Background: Throughout the COVID-19 pandemic, women carried, birthed, and cared for infants in a drastically changed world. For perinatal women, the sudden increase in stressors compounded an already vulnerable time where they are at an elevated risk of developing symptoms of psychopathology. Moreover, the pandemic exacerbated pre-existing racial health disparities and disproportionately impacted Black, Indigenous, and People of Color (BIPOC)— particularly perinatal BIPOC women, due to the intersection of their race and perinatal status.
This study investigated the relationships between COVID-19-related stressors and postpartum psychopathology using network analysis. Network analysis is used as an alternative technique for investigating the activation and maintenance of psychopathology and is increasingly used to examine the influence of external variables (e.g., stressors) on network dynamics. The relationship between psychological symptoms and stressors is typically examined in a unilinear manner—that is, stress causes psychopathology or vice versa. By using network analysis, we were able to investigate the bidirectional relationship between COVID-19-related stressors and postpartum psychopathology to reveal new insights into the individual stressor-symptom interactions that may underlie the emergence of psychological disorders for the perinatal population during the pandemic.
Methods: Participants (N=630) were recruited via social media and listservs and completed an online Qualtrics survey. Data quality measures were used to identify repeated, incomplete, and potentially fraudulent responses, which were removed prior to data analysis. Goldbricker, inter-item correlations, and variance inflation factor analyses were used to address topological overlap and identify statistically unique items to be included in the networks. A comorbidity symptom network was estimated to investigate the relationship between postpartum depression and anxiety symptoms in all participants. Bridge symptoms between the two conditions were identified using bridge analysis and clique percolation analysis. Next, an expanded model was estimated to investigate the relationship between postpartum symptoms and COVID-19-related stressors. Node-wise predictability and moderation analyses were used to investigate the effects of adding external variables (i.e., positive experiences, maternal functioning domains, and predictors of psychopathology) to the expanded model. Finally, moderated networks were estimated to investigate differences in the structure of the comorbidity network and the expanded network for mothers from different racial and ethnic groups.
Results: Fear-based symptoms were central in both the comorbidity and expanded networks and bridged postpartum anxiety and depression symptoms in the comorbidity network. The Depressed Mood and two Home Stress domains were central in the expanded network. Additional bridge symptoms in the comorbidity network included feeling overwhelmed, concentration difficulties, and feeling disliked by others, and in the expanded network included the Postpartum Stress, Emotional Stress, and Difficulty Adjusting domains. Moderation analyses revealed that the more mothers felt competent and the less challenging they perceived their infant’s temperament, the weaker the node connections were in their expanded networks. Furthermore, mothers with a history of prenatal depression, prenatal anxiety, or baby blues had denser expanded networks (i.e., stronger and more unique edges) compared to mothers with no history of these conditions. Contrary to expectations, moderation analyses revealed that: 1) social support and engaging in positive experiences during the pandemic strengthened connections between stressors and symptoms; 2) middle-income mothers had denser networks compared to low- and high-income mothers. Finally, racial network comparisons revealed that Black mothers' comorbidity and expanded networks were denser compared to all other racial groups.
Conclusions: Our findings highlight the influence of major contextual changes, such as the COVID-19 pandemic, on network dynamics—that is, previously established peripheral network nodes (e.g., fear) may shift to the center during large-scale events. Therefore, researchers cannot assume that previously identified central nodes will remain as the main drivers of psychopathology irrespective of changes in context, as this may lead to a misdirection of prevention and intervention efforts. Further, our findings underscore that people with multiple intersecting vulnerabilities may be disproportionately impacted by these major events.
| Identifer | oai:union.ndltd.org:columbia.edu/oai:academiccommons.columbia.edu:10.7916/srgs-sy57 |
| Date | January 2023 |
| Creators | Alhomaizi, Dalal |
| Source Sets | Columbia University |
| Language | English |
| Detected Language | English |
| Type | Theses |
Page generated in 0.002 seconds