Stress vulnerability and resilience in children facing COVID-19-related discrimination: A quasi-experimental study using polygenic, brain, and sociodemographic data

During the pandemic, perceived COVID-19-related discrimination aggravated children’s stress levels. The remaining question is to evaluate the individual variability in these effects and to identify vulnerable or resilient populations and why. Using the Adolescent Brain and Cognitive Development dataset ( N = 1,116) and causal machine learning approach – Generalized Random Forest, we examined the average and individual treatment effects of perceived discrimination on stress levels immediately and six months later. Their variability and key factors were also assessed. We observed significant variability in the acute effects of perceived discrimination across children and pinpointed the frontotemporal cortical volume and white matter connectivity (streamline counts) as key factors of stress resilience and vulnerability. The variability of these neurostructural factors partially originated from the environmental and genetic attributes. The finding was replicated in held-out samples ( N = 2,503). Our study has the potential for personalized prescriptive modeling to prevent children’s future psychopathology after the pandemic.