directly assessed in the present study. Future research should examine whether the observed associations persist after accounting for these additional factors related to familial dysfunction. Seventh, despite the longitudinal design of the study, not all participants contributed data spanning the entire age range under investigation. Although we used a multilevel growth modeling approach to handle unstructured time data, we recognize that this method may not fully capture the substantial heterogeneity in individuals’ longitudinal course of HED across development. Finally, we conducted analyses separately by genetically inferred continental group (EA and AA). We recognize that both broad continental group and self-identified race are imprecise proxies for the complex sociocultural and environmental factors that may contribute to differences across groups. Thus, our findings should be interpreted with caution. Although this study does not explore the underlying sociocultural factors, such as experiences of discrimination or systemic stressors, investigating these factors in future research will be important for advancing health equity.
