Vachuska Karl
Department of Sociology, University of Wisconsin-Madison, 1180 Observatory Drive, Madison, WI, 53703, USA.
SSM Popul Health. 2025 Jul 8;31:101838. doi: 10.1016/j.ssmph.2025.101838. eCollection 2025 Sep.
There are large and persistent racial inequalities in adverse birth outcomes in the United States. While much research presents "causal" effects of race on birth outcomes, such analyses are not counterfactually formulated and fail to inform policy in a meaningful way. Recent methodological advances in causal decomposition have developed double machine-learning-based approaches for understanding how a causal intermediate variable can potentially mitigate a gap between groups. Drawing on population-level birth data from California in 2018 and 2019, I consider the causal role of prenatal care timing in contributing to racial disparities in birth outcomes as measured by APGAR score and intrauterine growth restriction. My results suggest that prenatal care timing plays little to no role in contributing to racial inequalities in birth outcomes, indicating that research should focus on other factors driving disparities in neonatal health.
在美国，不良出生结局方面存在着巨大且持续的种族不平等现象。尽管许多研究呈现了种族对出生结局的“因果”影响，但此类分析并非基于反事实框架构建，无法以有意义的方式为政策提供参考。因果分解方面的最新方法进展开发出了基于双重机器学习的方法，用于理解因果中间变量如何可能缩小群体间的差距。利用2018年和2019年加利福尼亚州的人口层面出生数据，我考察了产前护理时机在导致以阿氏评分和宫内生长受限衡量的出生结局种族差异方面的因果作用。我的结果表明，产前护理时机在导致出生结局的种族不平等方面几乎没有起到作用，这表明研究应聚焦于驱动新生儿健康差异的其他因素。
