English language proficiency and patterns of use as related to cognitive change in native Spanish speakers of the Hispanics/Latinos of the Hispanic Community Health Study/Study of Latinos‐Investigation of Neurocognitive Aging

Abstract

AbstractBackgroundGiven mixed results regarding the role of bilingualism in cognitive aging, investigators are increasingly studying continuous measures of bilingualism (e.g., levels of language proficiency or use) as opposed to a dichotomous “yes/no” one, to better characterize the bilingual experience. We investigated associations of continuous measures of bilingualism and cognition.MethodEligible Hispanic Community Health Study/Study of Latinos (HCHS/SOL) participants self‐reported their and their parents’ birth outside the 50 United States and Spanish as their first language, and used Spanish for baseline cognitive testing at their HCHS/SOL visit (2008‐2011) and comparable cognitive testing 7‐years later as part of the SOL‐Investigation of Neurocognitive Aging (SOL‐INCA; 2015–2018). Composite scores for baseline self‐reported Spanish/English language proficiency (reading/speaking) and patterns of use (thinking/socializing) were rated using a Likert scale from 1 = only Spanish to 4 = English>Spanish. Cognitive testing at both visits included test‐specific and a global composite score(s) of verbal learning and memory, Word Fluency, and Digit Symbol Substitution (DSS). Survey linear regression models examined the associations between our two bilingualism scores (separately) and cognition (at follow‐up, and standardized measures of 7‐year change) adjusting for age, sex, education, depressive symptoms, income, Hispanic/Latino background, and field center.ResultThere were 4,005 eligible participants (unweighted mean SOL‐INCA population age∼56 years, 56% female). Higher levels of second‐language (English) proficiency (β = 0.050, SE = 0.015, p<0.001) and patterns of use (β = 0.035, SE = 0.010, p<0.001) were associated with higher levels of global cognition at follow‐up (Figure), specifically, Fluency and DSS (p‐values<0.001). Only higher levels of second‐language proficiency were associated with better than predicted change in cognition, specifically, Fluency (β = 0.060, SE = 0.023, p<0.01).ConclusionIn an analytic sample of mid‐ to late‐life native Spanish speakers currently living in the US, second‐language (English) proficiency and patterns of use at baseline were associated with better cognitive outcomes over seven years of follow‐up. Associations of bilingualism composites and cognition suggest distinctions in the bilingual experience may differentially relate to level versus 7‐year change in cognition for Hispanics/Latinos. Investigations of foreign language learning on cognition in older adults, some of which have found no cognitive benefit, may be enhanced by using continuous measures of second‐language proficiency and, perhaps, patterns of language use.