Abstract
Depression is a common mental disorder formed by a combination of various factors. Existing researches have already demonstrated that noise indeed impacts the level of depression, but their results were inconsistent. To reconcile seemingly contradictory findings, this study aims to investigate how noise affects individual depression using big data mining and analysis techniques. The individual data was obtained from the China Health and Retirement Longitudinal Study (CHARLS) over 3 years (2013, 2015, and 2018) totaling 9693 participants coming from 125 different cities. The Chinese version of the 10-item Center for Epidemiologic Studies Depression Scale (CES-D) was employed to assess depression scores, while the search index for noise-related keywords was obtained from Baidu Index to measure noise levels across different cities. A curvilinear model with fixed effects was applied to analyze the relationship between noise and depression. Additionally, moderating effect analyses were conducted to examine the influence of city size and green space. The results indicate a U-shaped relationship between depression and noise, wherein depression initially decreases with increased noise, then subsequently rises. The moderating effect analysis suggests that both city size and green space influence this U-shaped curve; notably, in cities with larger populations or higher green coverage rates, the curve flattens. This study reveals that the impact of noise on depression is complex, which is the result of a multifactorial synergy. It underscores the necessity for urban planning and management to prioritize the creation of friendly sound environments, which could enhance the physical and mental health of urban residents.