This longitudinal investigation examined reciprocal, process-oriented links between parent–adolescent closeness and adolescent depressive symptoms across three time points spanning the pre-pandemic, pandemic outbreak, and post-pandemic periods. The research used waves from the China Family Panel Study collected in 2018 (pre-pandemic), 2020 (pandemic outbreak), and 2022 (post-pandemic) to assess how changes in closeness and depression predicted one another over time.
The analysis included 1,780 adolescents: 841 girls (Mage = 14.99 years in 2018, SD = 2.04; 87.2% Han ethnicity) and 939 boys (Mage = 14.89 years, SD = 1.99; 87.1% Han ethnicity). Data were drawn from three waves covering the years 2018, 2020, and 2022, allowing the authors to study pre-to-post pandemic trajectories and reciprocal associations between relationship quality and internalizing symptoms.
The study focused on self-reported indicators of parent–adolescent closeness and adolescent depressive symptoms. To model the dynamic, bidirectional processes between these constructs across the three waves, researchers used bivariate latent change score models. This approach permits testing whether prior levels of one construct predict subsequent change in the other while accounting for prior change and stability.
Across models and regardless of the gender of parent or adolescent, lower prior levels of parent–adolescent closeness consistently predicted greater subsequent increases in adolescent depressive symptoms. This pattern held across the pre-pandemic, pandemic, and post-pandemic measurement points, suggesting an enduring protective effect of closeness on adolescents’ emotional well-being across a period of heightened environmental stress.
Reciprocal effects from adolescent depressive symptoms to later changes in parent–adolescent closeness were present but gender-specific. Prior higher adolescent depression predicted greater subsequent decreases in closeness only in the father–son relationship. That is, when boys reported higher depressive symptoms at an earlier wave, closeness with fathers tended to decline at follow-up.
The models revealed differentiated gendered patterns. The robust finding that lower closeness predicts later increases in depression was consistent across parent and adolescent gender. In contrast, the reverse pathway—where adolescent depression predicted later reductions in closeness—was observed uniquely for boys in relation to their fathers. The article’s plain-language summary also noted that girls who reported higher depression later reported greater closeness with mothers, indicating divergent parent-response patterns by child gender.
The authors reported that some effects were moderated by developmental age. The abstract indicates that developmental age moderated certain pathways, but the source summary does not provide detailed parameter estimates or specify which pathways were moderated or the direction and magnitude of moderation. For those full details, readers should consult the original article.
Taken together, the findings underscore that the quality of parent–adolescent relationships matters for adolescent mental health across the pandemic period. Lower perceived closeness with parents predicted worsening depressive symptoms over time, highlighting the protective role of supportive family ties. The observed decline in father–son closeness following higher levels of boys’ depression points to a potential relational vulnerability: fathers may pull away when sons are struggling, whereas mothers may move closer, particularly with daughters.
Clinically, these results suggest the value of family-focused approaches that foster and maintain parent–adolescent closeness, and of attention to father–son dynamics when boys present with depressive symptoms. Interventions that engage parents, encourage supportive responses to adolescent distress, and monitor relational strain may help buffer adolescents during prolonged stressors such as a pandemic.
The summary available in the source provides key sample descriptors, analytic method, principal findings, and a plain-language summary, but it does not enumerate all study limitations or full statistical details in the abstract. The authors declared no conflicts of interest. For full methodological specifics (measures, model fit indices, effect sizes, and detailed moderation results), readers should consult the full published article.