Parental Education, Own Education, and Cognitive Function in Middle-Aged and Older Adults

Importance: There is a lack of research examining the direct and independent associations of maternal and paternal education with later-life cognitive decline across diverse contexts. Objective: To evaluate the independent association of maternal and paternal education with cognitive decline over time among middle-aged and older adults in 4 countries, and to assess whether these associations are mediated by individuals' own education. Design, Setting, and Participants: This multicohort study used data from 4 nationally representative cohorts, including the China Health and Retirement Longitudinal Study (CHARLS; 2011-2015), the Health and Retirement Study (HRS; 2010-2018) in the US, the English Longitudinal Study of Aging (ELSA; 2010-2018), and the Mexican Health and Aging Study (MHAS; 2012-2018). The study population consisted of participants aged 50 years or older with complete data on parental education, cognitive function, and covariates and without cognitive impairment at baseline. Statistical analysis was performed from November 1, 2023, to April 30, 2024. Exposures: Participants self-reported both parents' education, which was further harmonized into 4 categories based on the 2011 International Standard Classification of Education (ie, less than primary, primary, lower secondary, and upper secondary or higher). Main Outcomes and Measures: The primary outcome was cognitive function, measured in 2 domains (ie, episodic memory and mental status). An overall cognitive state score was derived by combining scores from these 2 domains. All cognitive outcomes were standardized as z scores according to the baseline mean and SD within each cohort. Linear mixed-effects models were used to estimate the associations between parental education and cognitive decline over time. Generalized structural equation models were applied to examine the mediating role of participants' own education in such associations. Results: This study included 7898 participants from China (mean [SD] age, 61.0 [7.4] years; 4141 men [52.4%]), 12 402 from the US (mean [SD] age, 64.8 [9.7] years; 5304 men [42.8%]), 6603 from England (mean [SD] age, 65.3 [8.2] years; 2915 men [44.1%]), and 9162 from Mexico (mean [SD] age, 63.4 [8.5] years; 3863 men [42.2%]). Compared with less than primary education, maternal education at the upper secondary level or higher was associated with slower rates of decline in cognitive state (β = 0.082 [95% CI, 0.035 to 0.129] SD per year in CHARLS, β = 0.025 [95% CI, 0.012 to 0.038] SD per year in HRS, and β = 0.040 [95% CI, 0.016 to 0.064] SD per year in ELSA). In MHAS, only primary maternal education was associated with slower decline in cognitive state (β = 0.010 [95% CI, 0.004 to 0.017] SD per year). Similarly, paternal education at the upper secondary level or higher, compared with less than primary, was associated with slower cognitive state decline in China (β = 0.042 [95% CI, 0.021 to 0.063] SD per year), the US (β = 0.027 [95% CI, 0.014 to 0.039] SD per year), and England (β = 0.044 [95% CI, 0.012 to 0.077] SD per year), whereas in Mexico, a significant association was observed for primary paternal education (β = 0.010 [95% CI, 0.004 to 0.017] SD per year). Participants' own education significantly mediated the associations between parental education and annual decline in cognitive state in all cohorts, except for MHAS. Conclusions and Relevance: In this cohort study of middle-aged and older adults across 4 countries, both higher maternal and paternal education were generally associated with slower cognitive decline, and these associations were mediated by participants' own education. These findings highlight the long-term relevance of parental education for offspring cognitive health across diverse cultural or socioeconomic contexts, and support the potential benefits of improving educational attainment to reduce intergenerational disparities in late-life cognitive health.