The association between weekend catch-up sleep and obesity in U.S. adults: A cross-sectional analysis of NHANES 2017–2020

Weekend catch-up sleep (WCS) is a behavioral strategy employed to compensate for weekday sleep loss and has been reported to be associated with various metabolic outcomes. However, its association with obesity remains controversial. This study aimed to investigate the relationship between WCS and obesity among U.S. adults using data from the National Health and Nutrition Examination Survey. We analyzed data from 8279 individuals aged 20 years and older, derived from the 2017-2020 cycles of the National Health and Nutrition Examination Survey database. WCS was defined as the difference between self-reported sleep duration on weekends and weekdays, and subsequently divided into quartiles. Obesity was identified based on a body mass index threshold of 30 kg/m² or higher. To examine the relationship between WCS and obesity, we employed survey-weighted logistic regression models, treating WCS as a continuous variable. Restricted cubic spline curves and two-piecewise regression models were additionally explored for sensitivity, but statistical tests did not support a nonlinear association. Stratified analyses and interaction tests were performed as exploratory assessments to evaluate potential effect modifiers. Each additional hour of WCS was associated with a modestly higher odds of obesity (odds ratio = 1.03, 95% confidence interval: 1.00-1.06; P = .028) after adjusting for demographic, socioeconomic, behavioral, and clinical factors. Statistical tests did not support a nonlinear relationship (P for nonlinearity = .816), indicating that the association is best characterized as linear. Exploratory subgroup analyses suggested potentially stronger associations in females, non-Hispanic Black individuals, and nonsmokers; however, these findings were not statistically robust and should be interpreted with caution. WCS was positively associated with obesity in U.S. adults in a modest linear manner. Expl oratory models suggested a possible threshold, but statistical tests (P for nonlinearity = .816; likelihood ratio test P = .113) did not support this. Thus, the linear model provides the most parsimonious description of the data. These findings suggest that while moderate WCS may reflect an attempt to compensate for sleep debt, excessive or irregular sleep extension was associated with higher odds of obesity. However, given the cross-sectional design, these results indicate association rather than causation.