Objective: To evaluate the association between the exposure to bedroom light-at-night and depressive symptoms in children and adolescents. Methods: From April to December 2024, a study was conducted in students at grade 4-7 recruited from 3 schools in Shanghai. The 10-item Center for Epidemiological Studies Depression Scale was used to detect depressive symptoms, and a portable illuminance meter was used to measure the intensity of bedroom light-at-night for one week. The log-binomial regression model was applied to analyze the association between the exposure to bedroom light-at-night and depressive symptoms. Results: A total of 628 children and adolescents were included, including 314 boys and 314 girls. The average age of the students was (10.8±1.1) years. The detection rate of depressive symptoms was 26.4%. The median intensity of bedroom light-at-night from 21:00 to 07:00 was 5.1(2.2,10.5) lx, which was higher on schooldays [5.1(1.9,10.1) lx] than that on weekends [4.0(1.3,9.3) lx]. Log-binomial regression analysis showed that during 21:00 - 7:00 , compared with the Q1 group of exposure to light-at-night, the PR for depressive symptoms in the Q4 group was 1.431 (95%CI: 1.002-2.043) and during 21:00-22:00, compared with Q1 group, both the Q2 and Q4 groups showed higher PR for depressive symptoms [1.669 (95%CI: 1.136-2.451); 1.538 (95%CI: 1.020-2.319)]. Similar results were observed during 21:00-23:00. During 21:00-24:00, the PR for depressive symptoms in the Q4 group significantly increased compared with Q1 group [1.611 (95%CI: 1.064-2.440)]. Associations between different levels of exposure to bedroom light-at-night and depressive symptoms were observed in different times on school days and in girls. No significant association was found between the exposure to bedroom light-at-night and depressive symptoms in boys and on weekends (P>0.05). Conclusion: Higher exposure to bedroom light-at-night during 21:00-24:00 was associated with depressive symptoms in children and adolescents, especially in girls and on schooldays.