Testing Bidirectional Associations Between Maternal and Child Depression During Emerging Adolescence.

Importance: Research has consistently shown links between parent and child depression. The prevailing assumption is that parent depression precedes the onset, persistence, and even exacerbation of child depression. However, it is plausible, albeit infrequently tested, that child depression triggers subsequent parent depression. Clarifying the direction and developmental timing of these associations is critical for designing and advocating for family-centered approaches to pediatric care. Objective: To estimate the longitudinal bidirectional associations between maternal and child depression from middle childhood through emerging adolescence during a period that coincided with the COVID-19 pandemic, and to assess whether these associations are moderated by key sociodemographic factors. Design, Setting, and Participants: This study used data from the All Our Families (Calgary, Alberta, Canada) cohort across multiple waves during and beyond the COVID-19 pandemic when children were aged 10.3 (May 20 to July 15, 2020), 10.9 (March 4 to April 30, 2021), 11.6 (November 22, 2021 to January 17, 2022), and 12.8 (January 16 to July 7, 2024) years. Data were analyzed between March 1 and August 31, 2024, using random-intercept cross-lagged panel models. Exposures: Maternal and child report of depressive symptoms. Main Outcomes and Measures: The Behavior Assessment System for Children for child depression and the Center for Epidemiologic Studies Depression Scale-10 for maternal depression. Results: The sample included 1801 mother-child dyads (52% boys, 48% girls). Mothers were a mean (SD) 41.6 (4.4) years old at study entry and most had completed postsecondary education (80%), had an annual income more than CAD $100 000 (75%), and were married or in common-law relationships (71.4%). Consistent cross-sectional correlations were observed, reflecting stable between-participant associations for maternal depression and child depression throughout the study period. Within-participant increases in child depression scores at ages 10.3 and 10.9 years were associated with subsequent increases in maternal depression scores at child ages 10.9 (standardized coefficient 0.12; 95% CI, 0.02-0.22) and 11.6 (0.17; 95% CI, 0.07-0.26) years; however, this was not present for maternal depression. These patterns of associations were moderated by household income (difference test for χ212 = 23.0; P = .03) and within-participant increases in child depression were consistently associated with subsequent increases in maternal depression for the higher income group. Conclusions and Relevance: Contrary to prevailing assumptions, these findings suggest that children's depression over time may have contributed to worsening maternal depression, rather than the other way around. While these results should be replicated in nonpandemic contexts to confirm their generalizability, they highlight the need for family-centered approaches to mental health care.