The association between nighttime sleep duration and digestive diseases in middle-aged and older population in China: mediating role of depression

Current studies have not sufficiently explored the connection between nighttime sleep duration and digestive diseases, particularly the potential mediating effect of depression in this relationship. This study aims to clarify the interrelationships among the three variables. This study utilized the 2018 CHARLS (China Health and Retirement Longitudinal Study) dataset to examine the associations among nighttime sleep duration, depression, and digestive diseases. The analysis incorporated binary logistic regression, Spearman correlation analysis, sensitivity analysis, the restricted cubic spline method, subgroup analysis, and mediation effect analysis based on Baron and Kenny’s causal steps approach and the Karlson/Holm/Breen (KHB) method. Binary logistic regression analysis revealed that short sleep (< 6 h) was significantly associated with higher odds of digestive diseases compared to optimal sleep (6–8 h) (OR = 1.409, 95%CI: 1.179–1.683, P < 0.001). Restricted cubic spline analysis indicated that the lowest odds of digestive diseases occurred at approximately 7.8 h of nighttime sleep among middle-aged and older adults. In subgroup analyses, longer nighttime sleep duration was not significantly associated with lower odds of digestive diseases among individuals with hypertension, diabetes, or depression. Depression accounted for 20.98% of the association between nighttime sleep duration and digestive diseases. After full adjustment, depression still explained 17.26% of this association. This paper suggests a significant association between nighttime sleep duration and digestive diseases. Specifically, insufficient sleep time is associated with a higher prevalence of digestive diseases among middle-aged and elderly people. Furthermore, depression partially mediates the relationship between nighttime sleep duration and digestive diseases. The results are based on association analyses and do not imply causality. Therefore, the observed mediating effect should be interpreted as a statistical decomposition of associations rather than evidence of a causal pathway.