Background Adolescent depression is linked to daily activities like screen use, physical activity, and sleep. Most studies examine these factors separately. This study investigates the relationship between screen time, physical activity, sleep, and depressive symptoms in adolescents using compositional isotemporal substitution models (CISM). Methods A cross-sectional study collected data from 6,666 adolescents in Hefei, China, using self-administered questionnaires. Compositional Data Analysis was used to examine 24-h activity patterns and their association with depression. A compositional linear regression model was constructed to explore the relationships between screen time, low-intensity (LPA) and moderate-to-vigorous physical activity (MVPA), sleep (SLP), and adolescent depression. The CISM was then applied to analyze the effects of reallocating and substituting time between different activities on depression scores. Results The compositional linear regression model revealed a significant positive correlation between screen time and depression relative to the remaining activities (β ST = 0.693, P 0.001), while LPA, MVPA and SLP showed significant negative correlations with depression relative to the remaining activities (β LPA = −0.132, P 0.05, β MVPA = −0.293, P 0.001, β SLP = −0.981, P 0.001). Using a 10-min substitution as an example, replacing MVPA, LPA, or SLP with ST increased depression scores by 0.09, 0.06 and 0.03 points, respectively. Conversely, replacing ST with MVPA, LPA, or SLP decreased scores by 0.07, 0.05 and 0.03 points, respectively. Longer substitution durations amplified these effects. Conclusion The study highlights that reducing screen time and increasing physical activity or sleep can help alleviate depressive symptoms in adolescents.
Guo et al. (2026) studied this question.