The association between diabetes and excessive daytime sleepiness among American adults aged 20–79 years: findings from the 2015–2018 National Health and Nutrition Examination Surveys

Abstract

To examine the association between diabetes and excessive daytime sleepiness (EDS). Using the 2015-2018 cycles of the National Health and Nutrition Examination Survey, we applied survey-featured modified Poisson regression to estimate the association between diabetes and EDS among American adults aged 20-79 years, adjusting for confounding demographic, clinical and lifestyle variables. Effect modification by age, sex, race, education, income, sleep apnea and inadequate sleep was assessed. We performed sensitivity analyses using propensity score matched (PS) data and applied ordinal logistic regression using multiple levels of daytime sleepiness. Among people with diabetes, we assessed the association between EDS and diabetes care variables. Of the 6289 participants, 895 (10%) had diabetes. The estimated prevalence of EDS was higher among adults with diabetes (30.6%) than counterparts without diabetes (26.3%). After adjusting for confounding variables, diabetes remained associated with EDS (aPR:1.20; 95%CI:1.06 1.36). There was no statistically significant effect modification. Sensitivity analyses confirmed our main results. Among people with diabetes, there was limited evidence that the diabetes care variables were related to EDS. Among American adults, diabetes is associated with EDS after controlling for confounding variables. Although the cross-sectional design is a limitation, our findings support further exploration of the role of diabetes in EDS.