Trajectories of anxiety and depressive symptoms in high-risk children throughout the COVID-19 pandemic

• In Event: Family functioning and child mental health in the COVID-19 pandemic: Implications for risk, resilience, and recovery

Abstract

Introduction: The COVID-19 pandemic and resulting closures represents a large-scale traumatic event that has resulted in a number of ongoing family stressors, such as reduced economic resources, disrupted routines, and lack of childcare. Currently, little is known about how the pandemic has impacted child mental health. We investigate weekly fluctuations in anxiety and depression symptoms in high-risk children during the first 21 weeks of the COVID-19 pandemic in North Carolina.

Methods: Participants included 130 families with children 4-11 years old who previously participated in one of two research studies targeting low socioeconomic families and families of children at high risk for adversity. Caregivers completed at-home weekly surveys beginning one week after the stay-at-home order was issued in North Carolina (study site) during the COVID-19 pandemic. Mixed-effects regression ([MRM]/growth curve) models were conducted to assess patterns of child anxiety and depression symptoms over five months. Base models included: pre-pandemic family violence (FV) history, parental anxiety, parental depression, time (week) and covariates (child gender, recruitment group [0,1], child other symptom [i.e., anxiety symptoms when predicting depression, depression symptoms when predicting anxiety]). Interactions of time*pandemic-specific variables (e.g., parent employment changes; media consumption), and time*caregiver-specific variables (e.g., caregiver anxiety/depression symptoms, FV) were then tested in separate models to assess the impact of dynamic/time-variant experiences during the COVID-19 pandemic on anxiety/depression symptom trajectories in children.

Results: Child depression symptoms demonstrate a linear decrease over time (t = -2.37, p = .02), whereas time was not a significant predictor of anxiety symptom change (t = 1.26, p = .21). In COVID-specific MRM interaction models, parent employment change due to the pandemic was linked to change in depression (t = 2.72, p = .01), such that parental unemployment or inability to work full-time due to COVID was associated with higher child depression symptoms each week. Decreased childcare (t = -2.36, p =.02), increased child consumption of pandemic-related media coverage (t = -2.20, p = .03), and increased parent-child discussions about the dangers of COVID-19 (t = -2.44, p = .02) were all linked to increased child anxiety in interaction models. At the family/caregiver-level, time*caregiver ER (t = -2.44, p = .02) predicted depression symptom trajectories in youth. Additionally, family violence and parent depression, which have decreased linearly over time, demonstrated a main effect for child depression symptoms, such that higher FV (t = 4.66, p < .001) and parental depression symptoms (t = 8.90, p < .001) predicts greater overall symptoms, but not weekly variability.

Conclusion: Trajectories of child depression symptoms, while initially high, likely higher than baseline levels, have linearly decreased. In comparison child anxiety symptoms show no global trends and have remained variable over the five months following the initial COVID-19 pandemic lockdown in North Carolina. Results highlight possible specificity in pandemic- and family-specific variables that may predict depression and anxiety symptoms differentially in youth.

Authors

• Meredith Gruhn, University of North Carolina at Chapel Hill
  Presenting Author

• Laura Machlin, Harvard University
  Non-Presenting Author

• Summer Motton, University of North Carolina at Chapel Hill
  Non-Presenting Author

• Adam Bryant Miller, RTI International
  Non-Presenting Author

• Helen Milojevich, University of North Carolina at Chapel Hill
  Non-Presenting Author

• Margaret A. Sheridan, University of North Carolina at Chapel Hill
  Non-Presenting Author