2236-P: Nativity and English Language Use in Relation to Undiagnosed and Diagnosed Diabetes in U.S. Adults

Introduction and Objective: Diabetes detection varies across demographic and immigration-related groups. Foreign-born adults and individuals with limited English proficiency may be less likely to have diabetes identified, even when the disease is present. This study examined demographic, socioeconomic, and immigration-related correlates of undiagnosed and diagnosed diabetes among U.S. adults. Methods: Data from the National Health and Nutrition Examination Survey (NHANES, 2011-2023) were analyzed using survey-weighted multinomial logistic regression. Diabetes status was categorized as no diabetes, undiagnosed diabetes, or diagnosed diabetes based on self-report and laboratory criteria. Models sequentially adjusted for age, sex, race/ethnicity, nativity, English language use, income, and education. Results are presented as log-odds with 95% confidence intervals, accounting for the complex survey design. Results: Undiagnosed diabetes was more prevalent among foreign-born adults than native-born adults (4.7% vs. 2.9%), while diagnosed diabetes prevalence was similar (10.8% vs. 11.1%). Compared with non-Hispanic White adults, Mexican adults had higher odds of undiagnosed diabetes (odds ratio = 2.3). Older age was associated with higher odds of both diabetes categories. After adjustment, native-born adults had lower odds of undiagnosed diabetes compared with foreign-born adults (odds ratio = 0.80). Inclusion of English language use attenuated nativity associations with diagnosed diabetes, while limited English use remained significantly associated with undiagnosed diabetes. Higher income and educational attainment were associated with lower odds of undiagnosed diabetes. Conclusion: Variation in diabetes detection is associated with age, race/ethnicity, and immigration-related factors. The use of the English language appears to influence whether diabetes is identified, rather than whether it is actually present. Linguistically responsive screening strategies may support more timely diabetes diagnosis. Disclosure A. Kuerban: None.