Objective: To investigate the relationship between dietary carotenoid intake and sleep duration.
Methods: Adults enrolled in the National Health and Nutrition Examination Survey (NHANES) 2007-2018 without missing information on dietary carotenoid intake (α-carotene, β-carotene, β-cryptoxanthin, lycopene, and lutein + zeaxanthin), sleep duration, and covariates were included. Participants’ carotenoid consumption was divided into three groups by quartiles and sleep duration was grouped as short (<7 h/night), optimal (7-8 h/night), and long (>8 h/night). Multinominal logistic regression was constructed to examine the association between dietary carotenoid intake and sleep duration. Restricted cubic spline (RCS) regression was further utilized to explore their dose-response relationship. The weighted quantile sum (WQS) model was adopted to calculate the mixed and individual effect of 5 carotenoid sub-types on sleep duration.
Results: Multinominal logistic regression presented that people with higher intakes of α-carotene, β-carotene, β-cryptoxanthin, lycopene, and lutein + zeaxanthin were less likely to sleep too short or too long. Consistent with the findings from multinominal logistic regression, the RCS models suggested a reverse U-shaped relationship between sleep duration and carotenoid intakes. The mixed effects were also significant, where β-cryptoxanthin and lutein + zeaxanthin were the top 2 contributors associated with the decreased risks of short sleep duration, while β-carotene, α-carotene, and β-cryptoxanthin were the main factors related to the lower risk of long sleep duration.
Conclusion: Our study revealed that the American adults with optimal sleep duration were associated with more dietary carotenoid intake, in comparison to short or long sleepers.