The final sample included 491 women (n = 866 urine samples). Urinary BPA concentrations were available from 407 women at the first prenatal visit and 459 at the second prenatal visit; 375 women contributed BPA measurements at both prenatal visits. Demographic characteristics

were similar between women who provided one urine sample and women who provided two urine samples (data not http://www.selleckchem.com/products/jq1.html shown). Bilingual study staff conducted interviews in Spanish or English at each prenatal visit to collect maternal information on demographic characteristics, general dietary habits, and health. During the second prenatal visit, study staff also administered a modified version of the Block food frequency selleck chemical questionnaire to document participants’ dietary nutritional intake throughout the pregnancy (Harley et al., 2005). Spot urine samples were collected in polypropylene urine cups and aliquoted into glass vials. Samples were stored at − 80 °C until shipment to the CDC in Atlanta, GA for analysis. Concentration of total (free plus conjugated) species of urinary BPA was quantified

using automated online solid-phase extraction-high performance liquid chromatography-isotope-dilution tandem mass spectrometry using previously validated methods (Ye et al., 2005). Analytical runs included quality control (QC) samples (~ 3 μg/L and ~ 10 μg/L), which were analyzed with standards, blanks, and study samples. The coefficients of variation of repeated measurements of the QC materials ranged between 3.9 and 5.8%, depending on the concentration. An analysis of field blanks showed no detectable BPA contamination using our collection protocol; an analysis of reagent blanks indicated no BPA contamination during the laboratory sample processing. The limit of detection (LOD) was 0.4 μg/L. Concentrations below the LOD for which a signal was detected were reported as measured. Concentrations below the LOD with no signal detected were randomly imputed Phosphatidylinositol diacylglycerol-lyase based on a log-normal probability distribution

using maximum likelihood estimation (Lubin et al., 2004). Although some previous studies of BPA have accounted for urine dilution by adjusting urine concentrations by creatinine (Braun et al., 2011 and Calafat et al., 2008), this may not be appropriate particularly in populations undergoing rapid physiologic changes, such as pregnant women, due to high intra-individual variability in creatinine concentrations (Boeniger et al., 1993). Furthermore, as reported by Mahalingaiah et al. (2008), creatinine adjustment may not be appropriate for organic compounds such as BPA which are glucuronidated in the liver and eliminated by active tubular secretion. Other factors may also confound creatinine concentrations (e.g., muscularity, urine flow, age, exercise, diet, and diurnal variation) (Mahalingaiah et al., 2008).