Moderation of the Association between Primary Language and Health by Race and Gender: An Intersectional Approach

Abstract

In the United States (US), limited English proficiency is associated with a higher risk of obesity and diabetes. “Intersectionality”, or the interconnected nature of social categorizations, such as race/ethnicity and gender, creates interdependent systems of disadvantage, which impact health and create complex health inequities. How these patterns are associated with language-based health inequities is not well understood. The study objective was to assess the potential for race/ethnicity, gender, and socioeconomic status to jointly moderate the association between primary language (English/Spanish) and having obesity and diabetes. Using the 2018 Behavioral Risk Factor Surveillance System (n = 431,045), weighted generalized linear models with a logistic link were used to estimate the associations between primary language (English/Spanish) and obesity and diabetes status, adjusting for confounders using stratification for the intersections of gender and race/ethnicity (White, Black, Other). Respondents whose primary language was Spanish were 11.6% more likely to have obesity (95% CI 7.4%, 15.9%) and 15.1% more likely to have diabetes (95% CI 10.1%, 20.3%) compared to English speakers. Compared to English speakers, Spanish speakers were more likely to have both obesity (p < 0.001) and diabetes (p < 0.001) among White females. Spanish speakers were also more likely to have obesity among males and females of other races/ethnicities (p < 0.001 for both), and White females (p = 0.042). Among males of other racial/ethnic classifications, Spanish speakers were less likely to have both obesity (p = 0.011) and diabetes (p = 0.005) than English speakers. Health promotion efforts need to recognize these differences and critical systems−change efforts designed to fundamentally transform underlying conditions that lead to health inequities should also consider these critical sociodemographic factors to maximize their effectiveness.

Keywords: diabetes; intersectionality; language; obesity; socioeconomic status.

Figure 1

Predicted prevalence (and 95% confidence intervals) of obesity (left) and diabetes (right) by gender, race/ethnicity, income, and preferred language spoken.