Abstract
INTRODUCTION: The association between adult child educational attainment and older parent's cognitive health may vary across diverse contexts but cross-national comparisons have been limited by differences in outcome assessment, study design, and analytic choices. METHODS: We used harmonized data with comprehensive cognitive assessments from the United States (N = 3088), India (N = 3828), and Mexico (N = 1875) to estimate associations between adult child education and older adults' cognitive functioning using linear regression models adjusted for respondent and family-level socio-economic status (SES) in each study. RESULTS: Each additional year of offspring education was associated with 0.02 (Longitudinal Aging Study in India - Diagnostic Assessment of Dementia [LASI-DAD]) to 0.04 (Health and Retirement Study Harmonized Cognitive Assessment Protocol Survey [HRS-HCAP], Mexican Health and Aging Study Cognitive Aging Ancillary Study [Mex-Cog]) standard deviation (SD) units higher cognitive score (pooled estimate: 0.032 [95% confidence interval [CI]: 0.018-0.046]), comparable to about 1/3-1/4 of the association with respondents' own years of education. Differences by respondent gender were heterogeneous across contexts. DISCUSSION: Consistent overall estimates despite differences in context and potential confounding structures underscore the importance of offspring education for cognitive outcomes among older adults. HIGHLIGHTS: The harmonized study design allows for fair comparisons across diverse contexts. Effect sizes were largely consistent across the United States, India, and Mexico, despite differences in confounding structures. The pooled association between adult child education and parental cognitive functioning was about 1/3-1/4 of the association with respondent's own years of education. Heterogeneity in gender differences point to the potential effects of local culture and context.