Association of Parental Preconception Exposure to Phthalates and Phthalate Substitutes With Preterm Birth

Key Points español 中文 (chinese)

Question  Are paternal and maternal preconception urinary metabolite concentrations of phthalates and phthalate substitutes associated with singleton preterm birth?

Findings  In this preconception cohort study of 420 singleton infants, maternal preconception urinary concentrations of di-(2-ethylhexyl) phthalate and di(isononyl)cyclohexane-1,2-dicarboxylate (a nonphthalate plasticizer substitute) were associated with an increased risk of preterm birth. Fathers’ preconception concentrations of these biomarkers were not associated with preterm birth in offspring.

Meaning  Female exposure to select plasticizers during the preconception period may be a potential risk factor for adverse pregnancy outcomes.

Importance  Although phthalate exposure during pregnancy has been associated with preterm birth, the association of preconception exposure in either parent with preterm birth constitutes a knowledge gap.

Objective  To examine the association of paternal and maternal preconception urinary concentrations of biomarkers of phthalates and phthalate substitutes with singleton preterm birth.

Design, Setting, and Participants  This study, conducted at an academic fertility center in Boston, Massachusetts, included a prospective preconception cohort of subfertile couples comprising 419 mothers and 229 fathers and their 420 live-born singleton offspring born between January 1, 2005, and December 31, 2018. Statistical analysis was performed from August 1 to October 31, 2019.

Exposures  Urinary concentrations of metabolites of phthalates and phthalate substitutes obtained before conception.

Main Outcomes and Measures  Gestational age was abstracted from delivery records and validated using the American College of Obstetricians and Gynecologists guidelines for births after medically assisted reproduction. The risk ratio (RR) of preterm birth (live birth before 37 completed weeks’ gestation) was estimated in association with urinary concentrations of 11 individual phthalate metabolites, the molar sum of 4 di-(2-ethylhexyl) phthalate (ΣDEHP) metabolites, and 2 metabolites of 1,2-cyclohexane dicarboxylic acid diisononyl ester (DINCH, a nonphthalate plasticizer substitute) using modified Poisson regression models adjusted for covariates.

Results  The mean (SD) age of the 419 mothers was 34.7 (4.0) years, the mean (SD) age of the 229 fathers was 36.0 (4.5) years, and the mean (SD) gestational age of the 420 singleton children (217 boys) was 39.3 (1.7) weeks, with 34 (8%) born preterm. In adjusted models, maternal preconception ΣDEHP concentrations (RR, 1.50; 95% CI, 1.09-2.06; P = .01) and cyclohexane-1,2-dicarboxylic acid monohydroxy isononyl ester (MHiNCH, a metabolite of DINCH) concentrations (RR, 1.70; 95% CI, 0.89-3.24; P = .11) were associated with an increased risk of preterm birth. After additional adjustment for prenatal ΣDEHP or MHiNCH concentrations, the association of maternal preconception exposure to ΣDEHP and preterm birth remained robust (RR, 1.69; 95% CI, 1.17-2.44; P = .006), while the association of maternal preconception exposure to MHiNCH and preterm birth was attenuated (RR, 1.17; 95% CI, 0.49-2.81; P = .72). The remaining urinary metabolites examined in either parent showed no association with preterm birth.

Conclusions and Relevance  In this prospective cohort of subfertile couples, maternal preconception exposure to ΣDEHP metabolites was associated with an increased risk of preterm birth. The results suggest that female exposure to select phthalate plasticizers during the preconception period may be a potential risk factor for adverse pregnancy outcomes, which may need to be considered in preconception care strategies.

Preterm birth is the factor most strongly associated with neonatal mortality and long-term morbidity globally.^1-3 In the United States, 1 in 10 pregnancies is delivered preterm, accounting for approximately 380 000 births per year.⁴ Beyond the increased risk of early death, the long-term consequences among surviving infants may be associated with neurologic, respiratory, and gastrointestinal disorders as well as cardiometabolic disease during adulthood.^1,5-8

Preterm birth is a complex and heterogeneous condition with multiple etiopathogenic processes triggering early parturition.^3,9 Although some risk factors for preterm birth have been identified, including maternal age, race/ethnicity, socioeconomic status, smoking during pregnancy, infection, and multiple gestations, these factors account for less than half of all cases, and underlying mechanisms remain largely unknown.^1,3,10-13 There is increasing evidence of an association between environmental exposures during pregnancy (including air pollution and chemicals such as phthalates) and preterm birth.^14-20

Phthalates are a family of chemicals widely used in many consumer products. Phthalates are known reproductive and developmental toxicants in experimental animals^21,22 and are suspected to produce similar effects in humans.^23-25 Human exposure to phthalates is ubiquitous in the United States, Europe, and elsewhere.^26-28 Regulation of some phthalates has prompted the use of plasticizer replacement chemicals such as 1,2-cyclohexane dicarboxylic acid diisononyl ester (DINCH). Substitution of phthalates with DINCH warrants further screening in human populations²⁹ because its metabolites are biologically active and understudied.³⁰

The maternal preconception period remains an important but largely unexplored critical window of exposure for perinatal and infant outcomes.³¹ Even less is known about the association of paternal preconception exposures with offspring health.³² Environmental-epigenetic mechanisms in the preconception and periconception period are likely associated with the etiopathologic characteristics of preterm birth.^33-35 However, studies addressing the association of parental preconception exposure to phthalates with outcomes in gametes, fertilization, implantation, placentation, and gestation are limited.³¹ Therefore, we aimed to investigate whether higher paternal and maternal preconception urinary concentrations of metabolites of phthalates and phthalate substitutes were associated with an increased risk of preterm birth among couples undergoing fertility care.

The Environment and Reproductive Health (EARTH) Study is an ongoing prospective preconception cohort of couples seeking fertility evaluation and medically assisted reproductive treatment at the Massachusetts General Hospital Fertility Center. The EARTH Study was designed to investigate environmental and nutritional factors for both women and men across preconception and prenatal periods in association with fertility, pregnancy, and birth outcomes. The cohort has been described elsewhere.³⁶ In brief, women aged 18 to 46 years and men aged 18 to 55 years, using their own gametes, were eligible. Participants enrolled independently or as a couple and were followed up from study entry through their fertility care, pregnancy, and labor and delivery. The present analysis included 419 female and 229 male EARTH cohort participants who gave birth to a singleton infant between January 1, 2005, and December 31, 2018, for whom we had at least 1 urine sample quantified for biomarker metabolites during the period before conception of the index pregnancy. One singleton live birth was from a male participant enrolled without a female partner, thus leaving 228 couples (Figure). Trained staff explained the study details to participants and answered questions before obtaining written informed consent. The study was approved by the Massachusetts General Hospital, Harvard T.H. Chan School of Public Health, and the Centers for Disease Control and Prevention Institutional Review Boards. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Male and female participants provided 1 spot urine sample at study entry. Women provided up to 2 additional spot urine samples per fertility treatment cycle: one obtained during the follicular phase of the cycle (days 3-9) and the other on the day of the fertility procedure. Men provided an additional spot urine sample per cycle on the day when their female partner underwent the fertility procedure. Women also provided 1 spot urine sample per trimester at a median of 6, 21, and 35 weeks’ gestation. We used the biomarker concentrations from the multiple urine samples obtained per participant from study entry up to and including the samples from the treatment cycle of conception of the index birth to estimate the mean exposure in the preconception window.

Urine samples were collected in polypropylene specimen cups, and the specific gravity (SG) of each sample was quantified with a handheld refractometer (National Instrument Company Inc). The urine samples were then divided into aliquots, frozen for long-term storage at −80 °C, and shipped on dry ice overnight to the Centers for Disease Control and Prevention (Atlanta, Georgia). For each urine sample, the concentrations of metabolites of phthalates and phthalate substitutes were quantified using solid-phase extraction coupled with high-performance liquid chromatography–isotope dilution tandem mass spectrometry.³⁷ The concentrations of the following phthalate metabolites were measured: monoethyl phthalate, mono-n-butyl phthalate (MBP), mono-isobutyl phthalate (MiBP), monobenzyl phthalate (MBzP), mono(2-ethylhexyl) phthalate (MEHP), mono(2-ethyl-5-hydroxyhexyl) phthalate (MEHHP), mono(2-ethyl-5-oxohexyl) phthalate (MEOHP), mono(2-ethyl-5-carboxypentyl) phthalate (MECPP), mono(3-carboxypropyl) phthalate, monocarboxyisooctyl phthalate, and monocarboxyisononyl phthalate. The urinary concentrations of 2 DINCH metabolites, cyclohexane-1,2-dicarboxylic acid monohydroxy isononyl ester (MHiNCH) and cyclohexane-1,2-dicarboxylic acid monocarboxyisooctyl ester (MCOCH), were also measured in a subset of participants (Figure). The limits of detection (LOD) ranged from 0.1 to 1.2 ng/mL. Concentrations below the LOD were assigned the LOD divided by the square root of 2.³⁸

We calculated the molar sum of 4 di(2-ethylhexyl) phthalate (DEHP) metabolites by dividing each metabolite concentration by its molecular weight and then summing: ΣDEHP = {[MEHP × (1/278.34)] + [MEHHP × (1/294.34)] + [MEOHP × (1/292.33)] + [MECPP × (1/308.33)]}. We multiplied the molar sum by the molecular weight of MECPP (308.33) to convert ΣDEHP to nanograms per milliliter. We also calculated a summary measure of phthalate metabolites with antiandrogenic properties (ie, MEHP, MEHHP, MEOHP, MECPP, MBP, MiBP, and MBzP), as previously described.³⁹ The summary estimate (ΣAAPhthalates) was calculated by multiplying the SG-adjusted concentration of each of these 7 individual metabolites by their antiandrogenic potency and summing the weighted concentrations: ΣAAPhthalates = MBP + (0.24 × MiBP) + (0.26 × MBzP) + (0.61 × MEHP) + (0.61 × MEHHP) + (0.61 × MEOHP) + (0.61 × MECPP).^39,40

We abstracted gestational age in days from delivery records and validated it using the American College of Obstetricians and Gynecologists guidelines for dating births after medically assisted reproduction.⁴¹ The fertility treatment setting permitted us to estimate gestational age with high accuracy using in vitro fertilization protocol dates, substantially reducing the number of misclassifications of preterm births due to inaccuracies in pregnancy dating.⁴² For in vitro fertilization pregnancies, the gestational age was estimated as (Outcome Date − Transfer Date + 14 Days + Cycle Day of Transfer).⁴¹ For intrauterine insemination and nonmedically assisted or naturally conceived pregnancies, we used the birth date minus the cycle start date or the last menstrual period date. Preterm birth was defined as any live birth less than 37 completed weeks’ gestation (<259 days). We corrected for 3 pregnancies for which the medical delivery record estimates (criterion standard) differed from the American College of Obstetricians and Gynecologists–based estimates by more than 6 days through additional delivery record verification.

Data on paternal and maternal age, educational level, race/ethnicity, and smoking status were obtained from self-reported questionnaires administered at enrollment. Research study staff measured the height and weight of the participants at baseline, and body mass index was calculated as weight in kilograms divided by height in meters squared. The treating infertility physician diagnosed the underlying cause of infertility using the Society for Assisted Reproductive Technology (ART) definitions.^43,44 The type of medically assisted reproduction used in the conception cycle of the index birth was abstracted from the electronic medical records by trained study staff and dichotomized as ART procedures (all in vitro fertilization protocols, including intracytoplasmic sperm injection) vs non-ART protocols (all intrauterine insemination or ovarian stimulation protocols as well as nonmedically assisted or naturally conceived).

Statistical analysis was performed from August 1 to October 1, 2019. To account for urinary dilution, each biomarker concentration was multiplied by [(SGp − 1)/(SGi − 1)], where SGi is the SG of the participant’s sample and SGp is the mean SG for all male (mean, 1.016) or all female (mean, 1.015) participants included in the study.⁴⁵ The SG-adjusted biomarker concentrations were natural log-transformed to standardize the distribution and reduce the effect of extreme values. We estimated the geometric mean paternal and maternal preconception biomarker concentrations by averaging each participant’s natural log concentration obtained from study entry (baseline) and at each treatment cycle up to and including the cycle of the index conception of the singleton offspring. We calculated descriptive statistics for biomarker concentrations and the percentage of values below the LOD, as well as Spearman correlation coefficients for each natural log concentration between couples (paternal vs maternal preconception and preconception vs prenatal windows, using the mean concentration across 3 trimesters).

We examined the clinical and demographic characteristics, reported as mean (SD) values or as numbers and percentages, of study participants in the total cohort. We fit modified Poisson regression models to evaluate the association of continuous urinary biomarker concentrations with dichotomous preterm birth outcomes.⁴⁶ Modified Poisson models were fit by using a log-link function with a Poisson distribution to yield estimated risk ratios (RRs) and 95% CIs for preterm birth for every natural log unit increase in metabolite concentration. We fit a separate model for each of the 13 individual biomarkers of interest as well for the ΣDEHP and ΣAAPhthalates summary measures.

We selected covariates a priori as potential confounders based on substantive knowledge using a directed acyclic graph (eFigure in the Supplement) and examined unadjusted and covariate-adjusted results. All statistical models were adjusted for ART vs non-ART to control for mode of conception and indirectly for the underlying cause of infertility. Maternal preconception window covariate-adjusted models included maternal age and body mass index (continuous), maternal educational level (<college, college, or graduate degree), smoking status (never smoked or ever smoked, defined as a current or former smoker), and treatment type (ART or non-ART). Paternal preconception window covariate-adjusted models included paternal and maternal age and body mass index (continuous), paternal and maternal smoking (ever or never), maternal educational level (<college, college, or graduate degree), and treatment type (ART or non-ART). We further adjusted for partner’s preconception phthalate concentrations and maternal prenatal phthalate concentrations (averaged natural log concentrations across 3 trimesters), in additional covariate-adjusted models. All statistical analyses were performed with SAS, version 9.4 (SAS Institute Inc). Statistical tests were 2-tailed, and P < .05 was considered statistically significant.

First, we restricted the maternal preconception analyses to 228 couples to obtain more comparable results across models within couples. Second, to examine potential differences by infant sex, we stratified analyses and estimated sex-specific RRs and 95% CIs. Effect-measure modification P values were calculated for the interaction term (sex × urinary biomarker concentration). P < .20 was considered potential evidence of effect modification by infant sex on the multiplicative scale. Third, because our analyses on DINCH metabolites MHiNCH and MCOCH were limited by low detection rates, we dichotomized these parental preconception biomarkers by their median values and fit modified Poisson regression models for dichotomous DINCH metabolite concentrations and preterm birth, adjusting for covariates. We also fit multivariable general linear regression models for continuous parental preconception concentrations of MHiNCH and MCOCH with continuous gestational age to address concerns of lower power given the smaller sample size and therefore smaller number of cases of preterm birth in this subset in which DINCH biomarkers were measured. Coefficient estimates and 95% CIs represent the mean difference in gestational age for each natural log unit increase in urinary DINCH biomarker concentration. To assess the linearity assumption of our positive maternal DEHP results, we fit models across biomarker quartiles and estimated P values for trend across quartiles.

The study cohort included 419 mothers (mean [SD] age, 34.7 [4.0] years; mean [SD] body mass index, 24.1 [4.3]) and 229 fathers (mean [SD] age, 36.0 [4.5] years; mean [SD] body mass index, 27.7 [6.1]) (228 couples) at the time of enrollment (Table 1). Among the 420 singleton infants, the mean (SD) gestational age was 39.3 (1.7) weeks, with 34 infants (8%) born preterm (Table 2). The mean (SD) birth weight was 3363 (551) g, with 20 infants (5%) born with low birth weight (<2500 g).

In total, 1700 maternal preconception urine samples and 590 paternal preconception urine samples were analyzed for phthalate and DINCH metabolites. Women provided a mean (SD) of 4.1 (3.0) urine samles (median, 3 urine samples; interquartile range, 2-5 urine samples), and men provided a mean (SD) of 2.6 (1.7) urine samples (median, 2 urine samples; interquartile range, 1-3 urine samples). The distribution of metabolites, detection frequencies, and correlations can be found in eTable 1 and eTable 2 in the Supplement.

After adjusting for covariates, we found that maternal preconception urinary ΣDEHP metabolite concentrations were associated with increased risk of preterm birth (RR, 1.50; 95% CI, 1.09-2.06; P = .01). Risk ratios increased slightly in models accounting for maternal prenatal ΣDEHP concentrations (RR, 1.69; 95% CI, 1.17-2.44; P = .006) (Table 3). This association appeared to be stronger for male infants (RR, 2.01; 95% CI, 1.17-3.45) compared with female infants (RR, 1.22; 95% CI, 0.79-1.88) (effect-measure modification P = .17) (eTable 3 in the Supplement). Quartile analysis showed a positive dose-response association between maternal preconception urinary DEHP concentrations and preterm birth (eTable 4 in the Supplement). There was some suggestion that mothers with higher preconception urinary MHiNCH concentrations had an increased risk of preterm birth in the main covariate-adjusted model (RR, 1.70; 95% CI, 0.89-3.24; P = .11) (Table 3). This association was attenuated by additional adjustment for prenatal MHiNCH concentrations (RR, 1.17; 95% CI, 0.49-2.81; P = .72). In sensitivity analyses, maternal preconception MHiNCH concentrations above the median were associated with a suggested increased risk of preterm birth (RR, 4.02; 95% CI, 0.84-19.30; P = .08). Maternal preconception MHiNCH concentrations were associated with reduced gestational age (β = −2.01 days; 95% CI, −3.74 to −0.29 days; P = .02) (eTable 6 in the Supplement). No other maternal preconception metabolites were associated with the risk of preterm birth (Table 3).

Paternal urinary ΣDEHP metabolite concentrations were associated with an increased risk of preterm birth in covariate-adjusted models (RR, 1.41; 95% CI, 0.94-2.11; P = .09) (Table 4). However, this association was markedly attenuated toward the null in models accounting for maternal preconception ΣDEHP concentrations (RR, 1.06; 95% CI, 0.66-1.68; P = .82). The results of MHiNCH and MCOCH models in association with preterm birth were imprecise owing to low power. In sensitivity analyses, paternal preconception MHiNCH or MCOCH concentrations were not associated with continuous gestational age (eTable 6 in the Supplement). The remaining paternal preconception biomarkers showed little evidence of an association with preterm birth (Table 4).

In analyses restricted to 228 couples, associations of maternal preconception ΣDEHP metabolite concentrations with preterm birth remained robust in covariate-adjusted models (RR, 2.30; 95% CI, 1.46-3.60; P < .001), as well as in models additionally adjusting for prenatal (RR, 4.98; 95% CI, 2.31-10.75; P < .001) or paternal preconception (RR, 2.37; 95% CI, 1.39-3.70; P = .001) ΣDEHP concentrations (eTable 5 in the Supplement). Maternal preconception MHiNCH concentrations were associated with an increased risk of preterm birth among couples in an unadjusted model (RR, 3.48; 95% CI, 0.91-13.36; P = .07); however, associations became imprecise in models adjusting for covariates (RR, 3.15; 95% CI, 0.23-43.95; P = .39) (eTable 5 in the Supplement).

In this prospective cohort of subfertile couples, urinary ΣDEHP metabolite concentrations measured in mothers before conception were associated with a higher risk of singleton preterm birth. The results of ΣDEHP models were robust to adjustments for prenatal exposure. This association was more pronounced among male infants than female infants. Couple-based analyses confirmed the results for an association between maternal preconception ΣDEHP concentrations and increased risk of preterm birth. Maternal preconception MHiNCH concentrations were suggestively associated with an elevated risk of preterm birth. These results were further confirmed in a sensitivity analysis examining gestational age continuously. We found that a log-unit increase in maternal MHiNCH concentrations was associated with a reduction in gestational age by approximately 2 days. However, this association was partially explained by prenatal MHiNCH concentrations and should be interpreted cautiously owing to the small numbers and low detection frequencies. Future studies should confirm or rule out a potential association with this emerging phthalate substitute. We observed little evidence of associations between paternal preconception phthalate metabolites or biomarkers of plasticizer substitutes and preterm birth.

To our knowledge, this is the first study evaluating couples’ exposure to phthalate metabolites during the preconception window and its association with preterm birth. Previous human studies of preterm birth have assessed phthalate exposure during the prenatal window. Our maternal preconception ΣDEHP findings are compatible with most prior research,^{14,16,17,47,48} although not all,^49,50 on prenatal phthalate exposure. Ferguson and colleagues^14,17 used a nested case-control design and reported robust dose-response associations of ΣDEHP and MBP with increased odds of overall preterm birth and spontaneous preterm birth among North American women. A case-control study by Meeker et al¹⁶ also found a significantly increased risk of preterm birth in association with urinary concentrations of ΣDEHP and MBP among Mexican women. Gao et al⁴⁷ reported positive associations between prenatal DEHP metabolites in Chinese mothers and preterm birth. In contrast, Ferguson et al⁵⁰ found no association between prenatal DEHP metabolites and preterm birth in a recent analysis among Puerto Rican mothers, whereas Adibi et al⁴⁹ reported inverse associations between prenatal DEHP metabolites and preterm birth among North American mothers. In a recent systematic review of the literature, Radke et al¹⁸ concluded that exposure to DEHP was associated with the risk of preterm birth, with a moderate level of evidence. In addition, studies have evaluated prenatal phthalates in association with continuous gestational age with heterogeneous epidemiologic methods; some reported shorter gestations,^51-54 while others found inconsistent^55,56 or contrasting^49,57 results.

Our findings support a novel hypothesis: maternal phthalate exposure during the critical period before conception may be associated with shorter gestation. Although, to our knowledge, mechanistic data on preconception exposures are scarce, this latent association could be compatible with the established association of DEHP with the ovary and its related epigenetic modifications in oocytes.^22,23,58,59 In addition, DEHP metabolites may disrupt nuclear receptors, including peroxisome proliferator–activated receptors, the androgen receptor, and estrogenic receptors,^25,60-62 and they may increase oxidative stress and inflammation in the ovary and endometrium.^63-65 It is accepted that early disruption to peri-implantation processes such as alterations in embryo spacing or development, decidualization, and placentation may perpetuate throughout pregnancy, manifesting later as preterm birth, among other adverse outcomes.⁶⁶ Based on previous results from the EARTH Study team and those of others, including fertile populations, we hypothesize that an early action of DEHP metabolites at the ovary may interfere with normal fertility and implantation processes,^67,68 predisposing to a syndrome of complications throughout gestation that may be associated with altered placental function,^69,70 embryo and fetal growth restriction,⁷¹ preeclampsia,⁷² pregnancy loss,⁷³ and ultimately preterm birth. Of relevance for preventive care, this syndrome could have its roots in the preconception or periconception period.⁷⁴

This study has some strengths. A major strength was the opportunity to assess maternal and paternal phthalate exposure before conception. Although the generalizability of our findings to fertile couples is uncertain, our results are consistent with those of previous studies reporting phthalate-associated adverse pregnancy outcomes in both subfertile and fertile populations, including a preconception cohort of fertile couples in which maternal preconception phthalate exposure was associated with reductions in gestational age and birth weight.⁷⁵ In addition, the ongoing follow-up of the EARTH Study allowed for a timely assessment of the phthalate substitute DINCH; however, continued follow-up will allow us to strengthen this evidence. Another strength was the use of multiple urine samples, resulting in a more precise exposure assessment while reducing exposure misclassification and its expected attenuation bias.⁷⁶ However, some degree of misclassification cannot be ruled out given the short biological half-lives and episodic nature of exposure to these nonpersistent chemicals. There has been recent interest in examining the association between the modification of environmental chemicals and preterm birth by levels of stress in pregnancy.⁷⁷ Unfortunately, owing to the absence of any stress assessment in the EARTH Study, we could neither adjust for nor examine this association.

This study also has some limitations. One limitation was the modest number of preterm birth cases, which precluded us from studying clinical subtypes of preterm birth. Given that previous research has shown elevated odds of spontaneous preterm birth in association with prenatal phthalate exposure,¹⁴ future work with more cases should allow for the examination of subtypes of preterm birth. We also had limited power to detect sex-specific differences; these results should be interpreted with caution. We acknowledge that multiple comparisons were performed, and thus we cannot rule out that some of the associations could be due to chance. However, this possibility appears unlikely given the consistency of our results for the maternal preconception window showing positive associations with specific metabolites of DEHP (or phthalate substitutes), the absence of any associations within the paternal preconception window across all of the metabolites examined, and the attenuation of any potential paternal finding after accounting for maternal preconception biomarker exposure. Furthermore, the maternal preconception DEHP associations remained robust and internally consistent across all of the models analyzed. Last, our significant results have biological underpinnings from prior toxicologic studies and are consistent with previous epidemiologic evidence.

In this prospective study, maternal preconception urinary ΣDEHP metabolite concentrations were associated with an increased risk of preterm birth. Our results suggest that female exposure to DEHP before conception might be an unrecognized risk factor for adverse pregnancy outcomes, often overlooked in clinical practice. These findings may have important clinical and public health implications, given the ubiquity of DEHP exposure, the importance of the outcome, and that prevention strategies rarely focus on preconception care. Although future studies should validate these associations, it is appropriate to inform couples planning conception about measures to reduce phthalate exposure.

Accepted for Publication: February 10, 2020.

Published: April 7, 2020. doi:10.1001/jamanetworkopen.2020.2159

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Zhang Y et al. JAMA Network Open.

Corresponding Author: Carmen Messerlian, PhD, Department of Environmental Health, Harvard T.H. Chan School of Public Health, 665 Huntington Ave, Bldg 1, 1310C, Boston, MA, 02115 (cmesser@hsph.harvard.edu).

Author Contributions: Ms Zhang and Dr Messerlian had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Ms Zhang and Dr Mustieles contributed equally to this work.

Concept and design: Zhang, Yland, Williams, Hauser, Messerlian.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Zhang, Mustieles, Yland, Messerlian.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Zhang, Yland, Messerlian.

Obtained funding: Williams, Hauser, Messerlian.

Administrative, technical, or material support: Ford, Calafat, Hauser, Messerlian.

Supervision: Mustieles, Williams, Calafat, Messerlian.

Conflict of Interest Disclosures: Drs Williams and Hauser reported receiving grants from the National Institute of Environmental Health Science during the conduct of the study. Dr Braun reported receiving grants from the National Institutes of Health during the conduct of the study; and compensation from Quest Diagnostics outside the submitted work. No other disclosures were reported.

Funding/Support: This study was supported by grants ES009718 and ES000002 from the National Institute of Environmental Health Sciences. Dr Messerlian received funding from the Canadian Institutes of Health Research.

Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention (CDC). Use of trade names is for identification only and does not imply endorsement by the CDC, the Public Health Service, or the US Department of Health and Human Services.

Additional Contributions: We acknowledge all members of the EARTH Study team, including research staff Ramace Dadd and Myra Keller, BSN, Department of Environmental Health, Harvard T.H. Chan School of Public Health, and physicians and staff at Massachusetts General Hospital Fertility Center. Dadd and Keller were not compensated for their contributions. We are also grateful to all study participants. Manori Silva, PhD, Ella Samandar, BS, Jim Preau, BS, Tao Jia, MS, and Xiaoyun Ye, MS (deceased), Centers for Disease Control and Prevention, measured the urinary concentrations of phthalate biomarkers. They were not compensated for their contributions.