Abstract
The objective of this study is to determine how problematic social media use (PSMU), social media use frequency (SMU), and the number of followers (SMF) predict depressive symptoms evolution in adolescents, while evaluating the moderating effects of gender and age. Longitudinal data from 2,121 adolescents (52.38% female) were analysed at two time points (T1, T2) with a mean age (T1) of 13.79 years. Quantile regression models (τ = 0.25, 0.50, 0.75) revealed that problematic use and SMU were associated with increased depressive symptoms across all quantiles. Higher SMU predicted greater increases in depressive symptoms, particularly among younger adolescents. The interaction between SMF and SMU, also SMF and gender, was significantly associated with depressive outcomes; in particular, the effect of a higher number of followers was more pronounced among girls, and when combined with low use, it increased depressive symptoms. These findings suggest that SMU has a differential impact on depressive symptoms depending on the initial severity, with those exhibiting more severe baseline symptoms being more affected. Importantly, these results indicate that interventions for adolescent social media use may be tailored to their baseline mental health status, age, and gender; supporting current global regulatory trends aimed at increasing the age threshold for access to social media platforms.