Influencing factors of depressive symptoms in middle-aged and elderly people and its regional differences in china: a study based on Bayesian network model.

BACKGROUND: This study aims to explore the influencing factors and regional differences in the presence of depressive symptoms (DS) among middle-aged and elderly people in China, as well as to explore the network relationship among the influencing factors by constructing Bayesian networks (BNs) model. METHODS: A total of 3582 respondents were included using China Health and Retirement Longitudinal Study (CHARLS) 2020 data. Logistic regression analysis was used to screen for variables related to DS. And the intricate conditional dependencies were visualized by BNs. Differences in the spatial distribution of DS were visualized by a geographic information system. RESULTS: During the COVID-19 epidemic, the prevalence of DS in China was 46.5%. The high prevalence areas of DS were mainly clustered in the west (Sichuan, Qinghai) and middle (Hubei, Henan) regions of China. The results showed that the occurrence of DS among middle-aged and elderly people exhibits a direct probabilistic dependency with gender, body pain, self-reported health status, IADL (Instrument Activities of Daily Living), length of night sleep and region. Concurrently, age, education, chronic diseases, marital status and BADL (Basic Activities of Daily Living) were found to have an indirect probabilistic dependency with the prevalence of DS in these individuals. CONCLUSIONS: The overall prevalence of DS remains high among middle-aged and older adults. Predominantly, the hotspots for high prevalence of DS are concentrated in middle and west regions of China. By identifying the most direct risk factors for DS, healthcare providers and community workers can develop targeted interventions to prevent and manage this condition.