Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Jun 1.
Published in final edited form as: Health Psychol. 2016 Feb 4;35(6):625–633. doi: 10.1037/hea0000313

Diurnal Salivary Cortisol Patterns Prior to Pregnancy Predict Infant Birth Weight

Christine M Guardino 1, Christine Dunkel Schetter 1, Darby E Saxbe 2, Emma K Adam 3, Sharon Landesman Ramey 4, Madeleine U Shalowitz 5; Community Child Health Network (CCHN)6
PMCID: PMC4868628  NIHMSID: NIHMS749733  PMID: 26844584

Abstract

Objective

Elevated maternal psychosocial stress during pregnancy and accompanying changes in stress hormones may contribute to risk of adverse birth outcomes such as low birth weight and preterm birth. Relatedly, research on fetal programming demonstrates intriguing associations between maternal stress processes during pregnancy and outcomes in offspring that extend into adulthood. The purpose of this study was to test whether HPA patterns in mothers during the period between two pregnancies (i.e., the interpregnancy interval) and the subsequent pregnancy predict infant birth weight, a key birth outcome.

Methods

This study sampled salivary cortisol both before and during pregnancy in a diverse community sample of 142 women in the Community Child Health Network (CCHN) study.

Results

Using multilevel modeling, we found that flatter diurnal cortisol slopes in mothers during the interval between one birth and a subsequent pregnancy predicted lower infant birth weight of the subsequent child. This interpregnancy cortisol pattern in mothers also correlated with significantly shorter inter-pregnancy intervals, such that women with flatter cortisol slopes had more closely spaced pregnancies. After adding demographic covariates of household income, cohabitation with partner, and race to the model, these results were unchanged. For participants who provided both interpregnancy and pregnancy cortisol data (n = 73), we found that interpregnancy cortisol slopes predicted infant birth weight independent of pregnancy cortisol slopes.

Conclusions

These novel findings on interpregnancy HPA axis function and subsequent pregnancy outcomes strongly support lifespan health approaches and underscore the importance of maternal stress physiology between pregnancies.

Theories of fetal programming propose that the prenatal environment shapes the development of the fetus and the offspring’s health over the life course. For example, evidence confirms that maternal stress, trauma, alcohol use, malnutrition, and depression predispose the embryo and fetus to higher risk of developmental physical and mental health adversities (Barker, 2002; Pies, Kotelchuck, & Lu, 2014). In studies testing the “fetal origins hypothesis,” an infant’s birth weight often served as a proxy for the quality of the intrauterine environment because it is easily measured, routinely recorded, and reflects key factors that affect fetal growth such as maternal health and undernutrition (Barker, Winter, Osmond, Margetts, & Simmonds, 1989). Low birth weight (<2500 grams) carries an increased risk of infant mortality and impaired neurodevelopment in infants who survive (McCormick, 1985; Paneth, 1995) and also predicts health outcomes over the life course, including vulnerability to cardiovascular and metabolic disorders in adulthood (Barker, 2002). These findings present an important public health issue given that 8.2% of all babies in the U.S. are <2500 g, a figure that rises to 13.6% among African American babies (2009 data; Hamilton, Hoyert, Martin, Strobino, & Guyer, 2013).

Among the established medical and behavioral risk factors for low birth weight are maternal hypertension, malnutrition, and smoking (Kramer, 2003). A growing body of evidence has shown that maternal stress during pregnancy also contributes to lower birth weight (Dunkel Schetter & Lobel, 2012). Although the biological mechanisms underlying this association are not well established, the maternal hypothalamic-pituitary-adrenal (HPA) axis, through the actions of one of its key hormonal products, cortisol, is one of several possible pathways (Sandman, Wadhwa, Chicz-Demet, Dunkel-Schetter, & Porto, 1997).

Maternal cortisol may influence birth weight through multiple mechanisms, one of which is a direct influence on intrauterine growth rate. Maternal cortisol acts directly on the fetus and plays a key role in fetal development, notably organ maturation and neural generation (Challis et al., 2001). During pregnancy, passage of maternal cortisol through the placenta is regulated by the enzyme 11β-hydroxysteroid dehydrogenase (11β-HSD2; Benediktsson, Calder, Edwards, & Seckl, 1997), which allows a portion of maternal cortisol to cross the placental barrier, accounting for about 30–40% of fetal cortisol concentrations (Gitau, Cameron, Fisk, & Glover, 1998). Fetal exposure to elevated cortisol results in processes that reduce blood flow to the fetus and compromise fetal growth by restricting delivery of oxygen and nutrients. As evidence, women who receive synthetic corticosteroid infusions during pregnancy are at increased risk of delivering infants with fetal growth restriction and low birth weight (Bloom, Sheffield, McIntire, & Leveno, 2001; French, Hagan, Evans, Godfrey, & Newnham, 1999).

Elevated cortisol may also contribute to lower birth weight indirectly through earlier delivery (i.e., shortened gestation). Cortisol stimulates the production and release of placental corticotrophin releasing hormone (pCRH) (Sandman et al., 2006), which plays a central role in the physiological events involved in the onset of parturition (McLean & Smith, 2001). pCRH exerts direct effects on the uterus and cervix, supporting estrogen-initiated changes in these tissues (Wadhwa, Culhane, Rauh, & Barve, 2001); pCRH also interacts with prostaglandins and oxytocin, which are known mediators of uterine contractility (Challis, 2000).

Normal pregnancy is characterized by marked increases in maternal HPA products, including two- to four-fold increases in circulating cortisol levels over the course of gestation and notable changes in the feedback loops that regulate glucocorticoid secretion (Harville et al., 2007; Mastorakos & Ilias, 2003; Sandman et al., 2006). Despite these changes, some characteristics of human HPA activity are preserved during pregnancy. For example, the circadian rhythm of cortisol remains intact and generally follows a diurnal pattern characterized by peak levels 30–45 minutes after awakening, followed by gradual decline over the course of the day (Entringer, Buss, Andersen, Chicz-DeMet, & Wadhwa, 2011; Harville et al., 2007). There is, however, individual variability in diurnal cortisol rhythms that may be attributable, in part, to psychological influences. For example, women who report greater distress during pregnancy tend to have flatter cortisol slopes (Kivlighan, DiPietro, Costigan, & Laudenslager, 2008; O’Connor et al., 2014; Obel et al., 2005), which parallels findings of flatter circadian cortisol rhythms in individuals experiencing chronic stress in non-pregnant samples (Abercrombie et al., 2004; Adam & Gunnar, 2001). Furthermore, although the maternal stress response is progressively dampened as pregnancy progresses (de Weerth & Buitelaar, 2005; Entringer et al., 2010), levels of cortisol and other HPA hormones are influenced by psychological stress through the second and third trimester (Entringer et al., 2011; Giesbrecht, Campbell, Letourneau, & Kaplan, 2013; Nierop et al., 2006; Nierop, Wirtz, Bratsikas, Zimmermann, & Ehlert, 2008).

Most pregnancy studies involve convenience samples from prenatal care settings, often recruited after the critical early phases of embryonic and fetal development are over. Recent thinking is that risk factors for adverse birth outcomes could be better addressed if identified and managed prior to conception (Floyd et al., 2013; Johnson et al., 2006). The hypothesis that a woman’s health and life experiences prior to pregnancy potentially shape the development of the fetus and pregnancy outcomes parallels life-course models in maternal-child health (Lu & Halfon, 2003; Misra, Guyer, & Allston, 2003; Ramey et al., 2015). Studies examining pre-pregnancy health necessitate data collection efforts before a pregnancy is recognized, which may include the intervals between pregnancies. Large population-based studies using national registeries have linked preconception maternal stress to lower birth weight in Denmark (Khashan et al., 2008) and the U.S. (Strutz et al., 2014). Neither of these studies, however, explored possible biological mechanisms through which pre-pregnancy stress influenced birth outcomes.

The Present Study

We used data collected by the Eunice Shriver National Institute of Child Health and Development (NICHD)-funded Community Child Health Network (CCHN) to examine diurnal cortisol patterns in mothers prior to and during a pregnancy and test relationships between these patterns and infant birth weight. Mothers were recruited after the birth of an “index” child and studied at regular intervals for two years. During this time, a subset of mothers who became pregnant again was studied during their “subsequent” pregnancy. This study design intentionally afforded the rare opportunity to collect data during the interpregnancy interval, defined as the length of time between the birth of a child and conception of a subsequent child. The first aim was to test whether diurnal cortisol patterns in mothers predict infant birth weight. Based on the literature described above linking dysregulated maternal HPA axis activity to poorer birth outcomes, we hypothesized that flatter maternal cortisol slopes would be associated with lower infant birth weight. The second aim was to test the hypothesis that interpregnancy cortisol patterns predict infant birth weight over and above pregnancy cortisol patterns.

Method

Participants

CCHN is a five-site research network engaged in community-based participatory research methods in the process of planning and conducting a prospective longitudinal study to better understand disparities in maternal health and child development (Ramey et al., 2015). The full CCHN cohort includes 2,510 mothers and 1,436 fathers or co-parenting partners. Eligibility criteria, recruitment procedures, and cohort demographic characteristics are described elsewhere (Dunkel Schetter et al., 2013; Ramey et al., 2015). Briefly, mothers were recruited just after the birth of a child in one of five study sites: Washington, DC; Baltimore, MD; Los Angeles County, CA, Lake County, IL, and eastern North Carolina. The study catchment areas were predominantly low income. CCHN sampled only African American (54%), Latina (24%) and non-Hispanic white women (22%) and oversampled women who had delivered preterm infants.

Eighty-three percent of the mothers in the full CCHN cohort completed at least one of five possible follow-up study visits held at six month intervals between six months and two years after the birth of the index child (n = 2089). During at least one of these visits, 372 participants (18%) reported that they were currently pregnant. Participants were also asked to contact study staff if they became pregnant again between study visits, and an additional 44 subsequent pregnancies were identified in this manner for a total of 416 subsequent pregnancies during the follow-up period. Most participants (n = 343, 82%) consented to continued participation in the subsequent pregnancy follow-up study and completed at least one study visit during or shortly after the subsequent pregnancy. A majority of the remaining 73 subsequent pregnancies were among women lost to follow up (n = 64). A few women withdrew from the study for reasons such as moving out of the study catchment area (n = 2), death of the index child (n = 2), miscarriage (n = 3), and lack of time/interest (n = 2). There were no differences in terms of race/ethnicity or cohabitation status between women with subsequent pregnancies who withdrew or were lost to follow-up and those who were in this sample, except participants had a higher mean per capita household income ($10,407 vs. $14,816).

Neonatal hospital charts with birth weight data were available for 194 (57%) of the 343 subsequent pregnancies. All participants were invited to participate in the saliva sampling portion of the study at all time points; however, some participants did not return the saliva sampling kits and only participants who provided saliva samples at least once during the interpregnancy interval or subsequent pregnancy and for whom we were able to obtain newborn medical charts were included in the multilevel analyses (n = 142 women). Study sample characteristics are presented in Table 1. Compared to women with subsequent pregnancies who were excluded due to missing birth data, participants were significantly more likely to be Latina (34% of participants vs. 18% of those excluded) or White (34% vs. 21%) and less likely to be African American (32% vs. 60%). Participants also had a significantly higher per capita household income (mean $15,221 vs. $10,304). There were no significant differences between included and excluded participants in cohabitation status.

Table 1.

Descriptive Statistics (n = 142)

Descriptive variables n (%)
Race/ethnicity
 African American/Black 45 (31.7)
 White/Caucasian 49 (34.5)
 Hispanic/Latina 48 (33.8)
Cohabitating with baby’s father 111 (78.2)
Recruitment Site
 Baltimore 25 (17.6)
 Lake County, IL 84 (59.2)
 Los Angeles County 6 (4.2)
 Eastern North Carolina 9 (6.3)
 Washington, DC 18 (12.7)
Cortisol samples provided
 T2 (6 months postpartum) 109 (76.7)
 T3 (12 months postpartum) 93 (65.5)
 T5 (24 months postpartum) 30 (21.1)
 P1 (2nd trimester) 79 (55.6)
 P2 (3rd trimester) 83 (58.5)
Total cortisol sampling days
 1 27 (19.0)
 2 34 (23.9)
 3 32 (22.5)
 4 42 (29.6)
 5 7 (4.9)
M (SD) Range
Household income 15,221 (24224) 0—40,180
Infant birth weight (g) 3,278 (556.35) 1090—4750
Interpregnancy interval (days) 455.46 (233.22) 80—1363
Interpregnancy diurnal cortisol
 Cortisol at wake (nmol/L) 5.50 (4.94) .10—42.0
 Cortisol at wake +30 (nmol/L) 5.69 (4.75) .10—39.0
 Cortisol at bedtime (nmol/L) 2.23 (4.30) .10—25.0
Pregnancy diurnal cortisol
 Cortisol at wake (nmol/L) 6.59 (4.79) .10—42.0
 Cortisol at wake +30 (nmol/L) 7.08 (4.71) .20—39.0
 Cortisol at bedtime (nmol/L) 2.78 (2.77) .20—19.5
Open in a new tab

Note: Interpregnancy diurnal cortisol descriptives reflect a total of 152 sampling days across 98 participants. Pregnancy diurnal cortisol descriptives reflect a total of 242 sampling days across 126 participants. All cortisol values reflect raw values prior to log transformation.

Procedures

CCHN study visits occurred following a birth when index children were approximately 1 month (T1), 6 months (T2), 12 months (T3), and 24 months (T5) of age with an additional telephone interview at 18 months (T4). Mothers who became pregnant again during this study period were interviewed during the second (P1) and third (P2) trimesters of their subsequent pregnancies, and then 1 month after the birth of the subsequent child (P3). Biomarkers collected at the T2, T3, T5, P1, and P2 visits were examined in the present study. Also, some women were pregnant at the T-series visits and if so, their visits were considered pregnancy visits. The beginning of the pregnancy was dated using the first day of the last menstrual period (LMP) or ultrasound dating. We used this information to identify interpregnancy vs. pregnancy sampling days and to determine the trimester of pregnancy.

Salivary cortisol

During the study visits, research staff provided saliva sampling kits and verbal and written instructions on procedures for collecting saliva at three times over the course of the sampling day (upon waking, 30 minutes after waking, and bedtime), completed a practice sample with the participant, and answered questions. The kits included vials, labels, and straws for “passive drool” collection of saliva. The kits also included morning and bedtime diaries that were completed by participants on the sampling day and used to extract time of sampling and related variables such as shift work and cigarettes smoked, which were included in preliminary analyses but were not significant predictors of cortisol and did not affect study results. On the day after the study visit, participants were asked to self-collect samples by expelling saliva through straws into sterile 2 or 5 ml cryogenic vials, and to record the sampling time on accompanying labels. Completed materials were mailed back to each study office and stored at −80 degrees Celsius. Saliva samples were subsequently shipped to ZRT Laboratories on dry ice (Beaverton, OR) and assayed for cortisol by a competitive luminescence immunoassay (IBL-America, Minneapolis, MN) with reported detection limits of 0.015 μg/dl. The intra- and inter-assay coefficients of variance were 5.5% and 7.6%, respectively.

Infant birth weight

Research staff extracted birth weight from birth records/neonatal charts. Birth weight was treated as a continuous outcome and was divided by the standard deviation so that in interpreting results each one-unit change reflects a standard deviation.

Data analysis

Raw cortisol values were examined for extreme outliers, as they can bias the results. We took a conservative approach to outliers, using a two-step strategy as follows: First, all values > 3 SD from the sample mean were dropped, a common approach when cortisol values are extremely skewed (Dettling, Gunnar, & Donzella, 1999). Several outliers remained after truncation, so then we recalculated the sample mean and SD and winsorized any values > 3 SD above the mean by transforming them into the mean + 3 SD. Altogether, 59 values were dropped or winsorized, less than 1% of the full sample of >8000 samples. Finally, because significant skew remained (skewness = 3.47/SE = .03), we natural log-transformed all cortisol values.

In order to test diurnal slope, or the effect of time of day on cortisol levels, multilevel modeling (HLM 7.0; Raudenbush et al., 2011) was used to test a three-level model with cortisol levels as the outcome and sampling time as a predictor variable. This statistical approach is well suited for data that have a nested structure, such as cortisol sampling occasions within days within participants. It can adjust for saliva sample collection times, when these vary within and across participants, and allows for inclusion of all participants when some data are missing at the within-person level (Singer & Willett, 2003). Our Level 1 (sample-level) predictors included time and a dummy variable to reflect the cortisol awakening rise (CAR), coded as 1 for the wake + 30 sample or zero for the other samples (Adam, Hawkley, Kudielka, & Cacioppo, 2006). For further precision, we added one additional covariate at Level 1, which was the lab report of whether the sample was clear or discolored (dichotomously coded), which might indicate food or other contamination.

All available days from all participants were included in our final model. The intercept term was allowed to vary randomly and other effects were fixed. Restricted maximum likelihood estimation was used. At Level 2 (the day-level), of the full model, we modeled pregnancy trimester on the sampling day, coded as 0 for interpregnancy, 1 for first trimester, 2 for second trimester, and 3 for third trimester sampling days. Finally, at Level 3 (the person-level) we modeled infant birth weight. Race (coded as 1= African-American; 0=non African-American), household income adjusted for cost of living, cohabitation with the father of the index child during the first year of the child’s life, and interpregnancy interval (number of days between the index child’s birth and the first day of the subsequent pregnancy) were included as Level-3 covariates in the fully controlled model.1

Since we hypothesized that interpregnancy cortisol might be predictive of subsequent birth outcomes above and beyond the effects of pregnancy cortisol, we conducted a series of follow-up analyses to compare interpregnancy and pregnancy diurnal slopes to each other. First, we extracted Empirical Bayes (EB) coefficients from the HLM analyses for the separate pregnancy and interpregnancy models, by running these analyses without additional Level 2 and 3 covariates and then saving the residual files. EB coefficients are analogous to regression coefficients so the EB coefficient for the effect of sampling time of day on cortisol can be treated like an estimate of the diurnal slope (Raudenbush, 2004). These models included all available pregnancy and interpregnancy days such that if women contributed multiple cortisol sampling days, the EB coefficients reflects the effect of time of day on cortisol across all available sampling days for the given period of time (pregnancy or interpregnancy).

Results

Table 1 shows descriptive statistics for the study sample and for cortisol at waking, 30 minutes after waking, and bedtime for interpregnancy and pregnancy time periods. Cortisol values were obtained from 394 sampling days and 1127 cortisol sampling occasions. On average, women completed 2.67 sampling days (range=1 to 5, SD = 1.19; median = 3 sampling days). Ninety-one women completed at least one interpregnancy sampling day, and 119 women completed at least one pregnancy sampling day. Seventy-three women (51.4%) completed both an interpregnancy and a pregnancy sampling day. A more detailed breakdown of the number of participants who completed saliva sampling days at each time point is provided in Table 1. Since HLM is able to handle missing and unevenly spaced data, all 142 women with available birth weight who contributed at least one cortisol sampling day were included in the multilevel models of diurnal salivary cortisol. However, only the 73 women who contributed both interpregnancy and pregnancy sampling days were used in the follow-up regression analysis.

Multilevel Models of Diurnal Salivary Cortisol

First, we tested a model that included both interpregnancy and pregnancy saliva sampling days with the following variables included as Level 1 predictors: time, cortisol awakening response (CAR), and a dichotomous variable indicating whether the sample was clear or discolored. Pregnancy trimester (which was coded as zero for interpregnancy sampling days) was included as a Level 2 covariate. Table 2 shows the full model with race, income, and cohabitation status added as additional covariates at Level 3. Longer interpregnancy intervals, greater per capita household income, and higher birth weight were all associated with significantly steeper cortisol slopes. Additionally, African-American mothers had significantly lower morning cortisol and a smaller decline across the day. Pregnancy trimester was associated with higher morning values and flatter diurnal slopes.

Table 2.

Three-Level Model Showing Associations Between Infant Birth Weight, Interpregnancy Interval, and Diurnal Cortisol Slope Sampled Before and During Pregnancy: Fixed Effects With Robust Standard Errors and Additional Covariates (n = 142)

Fixed effects Estimate (SE) t ratio df p
Cortisol intercept (morning) 1.67 0.06 28.66 136 <0.001
 Level 2 covariates
  Pregnancy trimester 0.22 0.03 7.55 251 <0.001
 Level 3 covariates
  Infants’ Birth Weight 0.03 0.04 0.65 136 0.518
  Interpregnancy Interval 0.0001 0.0002 0.54 136 0.594
  Household income 0.000004 0.000002 1.71 136 0.090
  Cohabitation −0.06 0.11 −0.51 136 0.612
  African-American −0.39 0.10 −3.73 136 0.001
Cortisol slope (time) −0.08 0.00 −25.99 638 <0.001
 Level 2 covariates
  Pregnancy trimester 0.01 0.003 3.23 638 0.001
 Level 3 covariates
  Infants’ Birth Weight −0.01 0.003 −3.05 638 0.002
  Interpregnancy Interval −0.00003 0.00001 −2.27 638 0.023
  Household income −0.000000 0.000000 −2.13 638 0.033
  Cohabitation −0.01 0.01 −0.78 638 0.434
  African-American 0.07 0.007 10.62 638 <0.001
CAR 0.16 0.05 3.53 638 <0.001
Sample clarity 0.06 0.05 1.20 638 0.233
Open in a new tab

Notes. Pregnancy trimester coded as 0 for preconception study days, 1 for first trimester, 2 for second trimester, and 3 for third trimester. Interpregnancy interval calculated as difference between birth of index child and LMP. Sampling time is centered around wake time = 5 am. Household income is adjusted for cost of living differences across study site. Variable names for Level 1 indices (intercept, slope, CAR, and sample clarity) are presented in bold font.

We re-tested the model separately for interpregnancy and pregnancy cortisol samples in order to establish whether associations with birth weight differed as a function of time relative to conception. We did not include a Level 2 covariate for pregnancy trimester in the interpregnancy model as we did in the full model. In the pregnancy-only model, we included a covariate for number of days pregnant in order to use a more sensitive measure of pregnancy stage. As expected, being later in pregnancy was associated with a significantly higher cortisol intercept and flatter diurnal slope. Whether or not the interpregnancy interval was included as an additional Level 3 covariate, birth weight remained significantly associated with the diurnal decline of cortisol in both models. As Table 3 shows, when we added the demographic variables of income, cohabitation status, and race, birth weight remained significantly associated with diurnal slope in both the interpregnancy model and in the pregnancy model.

Table 3.

Three-Level Model Showing Associations Between Infant Birth Weight, Interpregnancy Interval, and Diurnal Cortisol Slope Sampled during Interpregnancy or Pregnancy: Fixed Effects With Robust Standard Errors

Model for Interpregnancy Cortisol (n = 99) Model for Pregnancy Cortisol (n = 114)
Fixed effects Estimate (SE) t ratio df p Estimate (SE) t ratio df p
Cortisol intercept (morning) 1.34 0.11 11.73 93 <0.001 1.51 0.11 14.06 108 <0.001
 Level 2 covariates
  Days gestation -- -- -- -- -- 0.002 0.00 3.65 110 <0.001
 Level 3 covariates
  Infants’ Birth Weight −0.03 0.06 −0.54 93 0.588 0.05 0.06 0.90 108 0.370
  Interpregnancy Interval −0.00003 0.00032 −0.10 93 0.920 0.0006 0.00 2.39 108 0.019
  Household income 0.000001 0.000002 0.58 93 0.561 0.00001 0.00 2.72 108 0.008
  Cohabitation −0.04 0.17 −0.22 93 0.826 −0.10 0.12 −0.82 108 0.415
  African-American −0.61 0.16 −3.80 93 <0.001 −0.31 0.12 −2.63 108 0.010
Cortisol slope (time) −0.09 0.01 −16.70 201 <0.001 −0.10 0.01 −13.10 327 <0.001
 Level 2 covariates
  Days gestation -- -- -- -- -- 0.0002 0.00 4.30 327 <0.001
 Level 3 covariates
  Infants’ Birth Weight −0.010 0.004 −2.23 201 0.027 −0.008 0.00 −2.15 327 0.032
  Interpregnancy Interval 0.00005 0.00003 1.80 201 0.074 −0.00004 0.00 −2.63 327 0.009
  Household income 0.000000 0.000000 −0.83 201 0.408 0.000000 0.00 −1.34 327 0.182
  Cohabitation −0.02 0.01 −1.68 201 0.094 0.004 0.00 0.49 327 0.628
  African-American 0.10 0.01 7.71 201 <0.001 0.064 0.00 7.98 327 <0.001
CAR 0.16 0.08 1.96 201 0.051 0.16 0.05 3.14 327 0.002
Sample clarity 0.21 0.11 1.87 201 0.063 0.01 0.06 0.18 327 0.858
Open in a new tab

Notes. Pregnancy trimester coded as 0 for preconception study days, 1 for first trimester, 2 for second trimester, and 3 for third trimester. Interpregnancy interval calculated as difference between birth of index child and LMP. Sampling time is centered around wake time = 5 am. Household income is adjusted for cost of living differences across study sites. Variable names for Level 1 indices (intercept, slope, CAR, and sample clarity) are presented in bold font.

Finally, we extracted Empirical Bayes (EB) coefficients from the HLM analyses for the separate pregnancy and interpregnancy models. Visual inspection of plotted slopes revealed one outlier interpregnancy value (slope of .27, 5 SDs beyond the mean), which we dropped from subsequent analyses. Consistent with the overall negative effect of time of day on cortisol, these coefficients were, on average, negative (M for pregnancy cortisol = −.07, SD = .05, range = −.19 to .02, and M for interpregnancy cortisol = −.10, SD = .05, range = −.20 to .03).

Consistent with the HLM results, zero-order correlations indicated that diurnal slope coefficients were negatively associated with birth weight such that steeper slopes were linked with higher birth weights (r (90) = −.36, p = .001 for the interpregnancy model; r (118) = −.23, p = .021 for the pregnancy model). Interpregnancy and pregnancy diurnal slopes were moderately positively correlated with each other, r (72) = .58, p = .001), suggesting some trait-level stability across time periods. In order to compare the effects of interpregnancy and pregnancy cortisol slopes on birth weight, we included only women who provided cortisol samples on at least one interpregnancy and one pregnancy day in a regression analysis with both slopes as predictors of birth weight. The effect of pregnancy diurnal cortisol slope on birth weight became non-significant (b (70, 2) = −.03, t = −.19, p = .85), while the interpregnancy diurnal cortisol slope remained a significant predictor of birth weight (b (70, 2) = −.39, t = −2.92, p = .005). As shown in Table 4, when we added the original HLM covariates (household income, cohabitation status, African-American race, interpregnancy interval) to this regression analysis, interpregnancy diurnal cortisol slope continued to predict birth weight but all other coefficients were non-significant (all p values > .27).

Table 4.

Regression model predicting infant birth weight from Empirical Bayes (EB) Coefficients for Interpregnancy and Pregnancy Slope (n = 73)

B (SE) t ratio p
Interpregnancy slope −5.39 2.44 −2.21 .03
Pregnancy slope −1.14 2.87 −.40 .69
Interpregnancy Interval −.00 .00 −1.12 .27
Household income .00 .00 −.09 .93
Cohabitation .10 .23 .45 .65
African-American −.13 .27 −.49 .27
Constant 5.67 .47 12.17 .00
Open in a new tab

Total adjusted R2 = .12

Discussion

Repeated diurnal assessments of maternal salivary cortisol collected during the interval between delivery of one child and onset of the next pregnancy allowed us to examine the impact of both interpregnancy and prenatal diurnal slopes on infant birth weight. The results show that mothers who had a longer interpregnancy interval and who delivered heavier babies had steeper diurnal declines in cortisol both before and during pregnancy. In contrast, a flatter slope of diurnal cortisol, which has previously been associated with greater chronic stress burden, more maladaptive relationship functioning, and poorer health outcomes (Adam & Gunnar, 2001; Saxbe, Repetti, & Nishina, 2008; Sephton, Sapolsky, Kraemer, & Spiegel, 2000), was associated with lower infant birth weight in the full model that included both interpregnancy and pregnancy cortisol values, as well as in further analyses that tested for separate effects of interpregnancy and pregnancy cortisol patterns. These results persisted when demographic covariates including race, household income, and cohabitation status were controlled. In follow-up analyses, interpregnancy cortisol slopes predicted infant birth weight independent of pregnancy cortisol slopes and demographic covariates. Thus, a particularly robust effect was found for cortisol slopes prior to the infant’s conception, which is notable given that these measurements were less proximal to the pregnancy and birth of the subsequent child than the pregnancy cortisol readings.

By examining cortisol in the interpregnancy interval, we provide the first prospective evidence that maternal cortisol patterns before a woman conceives a child may influence the growth and weight of her next child. This study is particularly novel in its measurement of cortisol during the interpregnancy period. Previously, researchers conducting longitudinal studies have not been in a position to study women during this time prior to conception. In addition to being the first study to examine interpregnancy/preconception biology, this is the largest sample to our knowledge on diurnal salivary cortisol patterns in pregnancy and infant birth weight. Three prior investigations that collected maternal salivary cortisol during pregnancy reported that dysregulated diurnal patterns during pregnancy were linked to lower infant birth weight, but each had fewer than 100 participants (Bolten et al., 2011; D’Anna-Hernandez et al., 2012; Hompes et al., 2012) and none had interpregnancy data.

Stress hormones produced by the HPA axis are necessary for reproductive processes. Our results support the hypothesis that dysregulation of this system before conception may contribute to adverse birth outcomes, specifically risk of reduced fetal growth. At the very least, maternal diurnal cortisol patterns may serve as a marker of broader dysregulation across multiple physiological systems that support fetal growth and, ultimately, influence birth weight. Potentially, these findings could have a number of interpretations: First, a mother’s life conditions lead to HPA dysregulation that influences physiology of pregnancy and fetal growth. That is, maternal cortisol dysregulation is secondary to other risk factors and ongoing stress that adversely affect the fetus. For example, if the flattened cortisol slopes are an overall marker of poor adaptation to stress, they might be linked with health behaviors during pregnancy. Second, HPA dysregulation is comorbid with inflammation (Chrousos, 1995; Miller, Cohen, & Ritchey, 2002; Silverman & Sternberg, 2012) and other processes that may influence fetal growth more directly (Challis et al., 2009; Redman, Sacks, & Sargent, 1999). Beyond these physiological mechanisms, flattened cortisol may reveal something about the overall stress burden, or the mother’s coping or resiliency, that we did not model explicitly in these analyses.

Elevated maternal cortisol during pregnancy has been associated with greater behavioral and physiological stress reactivity in fetuses, infants, and children (Davis et al., 2007; Matthews, 2000); decreased cognitive ability in infants (Davis & Sandman, 2010; Huizink, Robles de Medina, Mulder, Visser, & Buitelaar, 2003); and increased affective problems and larger amygdala volumes in young girls (Buss et al., 2012). There also is some evidence that programming of the fetal HPA axis mediates long-term health consequences of low birth weight (Phillips et al., 1998; van Montfoort, Finken, le Cessie, Dekker, & Wit, 2005). These findings highlight the need for a better understanding of neuroendocrine processes across the life course of both mothers and children. An in-progress follow-up study of the CCHN cohort is examining this in the mothers and subsequent children over the first five years of life.

Flatter slopes were also independently associated with a shorter interpregnancy interval, such that women with more closely spaced pregnancies had flatter cortisol slopes. Short interpregnancy intervals are defined as less than 18 months between the birth of a child and the beginning of a subsequent pregnancy and have been associated with adverse outcomes for both the mother and child including preterm birth, low birth weight, and preeclampsia (Conde-Agudelo, Rosas-Bermúdez, & Kafury-Goeta, 2006; Klerman, Cliver, & Goldenberg, 1998). Investigation of additional maternal health conditions related to cortisol patterns during the interpregnancy period and/or infant birth weight would be valuable now that we have demonstrated the potential importance maternal diurnal cortisol patterns during the interpregnancy period.

A strength of this study is the large, diverse community sample of women studied over a relatively large reproductive time period. Our statistical approach, a three-level hierarchical linear model, added precision to analyses. However, cortisol was collected on only one day at each time point, and published guidelines for ambulatory cortisol research typically recommend a minimum of 2 days with at least 3–6 samples per day in order to calculate diurnal slope (Saxbe, 2008). We used a one-day collection plan because community feedback emphasized the many life demands often at work and at home in this sample of low-income mothers of infants. We therefore developed a protocol that weighed the need for scientific rigor against participant burden and risk of noncompliance (Adam & Kumari, 2009). This issue limits the reliability of our slope estimates, although we attempted to control for some sources of error by controlling for shift work schedules and samples collected on different days (results not shown); even with controls, the results remained significant. Moreover, many participants contributed more than one cortisol sampling day, including multiple interpregnancy and pregnancy days, allowing us to explore the effects of pregnancy on cortisol patterns within participants. A limitation in the exploratory follow-up regression analyses is the sample size (n = 73) but it is rare to have study data on pregnancy and interpregnancy biomarkers and the sample was sufficient if not ideal. In addition, it was not possible to control for gestational age at time of sampling in these exploratory regression analyses due to the fact that all available collection days were used to estimate EB coefficients.

In summary, this study is consistent with a life course approach in examining how processes prior to conception and during pregnancy each influence later pregnancy outcomes (Lu & Halfon, 2003; Misra et al., 2003; Ramey et al., 2015). The results suggest that maternal cortisol patterns may play an important predictive role starting prior to pregnancy. Our results are consistent with the premise that maternal physiology before conception is relevant to birth outcomes and they add to emerging evidence suggesting that birth outcomes are shaped by maternal experiences and health before as well as during pregnancy.

Footnotes

1

Of note, these multilevel models testing the effect of diurnal cortisol slope on infant birth weight included cortisol as the “outcome” because there were multiple measures of cortisol over multiple days for each participant, and only one measure of birth weight. Though this modeling approach was necessary given the nested structure of the cortisol data to test our hypothesis, treating cortisol as the dependent variable does not imply a direction of causality but rather the effect of using nested data where birth weight is at a different level of nesting than cortisol.

References

  1. Abercrombie HC, Giese-Davis J, Sephton S, Epel ES, Turner-Cobb JM, Spiegel D. Flattened cortisol rhythms in metastatic breast cancer patients. Psychoneuroendocrinology. 2004;29(8):1082–92. doi: 10.1016/j.psyneuen.2003.11.003. [DOI] [PubMed] [Google Scholar]
  2. Adam EK, Gunnar MR. Relationship functioning and home and work demands predict individual differences in diurnal cortisol patterns in women. Psychoneuroendocrinology. 2001;26(2):189–208. doi: 10.1016/s0306-4530(00)00045-7. [DOI] [PubMed] [Google Scholar]
  3. Adam EK, Hawkley L, Kudielka B, Cacioppo JT. Day-to-day dynamics of experience–cortisol associations in a population-based sample of older adults. Proceedings of the National Academy of Sciences. 2006;103(45):17058–17063. doi: 10.1073/pnas.0605053103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Adam EK, Kumari M. Assessing salivary cortisol in large-scale, epidemiological research. Psychoneuroendocrinology. 2009;34(10):1423–36. doi: 10.1016/j.psyneuen.2009.06.011. [DOI] [PubMed] [Google Scholar]
  5. Barker DJP. Fetal origins of adult disease: strength of effects and biological basis. International Journal of Epidemiology. 2002;31(6):1235–1239. doi: 10.1093/ije/31.6.1235. [DOI] [PubMed] [Google Scholar]
  6. Barker DJP, Winter PD, Osmond C, Margetts B, Simmonds SJ. Weight in infancy and death from ischaemic heart disease. Lancet. 1989;2(8663):577–80. doi: 10.1016/s0140-6736(89)90710-1. [DOI] [PubMed] [Google Scholar]
  7. Benediktsson R, Calder AA, Edwards CRW, Seckl JR. Placental 11β-hydroxysteroid dehydrogenase: a key regulator of fetal glucocorticoid exposure. Clinical Endocrinology. 1997;46(2):161–166. doi: 10.1046/j.1365-2265.1997.1230939.x. [DOI] [PubMed] [Google Scholar]
  8. Bloom SL, Sheffield JS, McIntire DD, Leveno KJ. Antenatal dexamethasone and decreased birth weight. Obstetrics and Gynecology. 2001;97(4):485–90. doi: 10.1016/s0029-7844(00)01206-0. [DOI] [PubMed] [Google Scholar]
  9. Bolten MI, Wurmser H, Buske-Kirschbaum A, Papoušek M, Pirke KM, Hellhammer D. Cortisol levels in pregnancy as a psychobiological predictor for birth weight. Archives of Women’s Mental Health. 2011;14(1):33–41. doi: 10.1007/s00737-010-0183-1. [DOI] [PubMed] [Google Scholar]
  10. Buss C, Davis EP, Shahbaba B, Pruessner JC, Head K, Sandman Ca. Maternal cortisol over the course of pregnancy and subsequent child amygdala and hippocampus volumes and affective problems. Proceedings of the National Academy of Sciences. 2012;109(20):E1312–9. doi: 10.1073/pnas.1201295109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Challis JR. Mechanism of parturition and preterm labor. Obstetrical & Gynecological Survey. 2000;55(10):650–60. doi: 10.1097/00006254-200010000-00025. [DOI] [PubMed] [Google Scholar]
  12. Challis JR, Lockwood CJ, Myatt L, Norman JE, Strauss JF, Petraglia F. Inflammation and pregnancy. Reproductive Sciences (Thousand Oaks, Calif) 2009;16(2):206–15. doi: 10.1177/1933719108329095. [DOI] [PubMed] [Google Scholar]
  13. Challis JR, Sloboda D, Matthews SG, Holloway A, Alfaidy N, Patel FA, Newnham J. The fetal placental hypothalamic-pituitary-adrenal (HPA) axis, parturition and post natal health. Molecular and Cellular Endocrinology. 2001;185(1–2):135–44. doi: 10.1016/s0303-7207(01)00624-4. [DOI] [PubMed] [Google Scholar]
  14. Chrousos GP. The hypothalamic-pituitary-adrenal axis and immune-mediated inflammation. The New England Journal of Medicine. 1995;332(20):1351–62. doi: 10.1056/NEJM199505183322008. [DOI] [PubMed] [Google Scholar]
  15. Conde-Agudelo A, Rosas-Bermúdez A, Kafury-Goeta AC. Birth spacing and risk of adverse perinatal outcomes: a meta-analysis. JAMA. 2006;295(15):1809–23. doi: 10.1001/jama.295.15.1809. [DOI] [PubMed] [Google Scholar]
  16. D’Anna-Hernandez KL, Hoffman MC, Zerbe GO, Coussons-Read M, Ross RG, Laudenslager ML. Acculturation, maternal cortisol, and birth outcomes in women of Mexican descent. Psychosomatic Medicine. 2012;74(3):296–304. doi: 10.1097/PSY.0b013e318244fbde. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Davis EP, Glynn LM, Dunkel Schetter C, Hobel C, Chicz-Demet A, Sandman CA. Prenatal exposure to maternal depression and cortisol influences infant temperament. Journal of the American Academy of Child and Adolescent Psychiatry. 2007;46(6):737–46. doi: 10.1097/chi.0b013e318047b775. [DOI] [PubMed] [Google Scholar]
  18. Davis EP, Sandman CA. The timing of prenatal exposure to maternal cortisol and psychosocial stress is associated with human infant cognitive development. Child Development. 2010;81(1):131–48. doi: 10.1111/j.1467-8624.2009.01385.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. De Weerth C, Buitelaar JK. Physiological stress reactivity in human pregnancy--a review. Neuroscience and Biobehavioral Reviews. 2005;29(2):295–312. doi: 10.1016/j.neubiorev.2004.10.005. [DOI] [PubMed] [Google Scholar]
  20. Dettling AC, Gunnar MR, Donzella B. Cortisol levels of young children in full-day childcare centers: relations with age and temperament. Psychoneuroendocrinology. 1999;24(5):519–36. doi: 10.1016/s0306-4530(99)00009-8. [DOI] [PubMed] [Google Scholar]
  21. Dunkel Schetter C, Lobel M. Pregnancy and birth: A multilevel analysis of stress and birthweight. In: Revenson T, Baum A, Singer J, editors. Handbook of Health Psychology. 2. Psychology Press; 2012. pp. 431–464. [Google Scholar]
  22. Dunkel Schetter C, Schafer P, Lanzi RG, Clark-Kauffman E, Raju TNK, Hillemeier MM. Shedding light on the mechanisms underlying health disparities through community participatory methods: the stress pathway. Perspectives on Psychological Science. 2013;8(6):613–633. doi: 10.1177/1745691613506016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Entringer S, Buss C, Andersen J, Chicz-DeMet A, Wadhwa PD. Ecological momentary assessment of maternal cortisol profiles over a multiple-day period predicts the length of human gestation. Psychosomatic Medicine. 2011;73(6):469–74. doi: 10.1097/PSY.0b013e31821fbf9a. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Entringer S, Buss C, Shirtcliff Ea, Cammack AL, Yim IS, Chicz-DeMet A, Wadhwa PD. Attenuation of maternal psychophysiological stress responses and the maternal cortisol awakening response over the course of human pregnancy. Stress. 2010;13(3):258–68. doi: 10.3109/10253890903349501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Floyd RL, Johnson Ka, Owens JR, Verbiest S, Moore Ca, Boyle C. A national action plan for promoting preconception health and health care in the United States (2012–2014) Journal of Women’s Health (2002) 2013;22(10):797–802. doi: 10.1089/jwh.2013.4505. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. French NP, Hagan R, Evans SF, Godfrey M, Newnham JP. Repeated antenatal corticosteroids: size at birth and subsequent development. American Journal of Obstetrics and Gynecology. 1999;180(1 Pt 1):114–21. doi: 10.1016/s0002-9378(99)70160-2. [DOI] [PubMed] [Google Scholar]
  27. Giesbrecht GF, Campbell T, Letourneau N, Kaplan BJ. Advancing gestation does not attenuate biobehavioural coherence between psychological distress and cortisol. Biological Psychology. 2013;93(1):45–51. doi: 10.1016/j.biopsycho.2013.01.019. [DOI] [PubMed] [Google Scholar]
  28. Gitau R, Cameron A, Fisk NM, Glover V. Fetal exposure to maternal cortisol. Lancet. 1998;352(9129):707–708. doi: 10.1016/S0140-6736(05)60824-0. [DOI] [PubMed] [Google Scholar]
  29. Hamilton BE, Hoyert DL, Martin JA, Strobino DM, Guyer B. Annual summary of vital statistics: 2010–2011. Pediatrics. 2013;131(3):548–58. doi: 10.1542/peds.2012-3769. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Harville EW, Savitz Da, Dole N, Herring AH, Thorp JM, Light KC. Patterns of salivary cortisol secretion in pregnancy and implications for assessment protocols. Biological Psychology. 2007;74(1):85–91. doi: 10.1016/j.biopsycho.2006.07.005. [DOI] [PubMed] [Google Scholar]
  31. Hompes T, Vrieze E, Fieuws S, Simons A, Jaspers L, Van Bussel J, Claes S. The influence of maternal cortisol and emotional state during pregnancy on fetal intrauterine growth. Pediatric Research. 2012;72(3):305–15. doi: 10.1038/pr.2012.70. [DOI] [PubMed] [Google Scholar]
  32. Huizink AC, Robles de Medina PG, Mulder EJH, Visser GHA, Buitelaar JK. Stress during pregnancy is associated with developmental outcome in infancy. Journal of Child Psychology and Psychiatry. 2003;44(6):810–818. doi: 10.1111/1469-7610.00166. [DOI] [PubMed] [Google Scholar]
  33. Johnson K, Posner SF, Biermann J, Cordero JF, Atrash HK, Parker CS, Curtis MG. Recommendations to improve preconception health and health care--United States. Morbidity and Mortality Weekly Report. 2006;55(RR-6):1–23. [PubMed] [Google Scholar]
  34. Khashan AS, McNamee R, Abel KM, Pedersen MG, Webb RT, Kenny LC, Baker PN. Reduced infant birthweight consequent upon maternal exposure to severe life events. Psychosomatic Medicine. 2008;70(6):688–94. doi: 10.1097/PSY.0b013e318177940d. [DOI] [PubMed] [Google Scholar]
  35. Kivlighan KT, DiPietro JA, Costigan KA, Laudenslager ML. Diurnal rhythm of cortisol during late pregnancy: associations with maternal psychological well-being and fetal growth. Psychoneuroendocrinology. 2008;33(9):1225–35. doi: 10.1016/j.psyneuen.2008.06.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Klerman LV, Cliver SP, Goldenberg RL. The impact of short interpregnancy intervals on pregnancy outcomes in a low-income population. American Journal of Public Health. 1998;88(8):1182–1185. doi: 10.2105/ajph.88.8.1182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Kramer MS. The epidemiology of adverse pregnancy outcomes: an overview. The Journal of Nutrition. 2003;133(5 Suppl 2):1592S–1596S. doi: 10.1093/jn/133.5.1592S. [DOI] [PubMed] [Google Scholar]
  38. Lu MC, Halfon N. Racial and ethnic disparities in birth outcomes: a life-course perspective. Maternal and Child Health Journal. 2003;7(1):13–30. doi: 10.1023/A:1022537516969. [DOI] [PubMed] [Google Scholar]
  39. Mastorakos G, Ilias I. Maternal and fetal Hypothalamic-Pituitary-Adrenal axes during pregnancy and postpartum. Annals of the New York Academy of Sciences. 2003;997(1):136–149. doi: 10.1196/annals.1290.016. [DOI] [PubMed] [Google Scholar]
  40. Matthews SG. Antenatal glucocorticoids and programming of the developing CNS. Pediatric Research. 2000;47(3):291–300. doi: 10.1203/00006450-200003000-00003. [DOI] [PubMed] [Google Scholar]
  41. McCormick MC. The contribution of low birth weight to infant mortality and childhood morbidity. The New England Journal of Medicine. 1985;312(2):82–90. doi: 10.1056/NEJM198501103120204. [DOI] [PubMed] [Google Scholar]
  42. McLean M, Smith R. Corticotrophin-releasing hormone and human parturition. Reproduction. 2001;121(4):493–501. doi: 10.1530/rep.0.1210493. [DOI] [PubMed] [Google Scholar]
  43. Miller GE, Cohen S, Ritchey AK. Chronic psychological stress and the regulation of pro-inflammatory cytokines: a glucocorticoid-resistance model. Health Psychology. 2002;21(6):531–541. doi: 10.1037//0278-6133.21.6.531. [DOI] [PubMed] [Google Scholar]
  44. Misra DP, Guyer B, Allston A. Integrated perinatal health framework. American Journal of Preventive Medicine. 2003;25(1):65–75. doi: 10.1016/S0749-3797(03)00090-4. [DOI] [PubMed] [Google Scholar]
  45. Nierop A, Bratsikas A, Klinkenberg A, Nater UM, Zimmermann R, Ehlert U. Prolonged salivary cortisol recovery in second-trimester pregnant women and attenuated salivary alpha-amylase responses to psychosocial stress in human pregnancy. The Journal of Clinical Endocrinology and Metabolism. 2006;91(4):1329–35. doi: 10.1210/jc.2005-1816. [DOI] [PubMed] [Google Scholar]
  46. Nierop A, Wirtz PH, Bratsikas A, Zimmermann R, Ehlert U. Stress-buffering effects of psychosocial resources on physiological and psychological stress response in pregnant women. Biological Psychology. 2008;78(3):261–8. doi: 10.1016/j.biopsycho.2008.03.012. [DOI] [PubMed] [Google Scholar]
  47. O’Connor TG, Tang W, Gilchrist MA, Moynihan JA, Pressman EK, Blackmore ER. Diurnal cortisol patterns and psychiatric symptoms in pregnancy: short-term longitudinal study. Biological Psychology. 2014;96:35–41. doi: 10.1016/j.biopsycho.2013.11.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Obel C, Hedegaard M, Henriksen TB, Secher NJ, Olsen J, Levine S. Stress and salivary cortisol during pregnancy. Psychoneuroendocrinology. 2005;30(7):647–56. doi: 10.1016/j.psyneuen.2004.11.006. [DOI] [PubMed] [Google Scholar]
  49. Paneth NS. The Future of Children. 1. Vol. 5. Center for the Future of Children, the David and Lucile Packard Foundation; 1995. The problem of low birth weight; pp. 19–34. [PubMed] [Google Scholar]
  50. Phillips DI, Barker DJ, Fall CH, Seckl JR, Whorwood CB, Wood PJ, Walker BR. Elevated plasma cortisol concentrations: a link between low birth weight and the insulin resistance syndrome? The Journal of Clinical Endocrinology and Metabolism. 1998;83(3):757–60. doi: 10.1210/jcem.83.3.4634. [DOI] [PubMed] [Google Scholar]
  51. Pies C, Kotelchuck M, Lu MC. Advancing MCH life course [Special Issue] Maternal and Child Health Journal. 2014;18(2) doi: 10.1007/s10995-013-1408-5. [DOI] [PubMed] [Google Scholar]
  52. Ramey SL, Schafer P, Declerque JL, Lanzi RG, Hobel C, Shalowitz M, Raju TNK. The Preconception Stress and Resiliency Pathways model: A multi-level framework on maternal, paternal, and child health disparities derived by Community-Based Participatory Research. Maternal and Child Health Journal. 2015;19(4):707–719. doi: 10.1007/s10995-014-1581-1. [DOI] [PubMed] [Google Scholar]
  53. Raudenbush SW. HLM 6: Hierarchical linear and nonlinear modeling. 2004. [Google Scholar]
  54. Raudenbush SW, Bryk A, Cheong A, Fai Y, Congdon R, du Toit M. HLM 7: Hierarchical linear and nonlinear modeling. Lincolnwood, IL: Scientific Software International; 2011. [Google Scholar]
  55. Redman CWG, Sacks GP, Sargent IL. Preeclampsia: An excessive maternal inflammatory response to pregnancy. American Journal of Obstetrics and Gynecology. 1999;180(2):499–506. doi: 10.1016/S0002-9378(99)70239-5. [DOI] [PubMed] [Google Scholar]
  56. Sandman CA, Glynn L, Dunkel Schetter C, Wadhwa P, Garite T, Chicz-DeMet A, Hobel C. Elevated maternal cortisol early in pregnancy predicts third trimester levels of placental corticotropin releasing hormone (CRH): priming the placental clock. Peptides. 2006;27(6):1457–63. doi: 10.1016/j.peptides.2005.10.002. [DOI] [PubMed] [Google Scholar]
  57. Sandman CA, Wadhwa PD, Chicz-Demet A, Dunkel-Schetter C, Porto M. Maternal stress, HPA activity, and fetal/infant outcome. Annals of the New York Academy of Sciences. 1997;814(1 Neuropeptides):266–275. doi: 10.1111/j.1749-6632.1997.tb46162.x. [DOI] [PubMed] [Google Scholar]
  58. Saxbe DE. A field (researcher’s) guide to cortisol: tracking HPA axis functioning in everyday life. Health Psychology Review. 2008;2(2):163–190. doi: 10.1080/17437190802530812. [DOI] [Google Scholar]
  59. Saxbe DE, Repetti RL, Nishina A. Marital satisfaction, recovery from work, and diurnal cortisol among men and women. Health Psychology. 2008;27(1):15–25. doi: 10.1037/0278-6133.27.1.15. [DOI] [PubMed] [Google Scholar]
  60. Sephton SE, Sapolsky RM, Kraemer HC, Spiegel D. Diurnal cortisol rhythm as a predictor of breast cancer survival. Journal of the National Cancer Institute. 2000;92(12):994–1000. doi: 10.1093/jnci/92.12.994. [DOI] [PubMed] [Google Scholar]
  61. Silverman MN, Sternberg EM. Glucocorticoid regulation of inflammation and its functional correlates: from HPA axis to glucocorticoid receptor dysfunction. Annals of the New York Academy of Sciences. 2012;1261:55–63. doi: 10.1111/j.1749-6632.2012.06633.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Singer JD, Willett JB. Applied Longitudinal Data Analysis:Modeling Change and Event Occurrence: Modeling Change and Event Occurrence. Oxford University Press; USA: 2003. [Google Scholar]
  63. Van Montfoort N, Finken MJJ, le Cessie S, Dekker FW, Wit JM. Could cortisol explain the association between birth weight and cardiovascular disease in later life? A meta-analysis. European Journal of Endocrinology. 2005;153(6):811–7. doi: 10.1530/eje.1.02050. [DOI] [PubMed] [Google Scholar]
  64. Wadhwa PD, Culhane JF, Rauh V, Barve SS. Stress and preterm birth: neuroendocrine, immune/inflammatory, and vascular mechanisms. Maternal and Child Health Journal. 2001;5(2):119–125. doi: 10.1023/A:1011353216619. [DOI] [PubMed] [Google Scholar]

RESOURCES