A Multilevel Investigation of Migration and Dementia Risk Among Mexican Americans

The US-Mexico migration corridor is one of the busiest in the modern era. In the United States, over 15 million adults ages 50+ are foreign-born, most of Mexican origin. As immigration becomes more common, we urgently need better evidence on its health effects, social contexts that may remediate or exacerbate the health risks of immigrants, and biological mechanisms of these effects. Relatively little is known about the determinants of their risk for Alzheimer’s disease or related dementias (ADRD) among Mexican Americans. In addition to socioeconomic disadvantage, Mexican Americans may be exposed to unique migration-related stressors and experiences. Embodiment of migration-related stressors may manifest through accelerated biological aging or increased inflammation which, in turn, may elevate ADRD risk. Quantifying ADRD risk in Mexican Americans is not straightforward, however, due to the methodological challenge of disentangling the effect of migration from the selection factors that predict migration itself and ADRD. If unaccounted for, predictors of healthy in-migration and unhealthy out-migration (often offered as reasons for the Latino paradox) may mask true health disadvantages due to immigration. Additionally, research on the social epidemiology of migrant health, ADRD in particular, has underexamined the lifecourse stress pathways that shape the pace of biological aging. Addressing these gaps in the field and our knowledge demands methodological innovation and assessment of multilevel drivers of migration and risk of ADRD. Further, most data sources cannot disentangle selection/confounding from causation because they include only migrants (i.e., they enroll US residents and compare migrants to U.S.-born individuals). Often, the outcomes of those born in the communities of origin who do not migrate are never represented. In this proposal, we will leverage our team’s previous infrastructural work (under RF1AG055486) that built a binational dataset pooled across the US (from the Health and Retirement Study or HRS) and Mexico (from the Mexican Health and Aging Study or MHAS) that spans the life course (beginning from birth until age at migration or study entry). This dataset has built-in longitudinal propensity scores (weights) capturing pre- and post-migration characteristics, which is a crucial step for correcting bias due to selective migration, a major barrier to the effective study of migrant health. Using this unique dataset and innovative life course models, we propose to (Aim 1) estimate the effect of migration, residence in the US, and acculturation on cognitive function and dementia risk, (Aim 2) evaluate the embodiment of migration-related risk factors through stress pathways, and (Aim 3) assess embodiment through accelerated biological aging. We will use the Harmonized Cognitive Assessment Protocol (HCAP) which will allow major improvements on what was previously possible in this binational cohort. The study will provide novel insights into the multilevel drivers and underlying mechanisms of migration’s effects on ADRD.