Abstract
BACKGROUND: Chinese adolescents are navigating a complex digital environment amidst a growing mental health crisis. While health literacy is recognized as protective, previous research has primarily examined eHealth literacy (eHL) and mental health literacy (MHL) in isolation, leaving their synergistic effects and the integrative mechanism of mental eHealth literacy (MeHL) on social media addiction (SMA) and depression-anxiety-stress (DASS) largely unexplored. OBJECTIVE: This study aimed to investigate how eHL and MHL work through MeHL to reduce SMA and DASS in adolescents. We specifically tested whether having stronger MHL makes an adolescent's eHL more effective at building their overall MeHL. METHODS: A cross-sectional survey was conducted among 855 high school students (mean age 16.38, SD 0.86 years; 52.5%, 449/855 male) from 5 Chinese provinces, selected via stratified cluster random sampling. Participants completed validated scales for eHL, MHL, MeHL, SMA, and DASS. Data were analyzed using structural equation modeling (SEM) and Hayes' PROCESS macro (model 7) with cluster-robust SEs, controlling for age and gender. The significance of indirect and conditional effects was assessed using bias-corrected bootstrap CIs based on 5000 resamples. RESULTS: The results revealed distinct pathways. SEM path analysis indicated that both eHL (β=-0.152, 95% CI -0.270 to -0.034; P=.01) and MHL (β=-0.261, 95% CI -0.360 to -0.162; P<.001) were negatively associated with SMA. In contrast, only MHL demonstrated a significant negative association with DASS (β=-0.590, 95% CI -0.647 to -0.533; P<.001), whereas the link from eHL was nonsignificant (β=-0.011, 95% CI -0.102 to 0.080; P=.81). Bootstrapping tests confirmed that MeHL served as a significant mediator, with significant indirect paths from both eHL (SMA: β=-0.062, 95% CI -0.125 to -0.010, P=.01; DASS: β=-0.037, 95% CI -0.072 to -0.009, P=.008) and MHL (SMA: β=-0.045, 95% CI -0.090 to -0.008, P=.01; DASS: β=-0.027, 95% CI -0.053 to -0.007, P=.007). A key finding was the significant interactive association of eHL and MHL with MeHL (β=0.067, 95% CI 0.010-0.124; P<.05). This moderated mediation was supported, as the protective indirect paths from eHL (via MeHL) to both SMA and DASS were significantly stronger at higher levels of MHL. CONCLUSIONS: Unlike previous research that examined them separately, this study shows that eHL and MHL work best when combined. We found that MHL acts like a catalyst; it enhances the value of digital skills, helping teenagers become more competent at finding, understanding, and using online MeHL support effectively. In practice, this means integrated intervention programs that teach digital skills and mental health knowledge together are likely to be much more effective in fighting SMA and reducing distress than teaching either one alone.