Chemical exposures have been linked to a variety of chronic diseases. Food contact materials are considered an important contributor to overall chemical exposure. In particular, prenatal and early life exposures to certain hazardous chemicals are considered of relevance for later development of chronic disease.

The U.S. National Children’s Study (NCS) investigates the effects of early life chemical exposure on disease development later in life. In a first step, the NCS Vanguard Study evaluates feasibility, study procedures and outcome assessments to be used in the NCS’ main study design. The present study published in the February 2014 issue of the peer-reviewed scientific journal Environmental Research presents results on phenol exposure in pregnant women derived from the NCS Vanguard Study (Mortensen et al. 2014).

In order to show the data’s representativeness, the researchers from the U.S. National Center for Environmental Health compared the results to data from non-pregnant women of the U.S. National Health and Nutrition Examination Survey (NHANES) 2009-2010 and pregnant women from NHANES 2005-2010. The phenolic compounds measured in the study include the food contact substances bisphenol A (BPA, CAS 80-05-7), triclosan (CAS 3380-34-5), benzophenone-3 (BP-3, CAS 131-57-7), methyl- and propyl parabens (MP and PP, CAS 99-76-8 and 94-13-3), which are all authorized for plastic food contact under Regulation EC 10/2011 in Europe and of which BPA, BP-3 and PP are also explicitly authorized as indirect food additives in the U.S..

Methods. Mortensen and colleagues measured urinary levels of BPA, triclosan, BP-3, MP, PP, and 2,4- and 2,5-dichlorophenols (CAS 120-83-2 and 583-78-8) in 506 pregnant women aged 18-49. The women from 40 study locations completed questionnaires on age, race/ethnicity, annual household income, education and study site. Spot urinary samples were taken in the third trimester and analyzed for phenols using solid-phase extraction-high performance liquid chromatography-isotope dilution-tandem mass spectrometry.

Figure 1: Detection frequency

Results were analyzed using multiple regression analysis including age, race/ethnicity, education, income, study site and log urinary creatinine as covariates.

Results. The researchers detected all phenolic compounds in the majority of women (Figure 1). Mean levels of BPA, MP and PP were 1.4, 103.5 and 19.1 µg/l (Figure 2). Whereas urinary BP-3 concentrations were highest in non-Hispanic white women (µ=72.0 µg/l; 95% CI: 57.0, 90.9), triclosan levels were highest in Hispanic women (33.9; 21.7, 63.0). The researchers could not assess the impact of demographic factors including education, income, or study site due to the limited sample size.

Similarly to the NCS results, NHANES 2009-2010 detected all phenols other than triclosan in more than 90% of women.

Figure 2: Mean phenolic compound levels

Figure 2: Mean phenolic compound levels

Results were generally similar, though mean urinary BP-3 concentrations were higher in NCS women (88.2 vs. 41.5 µg/l) and non-Hispanic white NCS women had higher BP-3 levels than NHANES women (89.6 vs. 38.0 µg/l). Non-Hispanic black NCS women had lower PP and MP and higher 2,4- and 2,5-dichlorophenol urinary concentrations than their NHANES counterparts (MP, 151.6 vs. 359.7 mg/l; PP, 25.3 vs. 57.0 mg/l; 2,4 dichlorophenol, 3.8 vs. 1.4 mg/l; 2,5-dichlorophenol, 89.2 vs.20.6 mg/l). The geometric means for the urinary phenols of pregnant women from NHANES 2005-2010 were generally similar to the data from the NCS cohort. NHANES 2005-2010 data further indicates a slight decline in urinary triclosan concentrations between 2005 and 2008, though sample size was too small to analyze individual ethnic groups.

The results found by Mortensen and colleagues generally coincide with the well-established NHANES cohorts. This indicates that the cohort is representative for pregnant women from the general population, as well as for women in general, irrespective of pregnancy status. The authors hypothesize that elevated BP-3 levels in non-Hispanic whites can be related to increased sunscreen use, whereas the lower BPA exposure in old order Mennonite women indicate a link between BPA body burden and reduced exposure to plastic and canned food. The authors conclude that while the NCS cohort can provide abundant information on pregnant women and is representative for the general U.S. population, the cohort is limited by pregnancy status and unavailability of exposure information. However, the main NCS study will overcome this pitfall, and may provide important information on early life exposure and disease development later in life.


Mortensen et al. (2014). “Urinary concentration of environmental phenols in pregnant women in a pilot study of the National Children’s Study.” Environmental Research 129, 32-38.