the pathways model. Compared to those who completed primary school or less, those who completed any secondary school ({3global_cognition: 0.19, 95% Confidence Interval [CI]: 0.07, 0.31, 0.32) or high school or above ({3global_cognition: 0.64, 95% CI: 0.48, 0.80) had higher cognitive z-scores, accounting for both parental education and midlife poverty. The large magnitude of these estimates supports the sensitive periods model. Associations between midlife poverty and cognitive z-scores were modest to null. Our results point to the importance of educational attainment for midlife cognitive health among a subgroup of individuals historically excluded from cognitive aging research.
Publisher
Cold Spring Harbor Laboratory