Dados Bibliográficos

Resumo

Introduction The study employed a robust network analysis methodology to assess the effects of race and comorbidities on birth outcomes, using a dataset of 443,902 mothers in Alabama from 2014 to 2021. Methods Four multimorbidity networks corresponding to White, Black, Asian, and American Indian and Alaska Native groups were constructed to explore distinct comorbidity patterns. The nodes in these networks represented various diseases, while the edges, quantified by the Salton Cosine Index, depicted the associations between these conditions. Additionally, two separate networks were analyzed for low birth weight (LBW) and normal birth weight (NBW) to identify the differential impacts of specific diseases. Feature selection methods including random forest and logistic regression were applied to pinpoint crucial intersections between the LBW and NBW networks, enhancing the granularity of the analysis. Results The findings indicated significant racial disparities in the density of comorbidity networks, with more complex disease interactions observed among Black, American Indian and Alaska Native, and Asian groups compared to Whites. Preexisting hypertension and eclampsia emerged as significant risk factors for LBW in White and Black groups, while gestational hypertension was prevalent across multiple racial groups. The LBW network displayed greater density than the NBW network, highlighting the intricate connections between comorbidities leading to adverse birth outcomes. Conclusion These insights underline the necessity for healthcare interventions tailored to the distinct health profiles of each racial group to effectively address and reduce maternal health disparities.

Referências Citadas