A Causal Modelling Approach to Testing Longitudinal Associations Between Adolescent Internet Use and Symptoms of Depression

The extent to which digital media use by adolescents is a risk factor for poor mental health remains unclear. The purpose of the present study is to clarify the strength and direction of associations between adolescent internet use and the development of depression symptoms using a causal modelling approach. We also examine whether associations differ for boys and girls. Data are drawn from (N=1547) participants followed for the Quebec longitudinal Study of Child Development (QLSCD 1998-2020). Youth self-reported internet use in terms of the average hours of use per week at three age levels (ages 13, 15, and 17). Youth also self-reported depression symptoms at the same ages. After testing sex-invariance random intercepts, cross-lagged panel models stratified by sex, revealed that internet use by girls was associated with significant within-person (time-varying) change in depression symptoms. Girl’s internet use at age 13 was associated with increased depression symptoms at age 15 (ß=0.14) and internet use at age 15 increased depression at age 17 (ß=.11). For boys, internet use was not associated with significant time varying change in depression symptoms. The present findings support the hypothesis that internet use by adolescents can represent a significant and cumulative risk factor for the development of depressive symptoms, particularly in girls.