Joint associations of sleep quality and depressive symptoms with incident digestive disease in a nationally representative Chinese cohort

Abstract

Abstract Sleep disturbance and depressive symptoms are common and modifiable, yet their joint and longitudinal associations with incident digestive disease remain unclear. Using data from the China Health and Retirement Longitudinal Study, this national prospective cohort study included 3,711 Chinese adults free of digestive disease at baseline. Sleep quality (SQ) was derived from questionnaire-based sleep items, and depressive symptoms were assessed using the 10-item Center for Epidemiologic Studies Depression Scale (CESD-10). Cox proportional hazards models estimated associations with incident self-reported digestive disease after adjustment for sociodemographic, lifestyle, and comorbidity factors. Exploratory mediation analyses assessed model-based indirect associations, and joint trajectory-pattern analyses characterized longitudinal patterns of SQ and depressive symptoms. Sensitivity analyses examined alternative exposure definitions, restriction to participants without baseline comorbidities, and modified CESD-9 analyses after excluding the restless-sleep item. In separate models, poor SQ (HR = 1.67, 95% CI 1.39–2.01) and depressive symptoms (HR = 1.33, 95% CI 1.13–1.57) were both associated with higher risk. Compared with good SQ without depressive symptoms, elevated risks were observed for poor SQ without depressive symptoms (fully adjusted HR = 1.91, 95% CI 1.51–2.42) and poor SQ with depressive symptoms (HR = 1.80, 95% CI 1.45–2.23). Exploratory mediation analyses showed an attenuated indirect association through depressive symptoms for the SQ–digestive disease association after full adjustment (ACME = 0.006; P  = 0.106; proportion mediated = 7.8%), whereas the indirect association through SQ for the depressive symptoms–digestive disease association remained evident (ACME = 0.025; proportion mediated = 47.3%). Joint trajectory-pattern analyses showed elevated risks in the moderate worsening and marked worsening/lower SQ groups (HR = 1.50, 95% CI 1.24–1.82; and HR = 1.48, 95% CI 1.23–1.78, respectively). Sensitivity analyses were generally consistent, including analyses using modified CESD-9 after excluding the restless-sleep item. Poor SQ and depressive symptoms were associated with higher subsequent risk of incident self-reported digestive disease. The findings highlight a potential sleep–mood–digestive health link, although causal and interventional implications require further validation using objective sleep measures and refined disease phenotyping.