Association Between Abnormal Sleep Duration and Stroke in the United States

Abstract

Introduction: The new American Heart Association Life's Essential 8 construct of ideal cardiovascular health now includes sleep duration. Little is known, however, about sleep duration in individuals with prior stroke. Our objective was to compare sleep duration among individuals with and without prior stroke.

Methods: Using the National Health and Nutrition Examination Surveys (NHANES) database (2005-2018), individuals 18 years or older were identified (n = 37,987 without self-reported stroke; n = 1,572 with self-reported stroke). Prevalence of normal sleep duration (7-8 or 6-8 hours/night because of multiple definitions in the literature) was compared between persons with and without self-reported history of stroke using the Rao-Scott χ² test. Multinomial logistic regression analysis was used to evaluate the relationship between prior stroke and abnormal sleep before and after adjusting for demographic and clinical variables.

Results: Compared with individuals without stroke, those with prior stroke were more likely to report >8 hours/night (unadjusted odds ratio (OR) 2.03, 95% confidence interval (CI) 1.64-2.51), <7 hours/night (1.29, 1.08-1.53), and <6 hours/night (1.87, 1.53-2.29). After adjustment, these associations were attenuated (adjusted OR 1.54, 95% CI 1.22-1.94 and 1.15, 0.96-1.38 and 1.50, 1.21-1.85).

Discussion: US stroke survivors are more likely to have abnormal sleep duration than those without prior stroke. Limitations of this study include the cross-sectional and self-reported nature of the data.