Beverage-Derived Caffeine Intake, Sleep Duration, and Mental Health Among Korean Adults: A Cross-Sectional Analysis of KNHANES 2021–2024

Background/Objectives: To examine the association between caffeine intake and mental health- and sleep-related outcomes, including anxiety, sleep duration, and perceived stress, in a nationally representative sample of Korean adults. Methods: We analyzed 20,504 participants from the Korea National Health and Nutrition Examination Survey (KNHANES) 2021–2024. Caffeine intake was categorized into six groups, and mental health- and sleep-related outcomes included clinically significant anxiety symptoms (GAD-7 ≥ 10), sleep duration, and perceived stress. Survey-weighted multivariable logistic regression models and restricted cubic spline analyses were used to evaluate associations between caffeine intake and mental health- and sleep-related outcomes, adjusting for demographic, socioeconomic, behavioral, dietary, and clinical factors. Results: The prevalence of clinically significant anxiety symptoms was 4.7%. Restricted cubic spline analyses demonstrated a non-linear association between caffeine intake and anxiety (p-overall = 0.010; p-nonlinear = 0.016). Higher caffeine intake was associated with progressively increasing odds of short sleep duration (<6 h/day), indicating a clear dose–response relationship (p-overall < 0.001; p-nonlinear < 0.001). Higher caffeine intake was also associated with increased perceived stress (p-overall = 0.025), although evidence for a nonlinear relationship was weaker (p-nonlinear = 0.062). These findings were generally consistent across multivariable-adjusted models. Conclusions: Caffeine intake is differentially associated with mental health- and sleep-related outcomes, with excessive intake related to adverse effects on sleep and stress.