Impact of limited English proficiency on the control of diabetes and associated cardiovascular risk factors. The National Health and Nutrition Examination Survey, 2003–2018

Abstract

Language barriers pose a challenge to managing health conditions for various personal, interpersonal, and structural reasons. This study estimates the impact of limited English proficiency (LEP) on diabetes mellitus control and associated cardiovascular risk factors in a large representative sample of United States adults. Cross-sectional data from the National Health and Nutrition Examination Survey (NHANES, 2003-18) was used to estimate the impact of language proficiency on glycemic control (glycated hemoglobin [HbA1c]) and cardiovascular risk status (blood pressure [BP] and low-density lipoprotein [LDL]) in adult participants with known diabetes disease. The analysis included descriptive statistics and generalized linear models to adjust for sociodemographic characteristics. The study sample included 5017 participants with self-reported, physician-diagnosed diabetes mellitus. Most participants completed NHANES interview in English (90.8%), whereas some participants completed the interview in Spanish (LEP-Spanish; 6.6%) or requested an interpreter (LEP-interpreter; 2.6%). Compared to English-speaking participants, LEP-interpreter participants were more likely to have HbA1c ≥ 7% (OR = 1.6, 95% CI = 1.1, 2.4) or a combination of HbA1c ≥ 7%, LDL ≥ 2.6 mmol/L, and BP ≥ 130/80 mmHg (OR = 3.1; 95% CI = 1.2, 8.2). We observed no differences in the odds of diabetes control. between English-speaking and LEP-Spanish participants, whereas LEP-interpreter participants had worse diabetes control, possibly owing to the greater likelihood of patient-provider language discordance for non-English non-Spanish-speaking patients. Given that many patients, yet few providers, speak languages other than English or Spanish, innovative ways are needed to facilitate patient-provider communications (e.g., digital communication assistance tools).