The Cumulative Risk of Chemical and Nonchemical Exposures on Birth Outcomes in Healthy Women: The Fetal Growth Study

• Published in: International journal of environmental research and public health Vol. 16; no. 19; p. 3700
• Main Authors: Zilversmit Pao, Leah, Harville, Emily W, Wickliffe, Jeffrey K, Shankar, Arti, Buekens, Pierre
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 01.10.2019; MDPI
• Subjects: Adult; Alcohol; Birth weight; Cadmium; Education; Ethnicity; Exposure; Female; Fetal Development; Fetuses; Gestational age; Health care facilities; Health risk assessment; Heavy metals; Humans; Infant, Newborn; Infant, Small for Gestational Age; Infants; Intrauterine exposure; Logistic Models; Male; Maternal Exposure - adverse effects; Mercury; Mercury - administration & dosage; Metals - administration & dosage; Mothers; Organic chemistry; PCB; Perinatal exposure; Polychlorinated biphenyls; Pregnancy; Pregnancy Outcome - epidemiology; Researchers; Risk analysis; Risk Factors; Secondary analysis; Socioeconomic Factors; Stress; Stress, Psychological - epidemiology; Studies; Women's Health; Womens health; nonchemical exposures; chemical exposures; cumulative risk; perinatal health
• ISSN: 1660-4601; 1661-7827; 1660-4601
• DOI: 10.3390/ijerph16193700

Metals, stress, and sociodemographics are commonly studied separately for their effects on birth outcomes, yet often jointly contribute to adverse outcomes. This study analyzes two methods for measuring cumulative risk to understand how maternal chemical and nonchemical stressors may contribute to small for gestational age (SGA). SGA was calculated using sex-specific fetal growth curves for infants of pregnant mothers ( = 2562) enrolled in the National Institute of Child Health and Human Development (NICHD) Fetal Growth Study. The exposures (maternal lead, mercury, cadmium, Cohen's perceived stress, Edinburgh depression scores, race/ethnicity, income, and education) were grouped into three domains: metals, psychosocial stress, and sociodemographics. In Method 1 we created cumulative risk scores using tertiles. Method 2 employed weighted quantile sum (WQS) regression. For each method, logistic models were built with three exposure domains individually and race/ethnicity, adjusting for age, parity, pregnancy weight gain, and marital status. The adjusted effect of overall cumulative risk with three domains, was also modeled using each method. Sociodemographics was the only exposure associated with SGA in unadjusted models ((odds ratio) OR: 1.35, 95% (confidence interval) CI: 1.08, 1.68). The three cumulative variables in adjusted models were not significant individually, but the overall index was associated with SGA (OR: 1.17, 95% CI: 1.02, 1.35). In the WQS model, only the sociodemographics domain was significantly associated with SGA. Sociodemographics tended to be the strongest risk factor for SGA in both risk score and WQS models.