UCSF

By 2060, the number of individuals in the United States with Alzheimer's disease and related dementias (ADRDs) is projected to triple to about 13.9 million, with Latinos experiencing the largest increase. Research has shown that both parental and own education impact late-life cognitive health, and upward educational mobility from one generation to the next may partially compensate for the adverse cognitive health effects of low parental education. However, it remains unclear to what extent gains in educational attainment may help individuals with low parental education attain cognitive health benefits similar to those with multiple generations of high levels of education. Given the significance of education as a modifiable risk factor for ADRDs, it is critical to understand how generational increases in education in the US impact cognitive health, especially among marginalized racial and ethnic groups with historically limited access to education. Understanding how the cognitive health benefits of higher education compare for individuals who are first-generation and individuals who have benefited from multiple generations of high levels of education and how this varies across race and ethnicity can inform social policies to address cognitive health disparities due to low education levels. This dissertation aims to examine the relationship between intergenerational education and cognitive health using data from the US Health and Retirement Study (HRS) and the Study of Latinos – Investigation of Neurocognitive Aging (SOL-INCA), a large cohort of Latino older adults, and investigate whether the association differs for older Black, Hispanic, and White adults and within Latino heritage subgroups (e.g. Cuban, Dominican, etc.). Additionally, this dissertation makes a significant contribution to the literature as the first study to evaluate the relationship between highly educated offspring and cognitive health for Latinos in the US. The first chapter of my dissertation examined the association between being a first-generation high school graduate (i.e., neither parent graduated high school) vs. being a multi-gen high school graduate (i.e., at least one parent graduated high school) and cognitive performance, decline, and incidence of possible cognitive impairment no dementia (CIND) or probable dementia for Black, Latino, and White older adults. Compared to multi-generational high school graduates, first-generation graduates had notably lower baseline verbal learning and memory z-scores. Black and White first-generation graduates experienced a faster rate of decline in verbal learning compared to their multi-generational counterparts; rates did not differ for Hispanic graduates. First-generation high school graduates also had higher hazard ratios of possible CIND or probable dementia compared to multi-generational high school graduates, with the greatest difference among White respondents. Recognizing that Hispanic and Latino older adults are not a monolithic group, the second chapter examined the association between upward intergenerational educational mobility and cognitive outcomes within Latino subgroups. Specifically, we evaluated the association between being a first-generation (vs. multi-generation) high school graduate and cognitive performance and cognitive change, by Latino heritage group and nativity (US-born vs. non-US born). Compared to their multi-gen counterparts, first-generation Cuban, Mexican, and Puerto Ricans scored significantly lower on verbal learning. First-generation respondents born outside the US scored significantly lower across domain-specific and global cognitive outcomes. Finally, the third chapter focuses on the education of younger generations and examines how the education of older adults’ offspring is associated with their cognitive function, decline, and incidence of possible CIND or probable dementia. We found that each year of offspring education over 12 years was associated with higher baseline verbal learning and memory z-scores and a slightly faster rate of decline in verbal learning. Furthermore, Hispanic participants had a lower risk of possible CIND or probable dementia compared to their White counterparts with each additional year of offspring education over 12 years. Together, these studies provide further evidence into the importance of investing in education across generations, particularly when considering differences by race and ethnicity. Our findings suggest that dementia risk reductions attributable to higher levels of intergenerational educational attainment may continue to accrue in the future, which could help address inequities in cognitive health between populations with different educational backgrounds.