Objective: To investigate the association between different digital media usage types and depressive symptoms among junior and senior high school students, so as to provide a scientific reference for making precise mental health prevention and intervention strategies for adolescents. Methods: In October 2024, a convenience cluster sample of 3 225 students was collected from 2 junior high schools and 2 senior high schools in Shenyang City. Participants were investigated for its daily usage duration of various digital media types and depressive symptoms. Cochran-Armitage trend test was used to examine trend in depressive symptom detection rates across digital media usage types. Multivariate Logistic regression model was employed to assess associations, with stratified analyses by gender and educational stage. Results: The daily usage durations for educational/learning, social/chatting, gaming, and video/music/novel digital media were 30.0(12.1, 60.0), 22.9(9.3, 50.0), 17.1(0.0, 50.0), and 22.9(8.6, 55.7) minutes for junior and senior high school students, respectively. The detection rate of depressive symptoms was 46.1%. Among them, the detection rate of depressive symptoms of girls (49.8%) was higher than that of boys (42.3%), and that of senior high school students (53.2%) was higher than that of junior high school students (39.7%), and the differences were statistically significant ( χ 2= 18.35, 59.02, both P <0.01). Cochran-Armitage trend test revealed significant upward trends in depressive symptom detection rates with increasing usage of non educational digital media (social/chatting: Z =4.77; gaming: Z =3.24; video/music/novel: Z =7.30, all P <0.01). Logistic regression analysis showed that, compared to usage <1 h/d, using social/chatting digital media for 1-<2 h/d ( OR =1.66), 2-<3 h/d ( OR =1.80), and ≥3 h/d ( OR =2.68), gaming for 1-<2 h/d ( OR =1.48), 2-<3 h/d ( OR =1.90), and ≥3 h/d ( OR =2.93), and video/music/novel for 1-<2 h/d ( OR =1.76), 2-<3 h/d ( OR =2.00), and ≥3 h/d ( OR = 3.48) were all significantly associated with increased risks of depressive symptoms (all P <0.01). Conclusions Excessive use of non-educational digital media is a risk factor for depressive symptoms in adolescents. Regulating usage duration is beneficial for promoting adolescent mental health.