Nutritional Intake in Dichorionic Twin Pregnancies: A Descriptive Analysis of a Multisite United States Cohort

Abstract

Introduction

Twin gestations have greater nutritional demands than singleton gestations, yet dietary intakes of women with twin gestations have not been well described.

Methods

In a prospective, multi-site US study of 148 women with dichorionic twin gestations (2012–2013), we examined longitudinal changes in diet across pregnancy. Women completed a food frequency questionnaire during each trimester of pregnancy. We examined changes in means of total energy and energy-adjusted dietary components using linear mixed effects models.

Results

Mean energy intake (95% CI) across the three trimesters was 2010 kcal/day (1846, 2175), 2177 kcal/day (2005, 2349), 2253 kcal/day (2056, 2450), respectively (P = 0.01), whereas the Healthy Eating Index-2010 was 63.9 (62.1, 65.6), 64.5 (62.6, 66.3), 63.2 (61.1, 65.3), respectively (P = 0.53).

Discussion

Women with twin gestations moderately increased total energy as pregnancy progressed, though dietary composition and quality remained unchanged. These findings highlight aspects of nutritional intake that may need to be improved among women carrying twins.

Introduction

Women with twin pregnancies represent only ~ 3.3% of US deliveries (Martin et al., 2019), yet they are of significant public health importance due to their disproportionate attributable risk of complications and associated health care costs (Sherer, 2001). There is a need to understand factors that contribute to the health of twin pregnancies of which nutrition is an important one.

The 2014 Farm Bill mandated that the Dietary Guidelines for Americans (DGAs) include pregnancy-specific recommendations which has led to a comprehensive summary of the current state of evidence on diet and pregnancy outcomes (Stoody et al., 2019).The resulting 2020–25 DGAs present nutritional recommendations for pregnant women, echoing for total energy intake the Institute Of Medicine recommendations to increase calories by 340 kcal/day and 452 kcal/day in the second and third trimesters, respectively for women with a healthy pre-pregnancy weight (Rasmussen et al., 2010). Yet, recommendations specific to multifetal (twin) pregnancies have not been made at least in part due to the paucity of data in this subgroup. Cohorts of US pregnant women with repeated dietary assessment during pregnancy are rare (Brown et al., 1996; Haas et al., 2015; Oken et al., 2014; Savitz et al., 1999), and tend to include only or mostly women with singleton gestations.

A recent review on the dietary needs of twin pregnancies has highlighted the insufficient literature and several unknowns on the topic (Wierzejska, 2022). The increased physiological demand of supporting more than one fetus is agreed to lead to higher energy demand though consensus on the exact amount has not been established. Body stores of some micronutrients are also depleted faster than in singleton pregnancies (Zgliczynska et al., 2021). Yet, women with twin pregnancies experience earlier satiety as well as more nausea and vomiting which may deter them from higher intakes or cause them to change their overall dietary patterns. Critical questions remain regarding the general pattern of dietary intake of women with twin gestations and whether, how, and when during gestation they change aspects of their diets such as total energy, food groups, overall diet quality as well as macro- and micronutrients.

The aim of this analysis was to provide trimester specific estimates of dietary intake and quality, per the Healthy Eating Index 2010, among women with dichorionic twin pregnancies from multiple sites in the US, and to longitudinally examine changes across trimesters.

Methods

The study was a secondary analysis of a cohort that included women with dichorionic twin gestations recruited from eight US clinical centers between 2012 and 2013 as part of the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Fetal Growth Studies-Twins. Women 18–45 years of age were enrolled between 8 and 0 days and 13 weeks 6 days of gestation. Gestational age was based on the last menstrual period, which had to match that estimated by ultrasound for the larger twin, or for women who conceived with in vitro fertilization was calculated using the date and embryo age at transfer. Exclusion criteria were: planned fetal reduction, congenital anomalies detected during the first-trimester sonogram, increased nuchal translucency (≥ 3.5 mm) in either twin sibling, monochorionic twins, or crown-lump length discordance > 10%. Details of the protocol are described elsewhere (Grantz et al., 2016; Grewal et al., 2017). Ethical approval was obtained from the Institutional Review Boards of NICHD, the clinical sites, and the data and imaging coordinating centers.

At enrollment, women were interviewed for data on sociodemographic characteristics, reproductive history, and lifestyle factors. Pre-pregnancy weight and height were self-reported and used to calculate pre-pregnancy body mass index (BMI). Complications of the current pregnancy, such as gestational diabetes or hypertensive disorders of pregnancy, were extracted from medical records. Diet reflecting the past three months was assessed in each trimester (11.0–13.9 weeks, 19.0–24.9 weeks, and 31.0–34.9 weeks) using a self-administered semi-quantitative food frequency questionnaire (FFQ), which was a modified version of the Diet History Questionnaire II (Subar et al., 2001, 2003). Study coordinators examined each FFQ and contacted participants to complete any missing items. My Pyramid Equivalent Database (MPED) food group (Bowman et al., 2008) and nutrient intakes were determined using the Diet*Calc software (National Cancer Institute, Bethesda, MD). Diet*Calc matches FFQ items to a food composition database and estimates nutrient intake based on the reported frequency and serving size of the items. We defined implausible dietary data as total energy less than 600 kcal/day or greater than 6000 kcal/day (Radesky et al., 2008). The Healthy Eating Index (HEI)-2010 total score and component scores were calculated as described by Guenther et al., 2013 (Guenther et al., 2013). The HEI-2010 is an indicator of diet quality as quantified by adherence to the 2010 Dietary Guidelines for Americans – recommendations that are not pregnancy-specific, but target individuals ages 2 and older for achieving healthy, nutritionally adequate diets. HEI-2010 scores reflect greater adherence to the dietary guidelines relative to an individual’s energy intake with the maximum total score (highest diet quality) being 100 points.

As gestation progressed, the mean dietary intake was estimated using multivariable linear mixed effects regression models with random intercepts. All dietary variables except total energy, shares of energy from macronutrients, and the HEI-2010 total and component scores were energy adjusted. P-values were false discovery rate (FDR)-adjusted within categories of dietary variables to account for multiple comparisons (Benjamini & Hochberg, 1995). We quantified the percentage of women who had perfect adherence to the HEI-2010 components, and tested whether this changed significantly across trimesters using a logistic regression model with generalized estimating equations incorporating a variable indicating the trimester.

We applied inverse probability weighting (Mansournia & Altman, 2016) to account for potential bias that may result from potential differential loss to follow-up in all analyses. Specifically, a logistic regression model was used to estimate trimester-specific weights based on factors associated with loss to follow-up, including age, income, insurance, BMI category, race/ethnicity, parity, marital status, education, nativity, outcome of twins, gestational diabetes, and hypertension during pregnancy. These weights reflect the inverse of the probability of retention (giving more weight to individuals that are more likely to be lost to follow up) and were applied to the aforementioned models that longitudinally estimated continuous and categorical dietary outcomes.

The threshold for significant p-values was set at < 0.05. SAS version 9.4 (SAS Institute, Cary, NC) was used for all analyses.

Results

Of the 171 women enrolled in the full cohort, there were 148 women who had least one dietary assessment (n = 142, n = 127, and n = 81 in the first, second, and third trimester, respectively). After excluding implausible dietary data, there remained 148 women in the analytic sample (n = 138, n = 121, and n = 78 at each trimester); 64 women had plausible dietary data at all three trimesters though the longitudinal models did not require plausible dietary data in all three trimesters for each woman.

Mean (SD) age and pre-pregnancy BMI of women in the full cohort was 32 (6) years and 28 (7) kg/m², respectively (Supplemental Table 1). The cohort was comprised mostly of non-Hispanic White (54%), nulliparous (56%), those married or living with a partner (79%), women who reported being born in the US (84%), and those residing in a household with a family income ≥$100,000 (51%). After applying inverse probability weights accounting for characteristics associated with loss to follow-up, the analytic sample at each visit (n = 138, n = 121, and n = 78), used in all subsequent analyses, reflected the characteristics of the full cohort.

Total energy intake increased statistically significantly across trimesters (mean (95% CI): 2010 kcal/day (1846, 2175), 2177 kcal/day (2005, 2349), 2253 kcal/day (2056, 2450), respectively; P = 0.01; Table 1). Changes in all macronutrients were not statistically significant. Energy-adjusted intake of foods, food groups, and micronutrients also remained stable across trimesters as did the HEI-2010 total score (63.9 (62.1, 65.6), 64.5 (62.6, 66.3), 63.2 (61.1, 65.3), respectively; P = 0.53). Likewise, the percentage of pregnant women who met the recommended standard intake for each component of the HEI-2010 did not change significantly across trimesters (Fig. 1). The HEI-2010 components with the lowest percentage of women having perfect adherence across trimesters were whole grains (0% in all trimesters) and sodium (5% (2%, 11%), 3% (1%, 12%), 6% (2%, 17%), respectively; P = 0.66; Fig. 1).

Table 1:

Maternal Dietary Intake across Pregnancy among Women with Dichorionic Twin Pregnancies (NICHD Fetal Growth Studies – Twins)

Dietary Variables	Weighted a Mean b (95% CI)			P c
	1st Trimester Visit n = 138	2nd Trimester Visit n = 121	3rd Trimester Visit n = 78
Total energy (kcal)	2010 (1846, 2175)	2177 (2005, 2349)	2253 (2056, 2450)	0.01
Foods and food groups (MPED servings) d
Fruits and vegetables	4.0 (3.6, 4.4)	4.1 (3.7, 4.5)	3.8 (3.3, 4.3)	0.78
Total fruits	2.0 (1.7, 2.2)	2.2 (1.9, 2.5)	2.1 (1.7, 2.4)	0.60
Whole fruits	1.4 (1.2, 1.6)	1.6 (1.4, 1.8)	1.4 (1.2, 1.7)	0.60
Total vegetables	2.1 (1.8, 2.3)	1.9 (1.6, 2.1)	1.7 (1.5, 2.0)	0.41
Dark-green vegetables	0.4 (0.3, 0.5)	0.3 (0.2, 0.4)	0.3 (0.2, 0.4)	0.60
Orange vegetables	0.1 (0.1, 0.2)	0.1 (0.1, 0.2)	0.1 (0.1, 0.2)	0.71
Other vegetables	0.8 (0.7, 0.9)	0.7 (0.6, 0.9)	0.6 (0.5, 0.8)	0.60
White potatoes	0.3 (0.3, 0.3)	0.2 (0.2, 0.3)	0.3 (0.2, 0.3)	0.41
Starchy vegetables	0.2 (0.1, 0.2)	0.2 (0.1, 0.2)	0.1 (0.1, 0.1)	0.41
Tomatoes	0.3 (0.3, 0.4)	0.3 (0.2, 0.3)	0.3 (0.2, 0.3)	0.71
Whole grains	0.9 (0.8, 1.0)	0.9 (0.8, 1.0)	0.9 (0.8, 1.0)	0.95
Total dairy	1.9 (1.6, 2.2)	1.9 (1.6, 2.2)	2.2 (1.9, 2.6)	0.60
Milk	1.1 (0.8, 1.4)	1.1 (0.8, 1.4)	1.4 (1.0, 1.7)	0.60
Cheese	0.6 (0.5, 0.7)	0.6 (0.5, 0.7)	0.7 (0.6, 0.7)	0.60
Yogurt	0.2 (0.2, 0.3)	0.2 (0.2, 0.3)	0.2 (0.2, 0.3)	0.98
Processed meats	0.5 (0.4, 0.6)	0.4 (0.3, 0.5)	0.5 (0.3, 0.6)	0.60
Legumes	0.1 (0.1, 0.1)	0.1 (0.1, 0.1)	0.1 (0.1, 0.1)	0.98
Fish	1.2 (1.0, 1.4)	1.2 (1.0, 1.4)	1.0 (0.8, 1.3)	0.60
Macronutrients
Carbohydrates (% energy)	50.0 (48.6, 51.4)	49.3 (47.8, 50.8)	48.4 (46.6, 50.2)	0.32
Protein (% energy)	16.5 (15.9, 17.1)	16.6 (16.0, 17.2)	16.9 (16.2, 17.6)	0.54
Total fat (% energy)	35.5 (34.4, 36.7)	36.3 (35.1, 37.5)	36.8 (35.4, 38.2)	0.32
PUFA (% energy)	7.2 (6.9, 7.6)	7.4 (7.1, 7.8)	7.3 (6.9, 7.8)	0.67
SFA (% energy)	11.6 (11.1, 12.0)	12.0 (11.5, 12.5)	12.4 (11.9, 13.0)	0.13
MUFA (% energy)	14.0 (13.4, 14.6)	14.0 (13.4, 14.7)	14.2 (13.5, 15.0)	0.80
Carbohydrates (grams)	247.5 (239.0, 256.0)	248.5 (239.5, 257.5)	236.4 (225.8, 247.1)	0.17
Protein (grams)	82.2 (78.8, 85.6)	81.4 (77.8, 85.0)	85.1 (81.0, 89.3)	0.32
Total fat (grams)	79.8 (76.8, 82.8)	80.6 (77.5, 83.8)	83.9 (80.2, 87.5)	0.17
Added sugars (tsp equivalent)e	14.9 (12.8, 17.0)	15.9 (13.6, 18.1)	12.3 (9.6, 15.1)	0.17
Total dietary fiber (grams)	19.3 (18.0, 20.7)	19.3 (17.8, 20.7)	18.5 (16.9, 20.1)	0.66
Micronutrients
Iron (mg)	16.0 (15.3, 16.6)	15.7 (15.0, 16.3)	15.7 (14.9, 16.5)	0.94
Zinc (mg)	11.7 (11.2, 12.2)	12.2 (11.7, 12.7)	12.0 (11.3, 12.6)	0.49
Calcium (mg)	1004.6 (918.6, 1090.6)	991.0 (900.1, 1082.0)	1077.2 (971.1, 1183.3)	0.49
Potassium (mg)	3176.3 (3017.2, 3335.5)	3153.4 (2984.8, 3322.0)	3189.4 (2991.8, 3387.1)	0.94
Sodium (mg)	3192.4 (3091.8, 3293.0)	3080.9 (2974.2, 3187.6)	3100.1 (2974.5, 3225.7)	0.49
Total folate (μg, DFE)	539.8 (513.6, 566.0)	518.2 (490.6, 545.8)	532.5 (500.6, 564.4)	0.49
Vitamin B6 (mg)	2.0 (1.9, 2.1)	2.0 (1.9, 2.1)	2.0 (1.9, 2.2)	0.94
Vitamin B12 (μg)	5.4 (5.0, 5.8)	5.4 (4.9, 5.8)	5.5 (4.9, 6.0)	0.94
Vitamin C (mg)	159.0 (142.8, 175.3)	163.6 (146.3, 181.0)	154.2 (133.1, 175.2)	0.94
Vitamin A (μg, RAE)	948.1 (874.2, 1022.0)	876.2 (797.8, 954.7)	953.2 (860.5, 1046.0)	0.49
Vitamin D (mg)	5.8 (5.1, 6.5)	5.7 (5.0, 6.5)	6.1 (5.2, 7.0)	0.94
Healthy Eating Index-2010 f
Total score (100)	63.9 (62.1, 65.6)	64.5 (62.6, 66.3)	63.2 (61.1, 65.3)	0.36
Total vegetable score (5)	3.8 (3.6,4.0)	3.6 (3.4,3.8)	3.5 (3.3,3.8)	0.13
Greens and beans score (5)	2.9 (2.6,3.2)	2.9 (2.6,3.2)	2.9 (2.6,3.3)	0.95
Total fruit score (5)	4.2 (4.0,4.4)	4.3 (4.1,4.5)	3.9 (3.7,4.2)	0.13
Whole fruit score (5)	4.4 (4.2,4.6)	4.5 (4.3,4.7)	4.4 (4.1,4.6)	0.56
Whole grains score (10)	3.2 (2.9,3.5)	3.0 (2.6,3.3)	3.0 (2.6,3.4)	0.53
Dairy score (10)	6.2 (5.8,6.6)	6.7 (6.2,7.1)	6.9 (6.4,7.4)	0.12
Total protein foods score (5)	4.6 (4.4,4.7)	4.7 (4.5,4.8)	4.5 (4.4,4.7)	0.34
Seafoods and plant proteins food score (5)	4.0 (3.8,4.2)	4.1 (3.9,4.3)	4.0 (3.7,4.3)	0.64
Fatty acid ratio score (10)	5.0 (4.5,5.5)	4.9 (4.4,5.4)	4.6 (4.0,5.2)	0.56
Sodium score (10)	4.4 (3.9,4.8)	4.9 (4.5,5.4)	4.8 (4.3,5.3)	0.14
Refined grain score (10)	7.9 (7.5,8.3)	8.4 (8.0,8.8)	8.0 (7.5,8.4)	0.12
Empty calories score (20)	13.3 (12.5,14.1)	12.7 (11.8,13.5)	12.8 (11.8,13.7)	0.32

^aInverse probability weighting applied to the analytic sample at each trimester (n=138, n=121, n=78) to reflect the characteristics of the full cohort (n = 171). Variables used in calculating inverse probability weights included age, income, insurance, body mass index category, race/ethnicity, parity, marital status, education, nativity, twin outcomes, gestational diabetes, and hypertensive disorders of pregnancy.

^bAll means except total energy, % energy of macronutrients, and Healthy Eating Index-2010 were adjusted for energy to 2000 kcal/day

^cThe difference between first, second, third trimester diet was tested, and false discovery rate adjusted P-values within each dietary component section were computed

^dFood and food group servings/day estimated from the MyPyramid Equivalents Database (MPED) in cup equivalents for fruits, vegetables, and dairy and ounce equivalents for meat, poultry, fish, grains, and legumes

^eAdded sugars servings/day estimated from the MyPyramid Equivalents Database (MPED) in teaspoon equivalents

^fHealthy Eating Index-2010 recommended scores are shown in parentheses after each component

Abbreviations - MPED: MyPyramid Equivalents Database; CI: Confidence Interval; PUFA: Polyunsaturated Fatty Acids; SFA: Saturated Fatty Acids; MUFA: Monounsaturated Fatty Acids; DFE, dietary folate equivalents; RAE, retinol activity equivalents.

Fig. 1.

Inverse probability weighting applied to the analytic sample at each trimester (n=138, n=121, n=78) to reflect the characteristics of the full cohort (n=171). Variables used in calculating inverse probability weights included age, income, insurance, body mass index category, race/ethnicity, parity, marital status, education, nativity, twin outcomes, gestational diabetes, and hypertensive disorders of pregnancy. The difference across trimesters were tested and false discovery rate-adjusted P values were computed. All adjusted P values were nonsignificant (≥0.05). Percentage meeting the recommendations for whole grains in each trimester was zero.

^aPercentage of pregnant women in each trimester who met recommended standard intakes of the HEI-2010 components, NICHD Fetal Growth Studies – Twins. Diet at each trimester was assessed using a food frequency questionnaire.

^b HEI-2010 recommendations are not pregnancy-specific. Serving sizes are based on cup equivalents for fruit, vegetables, greens and beans, and dairy, and on ounce equivalents for whole grains, total protein foods, seafood and plant proteins, and refined grains estimated according to the MyPyramid Equivalents Database.

^cRatio of PUFAs and MUFAs to SFAs.

^dCalories from solid fats, alcohol, and added sugars.

Abbreviations – HEI-2010, Healthy Eating Index-2010; NICHD, Eunice Kennedy Shriver National Institute of Child Health and Human Development; MUFA, monounsaturated fatty acid; PUFA, polyunsaturated fatty acid; SFA: saturated fatty acid.

Discussion

In a multisite, diverse cohort of US pregnant women with dichorionic twin gestations, we found that women statistically significantly increased their mean total energy intake by 243 kcal/day from the first to the third trimester. However, there was no meaningful change in energy adjusted intake of foods, macronutrients, and micronutrients, nor diet quality across pregnancy. Diet quality per the HEI-2010 was suboptimal, particularly in whole grains, fatty acid ratio, sodium, and empty calories. These findings shed light on the diets consumed by women carrying twins and highlight specific aspects that may need to be improved.

To our knowledge, this is the only study to describe the nutritional intake of US women carrying twins using validated dietary assessments obtained during each trimester. Although some non-US twin cohorts with repeated dietary assessments exist (Loke et al., 2013; Tong et al., 2018), longitudinal changes in dietary intake during pregnancy have not been reported. One may expect some aspects of diet to change across pregnancy due to either first trimester nausea/vomiting and/or increased nutritional demands later in pregnancy. However, we found that many aspects of diet remained stable across trimesters. Previous studies among women with singleton pregnancies have also reported no changes in dietary patterns in preconception and across pregnancy (Crozier et al., 2009; Cuco et al., 2006) or between the second and third trimester (Hinkle et al., 2020).

Diet quality per the 100-point HEI-2010 – which is not pregnancy-specific but has been prospectively linked with adverse maternal and neonatal outcomes (Bodnar et al., 2017) – was suboptimal. Whole grains, fatty acid ratio, empty calories and sodium were specific components of concern with only 0–15% of women adhering to the recommended intakes throughout pregnancy.

Women in our cohort had total energy intakes of 2010 kcal/day, 2177 kcal/day, and 2253 kcal/day as assessed by an FFQ. The Institute of Medicine (IOM) recommends women with singleton pregnancies increase their total energy intake by 340 kcal/day and 452 kcal/day in the second and third trimesters respectively, with no recommendations specific to multifetal pregnancies. Luke et al. demonstrated that an intervention among women pregnant with twins that included BMI-specific energy intakes ranging from 4000 kcal/day for underweight women to 3000 kcal/day for women with obesity was associated with improved maternal and infant outcomes (Luke et al., 2003). Moreover, Gandhi et al. used energy balance methods among women pregnant with dichorionic twins to recommend an additional ~ 700 kcal/day during the second and third trimesters for supporting gestational weight gain and the rise in energy expenditure (Gandhi et al., 2018). Even with accounting for the largest reported energy underestimation bias of 42% for an FFQ compared to doubly-labelled water (Burrows et al., 2019), our cohort would have total energy intakes of approximately 2855, 3091, and 3199 kcal/day which still fall short of some of these recommendations. Nonetheless, the women in this cohort seemed to gain adequate weight as the mean total gestational weight gain was 18 kg, which was consistent with the IOM’s provisional recommendations for twin pregnancies (Hinkle et al., 2017; Rasmussen et al., 2010). In line with this, some investigators have questioned the appropriateness of current increased energy recommendations due to the high obesity and excessive gestational weight gain rate among women (Jebeile et al., 2016; Most et al., 2019). This uncertainty in energy recommendations seems to be reflected in clinical care; in one study of US women pregnant with twins, 63% of women recalled getting nutrition advice during prenatal care with an overall emphasis on healthy diets and protein, but less than 30% recalled provider advice on caloric intake (Whitaker et al., 2019). In a review of nutrition and multifetal pregnancies, Bricker et al. concluded that the optimal diet in these pregnancies is uncertain due to the lack of randomized trials, and highlighted the need for studies that compare special high-calorie diets to usual diets (Bricker et al., 2015).

The study’s strengths include recruitment from multiple clinical sites, and three measurements of diet during pregnancy using a validated instrument. In addition to overall energy intakes, we examined multiple dietary variables, including diet quality, food groups, and macro- and micronutrients. Analyses thereby provided a comprehensive descriptive report of the dietary intakes of women pregnant with twins. As with all self-reported dietary assessment methods, FFQs (which are valuable in estimating usual intake) have limitations such as underestimation of total energy intake (Burrows et al., 2019). However, we do not expect this bias to systematically vary by trimester and all other components of diet were energy-adjusted. Our study is the largest with repeated dietary data in US twin pregnancies. That said, the small sample size of the cohort limits the statistical power to detect important differences, especially given that not all women had dietary assessments at all three trimesters (~ 44%). Though we used inverse probability weighting to minimize bias from differential loss to follow-up (including those related to obstetric complications), the cohort overall may have limited generalizability as it may overrepresent those of higher socioeconomic status, which can be associated with improved access to healthy diets. Lastly, the data was collected in 2012–13 and may therefore not be reflective of potential population-level dietary changes in the past decade. However, given the paucity of twin pregnancy cohorts with longitudinally measured dietary data, it provides a valuable contribution to the literature.

In summary, we found that while energy increased through pregnancy, other aspects of diet remain largely unchanged among women with dichorionic twin pregnancies. A vital area of future research is identifying appropriate maternal nutrition to optimize outcomes among women with twin gestations.

Significance.

What is Already Known about the Subject?

Twin pregnancies have higher nutritional demands than singleton pregnancies. However, studies with longitudinal dietary data examining if, how, and at what points during gestation women may change their diets to support twin pregnancies have not been available.

What this Study adds?

Though total energy intake increased statistically significantly from the first to the third trimester, other aspects of diet remained stable across gestation. Similar to singleton pregnancies, diet quality was suboptimal among women pregnant with twins.

Data and code Availability

The data and codebook, along with a set of guidelines for researchers applying for the data, will be posted in the future to a data-sharing site, the NICHD/DIPHR Biospecimen Repository Access and Data Sharing [https://brads.nichd.nih.gov] (BRADS). The analytic code for this manuscript is available upon request.