JMIR Publications

ABSTRACT

Background:

In the digital era, mobile internet integration into daily routines presents a paradoxical relationship with mental health outcomes. While prior cross-sectional studies report inconsistent associations between mobile internet use (MIU) and depressive symptoms, the longitudinal mechanisms involving social support remain underexplored.

Objective:

This two-wave longitudinal study aims to examine the temporal relationships between mobile internet use (MIU), social support, and depressive symptoms among rural Chinese residents. Specifically, we hypothesize that (1) increased MIU will predict improved perceived social support over time, and (2) enhanced social support will mediate the relationship between MIU and reduced depressive symptoms. The findings are intended to inform digital health strategies that leverage online interactions to improve mental well-being in underserved communities.

Methods:

A two-wave longitudinal cohort study (4-year interval) was conducted among rural residents in Taierzhuang District, China (N=489 retained, mean age: 63.30 ± 13.34; 63.39% female). Multidimensional assessments included : 1) Demographic characteristics and MIU patterns via customized survey, 2) Social support was assessed using the Perceived Social Support Scale (PSSS), and 3) Depressive symptom severity via CES-D. Advanced analytical strategies were implemented: Least Absolute Shrinkage and Selection Operator (LASSO) regression for high-dimensional variable selection, complemented by cross-lagged panel modeling (CLPM) to disentangle temporal dependencies.

Results:

Our analysis found that the increase in mobile Internet use at baseline indicated the improvement of social support at follow-up (ρ = 0.097, 95% CI [0.001, 0.193]; P = 0.049). There is a bidirectional cross-lagged relationship between social support and depressive symptoms. An increase in baseline social support indicates a reduction in later depressive symptoms (ρ = -0.096, 95% CI [-0.183, -0.01]; P = 0.03). An increase in baseline depressive symptoms indicates a decrease in later social support (ρ = -0.109, 95% CI [-0.213, -0.004]; P = 0.041). And baseline social support has a significant impact on later depressive symptoms. Further analysis revealed that social support fully mediated the relationship between MIU and depressive symptoms. There was no direct effect between the two stages of MIU and depressive symptoms.

Conclusions:

This study used longitudinal data and developed cross-lagged models to improve the reliability of causal inferences and applied Lasso regression to improve the explanatory power and predictive accuracy of the models. The findings offer practical insights for designing digital mental health interventions, particularly in underserved rural areas. Specifically, our results support the development of mobile-based platforms that facilitate meaningful online social interactions to bolster perceived social support and thereby reduce depressive symptoms. We recommend that public health initiatives incorporate digital literacy training and promote online behaviors that strengthen real-life social connections. Furthermore, mental health professionals should consider individuals’ internet use patterns when designing personalized intervention strategies.