Benavides Jaime, Cohen Jacob W, Chillrud Lawrence G, Cohen Gali, DeSerisy Mariah, Ramphal Bruce, Goldsmith Jeff, Kioumourtzoglou Marianthi-Anna, Margolis Amy E
Department of Epidemiology, Brown University School of Public Health, Providence, Rhode Island.
Department of Environmental Health Sciences, Columbia University Mailman School of Public Health, New York, New York.
Environ Epidemiol. 2025 Sep 5;9(5):e417. doi: 10.1097/EE9.0000000000000417. eCollection 2025 Oct.
Prospective studies suggest that prenatal exposure to chemical neurotoxicants and maternal stress increase risk for psychiatric problems. However, most studies have focused on childhood outcomes, leaving adolescence-a critical period for the emergence or worsening of psychiatric symptoms-relatively understudied. The complexity of prenatal coexposures and adolescent psychiatric comorbidities, particularly among structurally marginalized populations with high exposure burdens, remains poorly understood.
We aimed to investigate correlated features in prenatal chemical and psychosocial exposures and in adolescent behavioral outcomes in a sample of economically disadvantaged urban youth.
Principal Component Pursuit, a pattern recognition technique, coupled with factor analysis identified profiles of prenatal chemical and psychosocial exposures and of adolescent behavioral outcomes. Linear regression tested associations between exposure and behavioral profiles, accounting for potential confounders.
Three prenatal exposure profiles were identified: one combining exposure to air pollutants, polycyclic aromatic hydrocarbons-DNA adducts, and psychosocial stressors, and two others separating exposure to endocrine-disrupting chemicals by molecular weight. Three adolescent behavioral profiles reflected two phenotypes: one characterized by attention problems only; two others characterized by attention, thought, substance use, and self-control problems, differentiated by informant (self vs. parent). The prenatal exposure profile involving air pollution and stress was associated with the phenotype of self-reported attention, thought, substance use, and self-control problems ( = 0.21, 95% confidence interval = 0.02, 0.40).
Principal Component Pursuit identified consistent patterns of multidimensional prenatal exposures and adolescent behavioral symptoms. Results support prior studies linking prenatal exposures associated with economic disadvantage to complex, self-reported transdiagnostic psychiatric problems in adolescence.
前瞻性研究表明，产前接触化学神经毒物和母亲压力会增加患精神疾病的风险。然而，大多数研究都集中在儿童期的结果上，而对青春期（精神症状出现或恶化的关键时期）的研究相对较少。产前共同暴露和青少年精神疾病共病的复杂性，尤其是在暴露负担高的结构性边缘化人群中，仍然知之甚少。
我们旨在调查经济弱势城市青年样本中产前化学和心理社会暴露以及青少年行为结果的相关特征。
主成分追踪法（一种模式识别技术）与因子分析相结合，确定了产前化学和心理社会暴露以及青少年行为结果的概况。线性回归测试了暴露与行为概况之间的关联，并考虑了潜在的混杂因素。
确定了三种产前暴露概况：一种是将接触空气污染物、多环芳烃 - DNA加合物和心理社会压力源结合在一起，另外两种是按分子量将接触内分泌干扰化学物质分开。三种青少年行为概况反映了两种表型：一种仅以注意力问题为特征；另外两种以注意力、思维、物质使用和自我控制问题为特征，根据报告者（自我报告与家长报告）进行区分。涉及空气污染和压力的产前暴露概况与自我报告的注意力、思维、物质使用和自我控制问题的表型相关（β = 0.21，95%置信区间 = 0.02，0.40）。
主成分追踪法确定了多维产前暴露和青少年行为症状的一致模式。结果支持了先前的研究，这些研究将与经济劣势相关的产前暴露与青少年复杂的、自我报告的跨诊断精神疾病问题联系起来。
