Specific types of screen exposure and adolescent emotional and behavioral problems: Longitudinal associations and moderation by individual, family, and school characteristics

Objectives The rise in television and digital media use among adolescents has become a major concern, but the impacts of specific types of screen exposure on mental health remain underexplored. This study examined whether specific types of screen exposure are associated with emotional and behavioral problems (EBPs), and whether individual, family, and school factors moderate these associations. Study design Longitudinal study. Methods Participants (N = 10,770; mean age 15.89 years, 51.6 % male) completed standardized questionnaires at baseline and follow-up. EBPs were measured using the Strengths and Difficulties Questionnaire, and screen exposure and contextual factors were collected via self-report questionnaires. Generalized linear mixed-effects models, restricted cubic splines, interaction analyses, and sensitivity analyses were conducted. Results Specific screen types use, particularly educational apps and shopping/food delivery app exposures, were associated with increased total EBP difficulties. Each additional hour of social networking (OR, 1.04; 95 % CI, 1.01–1.07), short video viewing (OR, 1.05; 95 % CI, 1.01–1.09), or video gaming (OR, 1.06; 95 % CI, 1.01–1.10) was associated with an elevated risk of emotional symptoms. Short videos and video games were also associated with greater hyperactivity/inattention. Moderation analyses showed father's education, academic achievement, health status, and family function significantly modified these associations (e.g., family function × short video use: OR interaction = 0.89, P interaction = 0.014). Conclusions These findings highlight the differential impacts of screen exposure types on adolescent mental health and underscore the importance of addressing specific screen behaviors while considering individual, family, and school characteristics that may modify these risks, although the study's observational design limits causal inference.