Skip to main content
NCBI home page
Biological Research for Nursing logoLink to Biological Research for Nursing
. 2015 May 6;18(1):31–42. doi: 10.1177/1099800415585157

Associations of Maternal and Infant Testosterone and Cortisol Levels With Maternal Depressive Symptoms and Infant Socioemotional Problems

June Cho 1,✉, Xiaogang Su 2, Vivien Phillips 3, Diane Holditch-Davis 4
PMCID: PMC5500224  NIHMSID: NIHMS871999  PMID: 25954021

Abstract

This study examined the associations of testosterone and cortisol levels with maternal depressive symptoms and infant socioemotional (SE) problems that are influenced by infant gender. A total of 62 mothers and their very-low-birth weight (VLBW) infants were recruited from a neonatal intensive care unit at a tertiary medical center in the southeast United States. Data were collected at three time points (before 40 weeks’ postmenstrual age [PMA] and at 3 months and 6 months of age corrected for prematurity). Measures included infant medical record review, maternal interview, biochemical assays of salivary hormone levels in mother-VLBWinfant pairs, and standard questionnaires. Generalized estimating equations with separate analyses for boys and girls showed that maternal testosterone level was negatively associated with depressive symptoms in mothers of boys, whereas infant testosterone level was negatively associated with maternal report of infant SE problems in girls after controlling for characteristics of mothers and infants and number of days post birth of saliva collection. Not surprisingly, the SE problems were positively associated with a number of medical complications. Mothers with more depressive symptoms reported that their infants had more SE problems. Mothers with higher testosterone levels reported that girls, but not boys, had fewer SE problems. In summary, high levels of testosterone could have a protective role for maternal depressive symptoms and infant SE problems. Future research need to be directed toward clinical application of these preliminary results.

Keywords: salivary testosterone, salivary cortisol, maternal depressive symptoms, infant socioemotional problems

Many factors, including maternal characteristics (e.g., education and income) and infant health characteristics (e.g., birth weight and gestational age [GA]), affect the interactions between mothers and very-low-birth weight (VLBW; birth weight less than 1,500 g) infants. Research has shown that infant gender significantly influences at least two of these factors, maternal depressive symptoms and infant socioemotional (SE) problems (Lagerberg & Magnusson, 2012; Sheldrick, Neger, & Perrin, 2012). Although findings have been inconsistent, mothers with greater depressive symptoms tend to experience more difficulty interacting with male infants than female infants, possibly because of the greater prevalence of SE problems in male infants (Weinberg, Olson, Beeghly, & Tronick, 2006). These findings raise important questions as to the underlying biological factors that might explain the vulnerability of male VLBW infants to the effects of elevated maternal depressive symptoms and suboptimal infant SE development. Answers to these questions could ultimately provide a basis for understanding and intervening in the interactions between mothers and their VLBW infants.

We selected testosterone as a biomarker of the gender-differentiated effects of maternal depressive symptoms and infant SE problems. According to gender-difference theories of brain development (Geschwind, 1987) and social relationships (Baron-Cohen, Knickmeyer, & Belmonte, 2005), exposure to high levels of testosterone is positively associated with typically male behaviors such as aggression and dominance. As the effect of testosterone on social behavior is presumably moderated by cortisol (Denson, Mehta, & Ho Tan, 2013) and cortisol is commonly used as a biomarker of psychological state such as depression and stress (Lommatzsch et al., 2006), we examined the association of maternal and infant steroid hormonal (testosterone and cortisol) levels with maternal depressive symptoms and infant SE problems concurrently.

Testosterone levels have been shown to affect psychological status in women, although the findings have not been consistent. Compared to women with normal testosterone levels, women with low testosterone have more psychological problems, including depressive symptoms, diagnosed depressive and anxiety disorders, social phobia, and agoraphobia (Giltay et al., 2012). In contrast, Buckwalter et al. (1999) found positive associations between testosterone levels and depression, mood disturbances, and anger in women across the pre- and postpartum periods. After delivery, testosterone was positively associated with mood disturbances and depression, while other hormones, such as progesterone and dehydroepiandrosterone, attenuated that association.

Research has also demonstrated associations between cortisol and psychological status in pregnant women, with one study finding higher cortisol levels in mothers with more stressors and depressive symptoms than in mothers with fewer stressors and symptoms (Lommatzsch et al., 2006). In addition, severity of depression was positively correlated with the late onset of depression after labor, psychosocial stress, and cortisol levels in another study (Saleh, El-Bahei, Del El-Hadidy, & Zayed, 2013), and these findings were more evident in minority and low-income women (Corwin et al., 2013). Cortisol levels, however, are altered by long periods of depression or fatigue. Women with chronic depression or fatigue had lower cortisol levels and less diurnal variation in cortisol than healthy controls (Corwin et al., 2013). Mothers who had post-traumatic stress disorder (PTSD) symptoms and those who developed dysregulation of the hypothalamic–pituitary–adrenal axis, both of which resulted in lower cortisol levels, perceived their infants as more difficult and initiated fewer interactions with them compared to mothers without PTSD (Brand, Engel, Canfield, & Yehuda, 2006).

The biological bases for infant SE problems, including genetic inheritance and prenatal environmental factors, are often considered to be stronger than external influences. Infant SE problems such as less self-regulation and more difficult temperament are associated with elevated postnatal testosterone levels (Saenz & Alexander, 2013) and with maladaptive responses of mothers to their infants (Muller-Nix et al., 2004). Positive maternal responses such as extended eye contact during alert intervals occurred more often between mothers and female infants or term infants as compared to male or preterm infants. Female infants had lower levels of testosterone in the amniotic fluid and showed significantly longer periods of eye contact with their mothers than male infants (Lutchmaya, Baron-Cohen, & Raggatt, 2002). Perinatal testosterone was also positively related to aggression and dominance in young children (Azurmendi et al., 2006). Thus, empirical evidence supports a positive association between testosterone levels and infant SE problems.

Infant SE problems have also been positively associated with infant cortisol levels (van Bakel & Riksen-Walraven, 2004). Prenatal cortisol in amniotic fluid predicted low birth weight and high levels of infant fear and distress at 3 months (Baibazarova et al., 2013). Cortisol levels of mothers and infants are not always positively related, however, higher maternal cortisol levels were associated with increased infant fear and negative affect at 2 months (Glynn et al., 2007). Mothers reported more SE problems in their 1-year-old infants when the infants had higher cortisol levels in their hair (Palmer et al., 2013). Infants and toddlers with more SE problems exhibited elevated cortisol levels under conditions of neglectful and abusive care (Gunnar & Donzella, 2002) and of environmental adversity associated with poverty (Blair et al., 2011).

As demonstrated by the studies cited above, researchers have investigated the role of testosterone and cortisol in various behavioral outcomes in infants and children (Azurmendi et al., 2006; van Bakel & Riksen-Walraven, 2004). However, little is known specifically about the associations of steroid hormones with maternal depressive symptoms or infant SE problems in the case of VLBW preterm infants who usually have more SE problems than normal birth weight and term infants (Forcada-Guex, Pierrehumbert, Borghini, Moessinger, & Muller-Nix, 2006). The purpose of the present study was to assess possible associations of these hormones with both maternal depressive symptoms and infant SE problems. We addressed three research questions: (a) Are levels of maternal testosterone and cortisol associated with maternal depressive symptoms longitudinally after adjusting for maternal (age, race, education, and income) and infant (gender, GA, Apgar scores, medical complications, intraventricular hemorrhage [IVH], hospitalization days, technology dependence, and degree of neurobiological insult) characteristics as well as the number of days post birth of saliva collection? (b) Are levels of testosterone and cortisol associated with infant SE problems longitudinally after adjusting for maternal and infant characteristics and number of days post birth of saliva collection? (c) Are infant SE problems associated with maternal depressive symptoms after adjusting for maternal and infant characteristics as well as number of days post birth of saliva collection? We hypothesized that high levels of testosterone and cortisol in VLBW infants would be positively associated with SE problems and the latter would be positively associated with maternal depressive symptoms.

Material and Method

We used a comparative longitudinal research design to answer the research questions.

Participants

We recruited 62 mother-VLBW infant pairs from the neonatal intensive care unit (NICU) of a major academic medical center in the southeast United States. Mothers were recruited if they were (1) older than 15 years, (2) able to communicate in English, and (3) primary caregivers of the infant. Mothers were excluded if they were (1) dependent on narcotics or other nonprescribed drugs or (2) HIV positive or if they had (3) a documented serious medical or psychological problem such as cancer or psychosis before or after delivery. Inclusion criteria for infants were (a) 6–10 days old, (b) less than 32 weeks’ GA (GA, which is defined as time elapsed between the first day of the last menstrual period and the day of delivery, “age terminology during the perinatal period,” 2004), and (c) less than 1,500 g at birth. Infants were excluded if they were going to be discharged from the hospital before 1 week of age or had a history or symptoms of substance exposure.

Measures

The characteristics of mother and infant (i.e., demographic information and health history) were obtained through maternal interview and prospective medical record review. Biochemical measurements of testosterone and cortisol levels were obtained through enzyme immunoassay (EIA) of saliva samples, as saliva allows for determining the concentration of the free hormones, unbound to any proteins. Maternal depressive symptoms and infant SE problems were obtained through standard questionnaires and an interview. We established the conceptual framework illustrated in Figure 1 based on the possible associations between biological factors (maternal and infant testosterone and cortisol levels) and maternal and infant well-being (maternal depressive symptoms and infant SE problems) after controlling for covariates (characteristics of mother and infant and number of days post birth of saliva collection).

Figure 1.

Figure 1.

Open in a new tab

Conceptual framework of the association between biological factors (testosterone [T] and cortisol [C] levels) and maternal and infant well-being (maternal depressive symptoms and infant socioemotional [SE] problems). Note. PMA = postmenstrual age.

Characteristics of mothers and infants

Demographic information collected on mothers included age, race, education, and income. History of pregnancy and labor were collected from medical records and individual interviews with the mothers. Infant information collected included gender, GA, Apgar scores, medical complications, IVH, hospitalization days, technology dependence, and degree of neurobiological insult as assessed by the Neurobiologic Risk Score (Brazy, Goldstein, Oehler, Gustafson, & Thompson, 1993). Neonatal history was obtained through medical-record review, and infant health history was updated through maternal reports of common health problems and continued technology dependence.

Biochemical measures

Maternal saliva was collected by having women use a straw to expectorate into a small tube. Infant saliva was collected using low-pressure suction with a 1-cc plastic syringe with a blunt tip. All samples were collected in the NICU before 40 weeks of postmenstrual age (PMA, which includes GA and chronological age for preterm infants during hospital stay; “Age terminology during the perinatal period,” 2004) because infants have relatively high and stable testosterone levels from 1 week through 3 months of age (Achermann, 2005).

Salivary testosterone levels were determined by EIA following the procedure provided by the suppliers (Salimetrics, LLC, State College, PA). Intra- and interassay coefficients of variation were 2.5% and 5.6%, respectively. To control the assay outcome for potential circadian rhythm effects, with higher levels in the morning than in the evening in adults for testosterone production and release, we obtained the samples between 8 a.m. and noon. This circadian variation may not exist in neonates, as the levels are affected by sleeping (Wada et al., 2012).

Salivary samples used for testosterone measurement were also used for cortisol determination. Salivary cortisol levels were determined by EIA (Salimetrics, LLC). The intra- and interassay coefficients of variation were 3.3% and 3.7%, respectively. Cortisol levels in adults are higher in the morning than in the evening. This variation may also exist in infants but not in neonates (de Weerth, Zijl, & Buitelaar, 2003).

Maternal depressive symptoms

The Center for Epidemiologic Studies Depression Scale (CES-D) was developed to assess the frequency of depressive symptoms—such as blues, difficulty concentrating, and sleep problems—during the previous week (Radloff, 1977). Respondents rate the 20 items on a 4-point scale from 0 (rarely) to 3 (frequently). Scores range from 0 to 60, with higher scores indicating more depressive symptoms. A score of 16 is the critical point for identifying individuals at risk for experiencing depression. The scale was originally designed for use with the general population, but researchers have subsequently validated it for use with mothers of medically at-risk infants (Cho, Holditch-Davis, & Miles, 2008). In the original study validating the scale, Spearman-Brown split-half α coefficients were .85 for the general population and .90 for the patient sample (Radloff, 1977). The CES-D is correlated with other measures of depression (Shafer, 2006), and the four-factor structure of the CES-D—depressed affect, positive affect, somatic activity, and interpersonal relations—is supported for women and longitudinally invariant (Ferro & Speechley, 2013). In the present study, mothers with elevated depressive symptoms (CES-D score > 16) were screened for suicidal ideation and given referrals for treatment by a psychologist.

Infant SE problems

The Ages & Stages Questionnaires: Social-Emotional (ASQ: SE), a behavioral screening tool, was developed to identify social and emotional problems in infants, toddlers, and young children and consists of a series of eight questionnaires (for ages 6, 12, 18, 24, 30, 36, 48, and 60 months) designed to be completed by parents (Squires, Bricker, & Twombly, 2002). Each questionnaire has a 3-month period of flexibility. Thus, in the present study, we used the 6-month ASQ: SE with infants’ age ranging from 3 months through 8 months. The 6-month ASQ: SE consists of five areas (self-regulation, communication, adaptive functioning, affect, and interactions with people) and 19 items, with an additional 3 items generated from the mother’s general concerns and comments. Mothers rate each of the first 19 items on a 3-point scale: Z, rarely or never, 0 points; V, sometimes, 5 points; and X, most of the time, 10 points. The additional 3 items are scored as 5 points each if mothers express that concern about the infants. Scores range from 0 to 205 (190 + 15), with higher scores indicating greater risk for SE problems (Squires et al., 2002). A score of 45 is the cutoff for identifying infants at risk for SE problems. The ASQ: SE was standardized on a national random sample that was representative of the U.S. population including 3,014 preschool-age children and their families. Internal consistency of the ASQ: SE ranged from .67 to .91. Concurrent validity was reported as 81–95%, with an overall agreement of 93% (Squires et al., 2002). In the present study, mothers of the infants with elevated SE problems (ASQ: SE score > 45) were given referrals for consultation by a developmental behavioral pediatrician and psychologist.

Procedure

The Institutional Review Board at the university approved the study. We identified potential infant participants on a daily basis through the NICU admission log. A research nurse discussed the study with the mothers and provided the informed-consent form for review.

Baseline data collection before 40 weeks’ PMA

After maternal consent, the research nurse collected baseline data and saliva samples from both mothers and infants in the NICU before 40 weeks’ PMA, as described previously. The research nurse also collected CES-D data on the day of baseline saliva sample collection.

Saliva-sample collection and analysis

Maternal saliva was obtained 1 hr before or after any oral intake. Three samples of maternal saliva, of 0.1 ml each, were collected within a 2-hr period and measures were averaged to minimize the influence of the episodic pattern of steroid hormone secretion (Rosner, Auchus, Azziz, Sluss, & Raff, 2007). As with the mothers, three 0.1-ml samples of infant saliva were obtained over a 2-hr period. All saliva samples were labeled with study identification numbers, placed on ice, and stored in a freezer at −80°C. Assays were duplicated in batches after all samples were collected. Testosterone and cortisol levels were determined using EIA procedures by a laboratory technician who was blinded to all characteristics of the participants.

Follow-up data collection at 3-months’ and 6-months’ corrected age (CA)

Mothers were asked to visit the research project office with their infants when the infants were at 3-months’ and 6-months’ CA for prematurity. For preterm infants, the term CA is recommended until 3 years of age (Age terminology during the perinatal period, 2004). The research nurse took the infant’s history and physical measurements and then administered the CES-D and ASQ: SE to the mother. The research nurse also addressed any concerns the mother expressed regarding her infant’s health and development and necessary referrals.

Data Analysis

Two-sample t-tests and χ2 tests for contingency tables were used to compare demographic and health characteristics of mothers and infants by infant gender. Analysis of variance was used to compare maternal depressive symptoms at 40 weeks’ PMA and 3 months’ and 6 months’ CA. Two-sample t-tests were used to compare infant SE problems at 3 months’ and 6 months’ CA. Generalized estimating equations (GEEs; Hedeker & Gibbons, 2006) for the entire sample and GEE with subgroup analysis for each infant gender were the main analytic tool for estimating the associations between steroid hormonal levels and repeated measures of maternal depressive symptoms at three time points (40 weeks’ PMA and 3 months’ and 6 months’ CA) and infant SE problems at two time points (3 months’ and 6 months’ CA). GEE is a popular method for analyzing correlated data, particularly useful here in assessing the population-averaged effect of testosterone and cortisol on longitudinal outcomes that were measured at different time points. Maternal characteristics (age, race, education, and income) and infant variables (gender, GA, Apgar scores, medical complications, IVH, hospitalization days, technology dependence, and degree of neurobiological insult) and number of days post birth of saliva collection were controlled statistically. To alleviate multiple testing problems, we applied the false discovery rate-based adjustment to the p values obtained. Adjusted p values of <.05 were deemed statistically significant.

Results

Characteristics of Mothers and Infants by Infant Gender

As shown in Table 1, t-tests and χ2 tests indicate that none of demographic and health characteristics of mothers or infants differed as a function of infant gender, except that female VLBW infants had greater technological dependence and weighed less than males at 3 months’ CA and were shorter than males at 6 months’ CA. Neither testosterone nor cortisol levels of mothers or infants differed by infant gender. In the present study, 56% of infants were the first child and lived with other family members (number of other family members M = 3.34, range 1–11). They lived with other children (M = 1.05, range 0–8) including (step)siblings, an uncle and aunt, or the mothers’ nieces/nephews whose ages ranged between 7 months and 21 years. Family income of 34% of the mothers was either smaller than US$10,000 or between US$10,000 and US$29,999, and 85% of mother–infant pairs were in the Women, Infant, and Children supplemental nutrition program. About 5% of mothers had a family income higher than US$70,000.

Table 1.

Characteristics of Mothers and Infants by Infant Gender.

Overall Male Female
Variable (N = 62) (n = 28) (n = 34) p Valuea
Infants
 Salivary testosterone (pg/mL) 314 (112) 285 (100) 338 (118) .07
 Salivary cortisol (μg/dL) 0.3 (0.3) 0.2 (0.1) 0.4 (0.4) .09
 Gestational age (weeks) 29 (2) 29 (2) 29 (2) .60
 1-min Apgar score, median (IQR) 4 (2) 4 (3) 4 (2) .85
 5-min Apgar score, median (IQR) 7 (2) 6 (2) 7 (1) .18
 CPR at birth (0–6) 2.0 (1.0) 2.0 (1.1) 2.0 (0.8) .98
 Medical complications at <40 weeks’ PMA (0–20) 3.1 (2.4) 3.1 (2.6) 3.2 (2.3) .95
 Health problems at 3 months 2.1 (1.3) 1.8 (1.2) 2.4 (1.4) .60
 Health problems at 6 months 2.3 (1.6) 2.1 (1.5) 2.5 (1.7) .35
 NBRS (0–28) 1.8 (2.5) 2.0 (3.0) 1.7 (2.0) .61
 Days of hospitalization 65 (27) 64 (30) 64 (24) .98
 TD at discharge (0–6) 0.4 (0.9) 0.3 (0.7) 0.4 (1.0) .53
 TD at 3 months (0–6) 1.1 (1.2) 0.8 (1.0) 1.4 (1.3) .05
 TD at 6 months (0–6) 1.0 (1.1) 1.1 (1.2) 1.0 (1.0) .73
 Weight at birth (g) 1,100 (308) 1,129 (315) 1,076 (303) .51
 Length at birth (cm) 37 (4) 37 (4) 37 (4) .72
 Head circumference at birth (cm) 26 (2) 26 (2) 25 (2) .41
 Weight at 3 months (kg) 5.4 (0.9) 5.7 (1.0) 5.2 (0.7) .03
 Length at 3 months (cm) 58 (3) 58 (3) 57 (2) .14
 Head circumference at 3 months (cm) 40 (1) 41 (1) 40 (1) .10
 Weight at 6 months (kg) 6.9 (1.0) 7.1 (0.9) 6.8 (0.7) .15
 Length at 6 months (cm) 64 (3) 65 (3) 63 (3) .01
 Head circumference at 6 months (cm) 43 (1) 43 (1) 43 (1) .09
Mothers
 Salivary testosterone (pg/mL) 52.1 (20.8) 49.2 (14.9) 54.5 (24.6) .32
 Salivary cortisol (μg/dL) 0.2 (0.1) 0.2 (0.1) 0.4 (0.4) .09
 Age (years) 24.9 (5.3) 24.1 (5.0) 25.6 (5.5) .26
 Education (1–5)b 2.6 (1.1) 2.6 (1.0) 2.5 (1.1) .88
 Married (%) 37.0 47.8 35.3 .67c
 Race: White (%) 50.0 42.9 55.9 .31c
 BMI (kg/m2) 32.4 (9.4) 32.7 (8.8) 32.2 (10.0) .84
 Gravida 2.4 (1.8) 2.2 (1.4) 2.6 (2.1) .43
 Parity 1.9 (1.4) 1.6 (1.1) 2.1 (1.6) .14
 No. of complications during pregnancy (0–6) 1.2 (0.8) 1.2 (0.8) 1.3 (0.9) .60
 No. of complications at delivery (0–9) 4.0 (1.3) 3.8 (1.2) 4.3 (1.3) .11
 Delivery: C/S (%) 54.8 57.1 52.9 .74c
Open in a new tab

Note. Values are expressed as mean (SD) unless otherwise indicated. CPR at birth was scored by the presence of treatment including oxygen, bagging/mask, continuous positive airway pressure (CPAP), intubation, chest compression, and epinephrine use. Medical complications were scored before 40 weeks’ PMA based on the presence of medical diagnosis or treatment use including surfactant, antenatal steroid, steroid for broncopulmonary dysplasia (BPD), indomethancin, ventilation/CPAP, patent ductus arteriosus (PDA), surgery, seizure, anticonvulsant, intraventricular hemorrhage (IVH), periventricular leukomalacia (PVL), brain cyst, sepsis, meningitis, parenteral hyperalimentation, necrotizing enterocolitis (NEC), gastrointestinal (GI) perforation, and retinopathy of premature (ROP; examination, diagnosis, and intervention of ROP). Health problems at 3 and 6 months’ corrected age (CA) were scored based on maternal reporting of presence of infant health problems including diarrhea, vomiting, reflux, ear infections, upper respiratory infection, or wheezing. TD was scored based on the continuous use of oxygen, apnea monitor, gastrostomy-tube, ventilator, tracheostomy, or medications. BMI was measured on admission prior to delivery. BMI = body mass index; C/S = Caesarean section; CPR = cardiopulmonary resuscitation; IQR = interquartile range; NBRS = Neurobiologic Risk Score; PMA = postmenstrual age; TD = technology dependence.

a P values are for the t-test for variables between infant genders except those for marital status, race, and delivery. bEducation: 1 = <12th grade; 2 = high school degree; 3 = partial college; 4 = college degree; and 5 = graduate degree. c p Values are for the χ2 test.

Association of Testosterone and Cortisol Levels With Maternal Depressive Symptoms

We examined the associations of maternal testosterone and cortisol levels with maternal depressive symptoms before 40 weeks’ PMA and at 3months’ and 6 months’ CA using GEE and GEE with subgroup analysis separately for each infant gender after adjusting for the covariates. In the entire-sample analysis, maternal depressive symptoms were not associated with either steroid hormonal (testosterone and cortisol) levels or the covariates. In the separate-sample analysis shown in Table 2, maternal testosterone levels were negatively associated with depressive symptoms in mothers of boys but not in mothers of girls. Maternal cortisol levels were not associated with depressive symptoms in mothers of boys or girls. Mothers of boys reported more depressive symptoms when their sons had a lower 5-min Apgar score, more medical complications, more technology dependence, and fewer neurological insults. Depressive symptoms in this group were also associated with less education and lower income. On the other hand, mothers of girls reported more depressive symptoms only when their daughters had lower 1-min Apgar scores. Maternal demographic variables were not associated with depressive symptoms in mothers of girls.

Table 2.

Association of Maternal Testosterone and Cortisol Levels With Maternal Depressive Symptoms (CES-D) Using GEE With Subgroup Analysis for Infant Gender From <40 Weeks’ Postmenstrual Age to 6 Months’ CA.

Mothers of Male Infants (n = 28)
 Mothers of Female Infants (n = 34)
Variable β SE Z p Value β SE Z p Value
Testosterone −0.151 0.066 −2.302 .03 −0.008 0.067 −1.123 .90
Cortisol 8.130 6.976 1.165 .24 −8.317 7.818 −1.064 .29
Months −0.355 0.376 −0.943 .35 0.136 0.388 0.352 .72
Days post birth of saliva collection 0.116 0.076 1.537 .12 −0.078 0.088 −0.889 .37
GA 0.052 0.079 0.653 .51 −0.030 0.184 −0.164 .87
1-min Apgar 0.074 0.533 0.138 .89 −1.694 0.697 −2.429 .02
5-min Apgar −1.174 0.424 −2.770 .01 0.133 1.011 0.132 .90
Medical complications 1.668 0.643 2.594 .01 −0.569 1.086 −0.524 .60
IVH 0.046 1.621 0.028 .98 0.286 1.773 0.161 .87
DOH 0.014 0.045 0.304 .76 0.057 0.083 0.685 .49
TD 4.430 1.617 2.740 .01 −3.085 2.030 −1.520 .13
NRBS −1.354 0.393 −3.446 .01 0.101 0.921 0.109 .91
Age −0.093 0.177 −0.527 .60 0.488 0.369 1.321 .19
Race (White) −2.371 2.237 −1.060 .29 0.010 3.203 0.003 1.00
Education −5.098 0.703 −6.980 .00 0.006 2.021 0.003 1.00
Income −1.232 0.548 −2.250 .03 −1.775 1.530 −1.160 .25
Open in a new tab

Note. Analyses were performed with adjustment for maternal and infant characteristics and days post birth of saliva collection. CES-D were collected at three time points (<40 weeks’ PMA, and 3 months and 6 months’ CA), and combined scores were used for data analysis. CES-D = Center for Epidemiological Depression Scale; CA = corrected age; DOH = days of hospitalization; GA = gestational age; GEEs = generalized estimating equations; IVH = intraventricular hemorrhage; PMA = postmenstrual age; NBRS = Neurobiologic Risk Score; TD = technology dependence (oxygen, apnea, gastrostomy tube, ventilator, tracheostomy, and medications).

As shown in Figure 2, mothers reported the highest level of depressive symptoms when their infant was hospitalized in the NICU. Maternal depressive symptoms decreased from before 40 weeks’ PMA to 3 months’ CA but then increased from 3months’ to 6 months’ CA. Mothers of boys reported fewer depressive symptoms than mothers of girls at all time points, although the differences were not statistically significant. By 6 months’ CA, mothers of boys reported significantly fewer depressive symptoms than they had before 40 weeks’ PMA (β = −.83, standard error [SE] = 0.42, Z = −1.99, p = .05), whereas there was no significant change in depressive symptoms reported by mothers of girls.

Figure 2.

Figure 2.

Open in a new tab

A plot of the original Center for Epidemiological Studies Depression Scale (CES-D) data for mothers of male infants and mothers of female infants before 40 weeks’ postmenstrual age (PMA) and at 3 months’ and 6 months’ corrected age (CA).

Association of Testosterone and Cortisol Levels With Infant SE Problems

We examined the associations between infant steroid hormones (testosterone and cortisol) and infant SE problems at 3 months’ and 6 months’ CA using GEE and GEE with subgroup analysis for each infant gender after adjusting for the covariates. In the entire-sample analysis, we found no associations between steroid hormonal levels and SE problems. Among the covariates, infant SE problems were negatively associated with male gender (β = −8.80, SE = 3.64, Z = −2.42, p = .02) and 5-min Apgar scores (β = −2.86, SE = 1.35, Z = −2.12, p = .04) and positively associated with infant medical complications (β = 3.40, SE = 1.34, Z = 2.55, p = .02). In the separate-sample analysis, as shown in Table 3, maternal report of SE problems in boys was associated with higher GA, more infant medical complications, and longer hospitalization. Mothers reported that SE problems in sons decreased from 3 months’ to 6 months’ CA (22.04 and 17.41, respectively, β = −2.57, SE = 1.05, Z = −2.44, p = .02). Mothers with higher levels of education were less likely to report that their son had SE problems, though this association did not reach significance (p = .06). Maternal report of SE problems in girls was associated with lower infant testosterone levels, lower GA, lower 5-min Apgar scores, more medical complications, lower IVH grade, and shorter hospitalizations. No time effect was found for the interval between 3 months’ and 6 months’ CA (25.73 and 25.15, respectively, β = .02, SE = 0.71, Z = 0.04, p = .97). No maternal demographic variables were associated with infant SE problems in mothers of girls. Cortisol levels were not associated with SE problems in either boys or girls.

Table 3.

Association of Infant Testosterone and Cortisol Levels With Infant Socioemotional Problems (ASQ: SE) Using GEE With Subgroup Analysis for Infant Gender From <40 Weeks’ Postmenstrual Age to 6 Months’ CA.

Male Infants (n = 28)
 Females Infants (n = 34)
Variable β SE Z p Value β SE Z p Value
Testosterone −0.013 0.022 −0.576 .56 −0.081 0.023 −3.555 .01
Cortisol −30.853 20.322 −1.518 .13 11.464 6.745 1.700 .09
Months −2.571 1.052 −2.443 .02 0.025 0.712 0.035 .97
GA 0.529 0.248 2.134 .04 −0.447 0.210 −2.126 .04
1-min Apgar 0.363 1.093 0.332 .74 −0.935 0.869 −1.076 .28
5-min Apgar −3.193 1.937 −1.648 .10 −4.500 1.766 −2.548 .02
Medical complications 3.511 1.589 2.209 .03 5.952 2.362 2.520 .02
IVH 0.538 8.285 0.065 .95 −9.441 3.307 −2.854 .01
DOH 0.281 0.098 2.849 .01 −0.499 0.133 −3.741 .01
TD 0.418 4.040 0.103 .92 2.693 2.162 1.245 .21
NRBS −2.743 2.632 −1.042 .30 −1.705 1.729 −0.986 .32
Age −0.123 0.410 −0.300 .76 −0.027 0.484 −0.055 .96
Race 0.110 3.213 0.034 .97 2.803 4.613 0.608 .54
Education −5.675 3.023 −1.877 .06 2.948 2.643 1.116 .26
Income −2.993 2.412 1.241 .21 −1.488 2.175 −0.684 .49
Open in a new tab

Note. Analyses were performed with adjustment for maternal infant characteristics and days post birth of saliva collection. ASQ: SE were collected at two time points (3 months’ and 6 months’ CA), and combined scores were used for data analysis. ASQ: SE = Ages & Stages Questionnaires: Social-Emotional; CA = corrected age; DOH = days of hospitalization; GA = gestational age; GEE = GEE = generalized estimating equation; IVH = intraventricular hemorrhage, NBRS = Neurobiologic Risk Score; TD = technology dependence (oxygen, apnea, gastrostomy tube, ventilator, tracheostomy, and medications).

Association Between Maternal Depressive Symptoms and Infant SE Problems

We examined the associations between maternal depressive symptoms and maternal report of infant SE problems at 3 months’ and 6 months’ CA using GEE and GEE with subgroup analysis separately for infant gender after adjusting for the covariates. In the entire-sample analysis, mothers with more depressive symptoms reported that their infants had more SE problems than mothers with fewer depressive symptoms (β = .57, SE = 0.13, Z = 4.27, p = .0001). Maternal report of SE problems was negatively associated with male gender (β = −6.23, SE = 2.70, Z = −2.31, p = .03) and 5-min Apgar scores (β = −3.46, SE = 1.16, Z = −2.99, p = .01) and positively associated with GA (β = .27, SE = 0.12, Z = 2.22, p = .03) and infant medical complications (β = 2.99, SE = 1.06, Z = 2.82, p = .01). In the separate-sample analysis, as shown in Table 4, mothers with more depressive symptoms reported that their sons had more SE problems than did mothers with fewer depressive symptoms. Mothers also reported that their sons had more SE problems if they had higher GA, more medical complications, or longer hospitalizations. Mothers reported that their sons had fewer SE problems at 6 months’ CA than at 3 months’ CA. Mothers of girls also reported their daughters had more SE problems when they had more depressive symptoms, themselves, as well as lower maternal testosterone levels. Mothers also reported more SE problems when they had the greater number of days post birth that saliva sample collection occurred. Mothers of girls reported that their infant had more SE problems if the infant had a lower 5-min Apgar score, more medical complications, lower IVH grade, shorter hospitalization, and more technology dependence.

Table 4.

Association Between Maternal Depressive Symptoms and Infant SE Problems Using GEE With Subgroup Analysis for Infant Gender From 3 Months to 6 Months’ CA.

Male Infants (n = 28)
 Females Infants (n = 34)
Variable β SE Z p Value β SE Z p Value
CES-D 0.633 0.237 2.677 .01 0.400 0.140 2.853 .01
Months −2.570 1.077 −2.387 .02 −0.031 0.738 −0.042 .97
No. of days post birth of saliva collection 0.133 0.098 1.351 .18 0.437 0.120 3.651 .01
Maternal testosterone 0.029 0.064 0.451 .65 −0.210 0.075 −2.797 .01
GA 0.570 0.257 2.217 .03 0.089 0.196 0.455 .65
1-min Apgar 0.518 0.940 0.551 .58 −0.802 0.736 −1.091 .28
5-min Apgar −3.404 1.951 −1.745 .09 −3.236 1.296 −2.497 .02
Medical complications 3.145 1.548 2.032 .05 5.995 1.672 3.585 .01
IVH −1.175 7.762 −0.151 .88 −7.892 2.830 −2.789 .01
DOH 0.198 0.080 2.466 .02 −0.626 0.136 −4.606 .00
TD 1.469 2.437 0.603 .55 8.333 2.774 3.004 .01
NRBS −2.016 2.387 −0.844 .40 −0.332 1.152 −0.288 .77
Age 0.317 0.377 0.840 .40 0.172 0.358 0.482 .63
Race (White) −2.126 2.200 −0.966 .33 2.116 4.556 0.464 .64
Education −4.551 3.035 −1.500 .13 2.341 2.120 1.104 .30
Income 2.839 1.842 1.541 .12 −2.629 2.579 −1.019 .31
Open in a new tab

Note. CES-D and ASQ: SE at two time points (3 months’ and 6 months’ CA) were used in this data analysis. ASQ: SE = Ages & Stages Questionnaires: Social-Emotional; CES-D = Center for Epidemiological Studies Depression Scale; CA = corrected age; DOH = days of hospitalization; GA = gestational age; GEE = generalized estimating equation; IVH = intraventricular hemorrhage; NBRS = neurobiologic risk score; SE = socioemotional; TD = technology dependence (oxygen, apnea, gastrostomy tube, ventilator, tracheostomy, and medications).

Discussion

We examined the associations of two steroid hormones, testosterone and cortisol, with two measures of maternal and infant well-being, maternal depressive symptoms, and maternal report of infant SE problems, in mothers and VLBW infants. These associations are known to play a role in mother–infant interactions, parenting (Gray, Edwards, O’Callaghan, Cuskelly, & Gibbons, 2013), and infant cognitive development (Bergman, Sarkar, O’Connor, Modi, & Glover, 2007), which, in turn, are influenced by infant gender (Lagerberg & Magnusson, 2012; Sheldrick et al., 2012). Based on the theories of gender differences related to brain development and social relationships (Baron-Cohen et al., 2005; Geschwind, 1987), we had expected that high levels of testosterone and cortisol in VLBW infants would be positively associated with infant SE problems. However, we found that testosterone levels did not differ by infant gender and were negatively associated with infant SE problems in girls, but not in boys, after adjusting for maternal (age, race, education, and income) and infant (gender, GA, Apgar scores, medical complications, IVH, hospitalization days, technology dependence, and degree of neurological insult) characteristics as well as number of days post birth of saliva collection. Cortisol levels also did not differ by infant gender and did not affect infant SE problems after adjusting for the covariates. As we had expected, mothers with a greater number of depressive symptoms reported that their infants had more SE problems. Mothers of boys who had high maternal testosterone levels reported fewer depressive symptoms, whereas mothers of girls who had high infant testosterone levels reported that their infants had fewer SE problems.

Out of our total sample, 32.3% and 14.5% of mothers were at high risk for depressive symptoms (CES-D score >16) before 40 weeks’ PMA and at 6 months’ CA, respectively, while 8.1% and 6.5% of infants were at high risk for SE problems (ASQ: SE score >45) at 3 months’ and 6 months’ CA, respectively. The prevalence of SE problems in VLBW infants in the present sample was lower than what researchers have reported for toddlers or preschoolers, which makes sense, as the problems are more likely to increase with a child’s age (Brown, Copeland, Sucharew, & Kahn, 2012).

Our finding of the negative association between maternal testosterone levels and depressive symptoms in mothers of male infants is consistent with other studies reporting that women with lower testosterone levels reported greater levels of depression and anxiety (Colangelo et al., 2012; Giltay et al., 2012). In the present study, mothers with high testosterone levels showed fewer depressive symptoms, especially mothers of boys, and fewer infant SE problems, especially mothers of girls. Greater depressive symptoms in mothers of boys was associated with the infant characteristics of lower 5-min Apgar score, more medical complications, more technology dependence, and fewer neurological insults. These mothers also showed greater depressive symptoms when they had fewer years of education and lower income. These findings are consistent with other studies showing that male vulnerability to any developmental disability was worsened by lack of maternal resources (Boyle et al., 2011). Depressive symptoms in mothers of girls were associated only with low 1-min Apgar scores. The influences we found of infant health and maternal demographic variables on maternal depression are consistent with the findings of other studies showing that maternal depressive symptoms are positively associated with poor pregnancy outcomes and negatively associated with availability of maternal resources (Field, 2011). Since testosterone levels are likely to be independent between mothers and infants (Cho, Carlo, Su, & McCormick, 2012), the present findings might be understood based on maternal testosterone levels rather than infant testosterone levels or gender. Maternal cortisol levels were not associated with depressive symptoms, although previous studies have found both positive and negative associations (Corwin et al., 2013; Lommatzsch et al., 2006). It is possible that maternal testosterone levels might be more sensitive than cortisol levels in capturing the differences in maternal depressive symptoms during the postnatal period.

Mothers reported the most depressive symptoms while their infants were hospitalized in the NICU. Depressive symptoms decreased by 3 months’ CA but increased again by 6 months’ CA, especially in mothers of girls. Mothers of VLBW preterm infants would be expected to show improvement in depressive symptoms by 6 months’ CA as their infants recover from medical problems related to prematurity and they, themselves, become familiar with their caregiving role (Kersting et al., 2004). A possible explanation for our finding of an increase in depressive symptoms in mothers of girls from 3 months’ to 6 months’ CA is that the girls in the present study were significantly more dependent on technologies than the boys at 3 months’ CA. It’s also possible that mothers of girls worry about their infants more than mothers of boys, resulting in more depressive symptoms at 6 months’ CA. Mothers of boys showed fewer depressive symptoms than mothers of girls during the study, although the differences were not statistically significant. Depressive symptoms decreased over time only in mothers of boys.

We also examined the associations of infant salivary testosterone and cortisol levels with infant SE problems during the first 6 months’ CA. For boys, we found no association of either testosterone or cortisol level with infant SE problems after adjusting for the covariates. For girls, infant testosterone levels were negatively associated with infant SE problems after adjusting for the covariates. It is possible that the positive association between testosterone and infant SE problems that previous researchers have reported (Azurmendi et al., 2006) is restricted to term healthy boys (Saenz & Alexander, 2013) and does not apply to VLBW preterm infants such as the ones in the present study. Infant cortisol levels were not associated with SE problems in either boys or girls in the present study, possibly because the cortisol levels were not high enough to capture the differences. In general, cortisol levels are lower in preterm than term infants while the preterm infants are in the NICU and up to 3 months’ CA (Grunau et al., 2007; Watterberg, Gerdes, & Cook, 2001). Regarding testosterone, the postnatal period in term infants is a time during which one of three testosterone surges that occur throughout the life course happens, with the other two being in the second trimester of fetal development and puberty (Bouvattier et al., 2002).

After adjusting for the covariates, we found that mothers of boys reported that their infants had more SE problems if the infant was born at an older GA or had more medical complications or a longer hospitalization. The mothers reported that boys, but not girls, had fewer SE problems over time. On the other hand, mothers of girls reported that their infants had more SE problems if the infant was born at a younger GA or had a lower 5-min Apgar score, more medical complications, lower IVH grade, and fewer hospitalization days. After adjusting for the covariates, we also found that mothers of girls who, themselves, had higher testosterone levels reported that their infants had fewer SE problems than did mothers of girls with lower testosterone levels. Number of medical complications was the only factor positively associated with infant SE problems in both genders, suggesting an influence of infant health (and possibly worry about infant health) on maternal report of infant SE problems.

Finally, we examined the associations between maternal depressive symptoms and infant SE problems. As expected, maternal depressive symptoms were positively associated with maternal report of infant SE problems, regardless of infant gender, even after controlling for the covariates. This finding is consistent with those of other studies, showing that mothers with more psychological problems perceived their infants as more of a burden and more difficult (Austin, Hadzi-Pavlovic, Leader, Saint, & Parker, 2005; Palmer et al., 2013). Along with the negative association we found between maternal testosterone levels and infant SE problems, this finding suggests that high maternal testosterone level could reduce maternal perception of infant SE problems possibly by reducing the risk of depressive symptoms.

A limitation of the present study is that 6 months’ CA does not provide sufficient time to explore all possible associations between steroid hormonal levels and psychological and behavioral problems of mothers and infants. Though we applied the false discovery rate-based adjustment to control the proportion of false positives, which could have occurred because of the multiple testing, the findings from this exploratory study require replication before clinical applications are explored. We collected saliva samples only once in the postnatal period, but it is necessary to examine longitudinal associations between steroidal hormone levels and maternal depression and infant/child SE problems during late infancy and early childhood before making any further conclusions because different levels of hormones might change the associations. However, our findings do suggest that a high level of maternal testosterone could have a protective effect for the variables of maternal depressive symptoms and maternal report of infant SE problems in mothers of VLBW infants to 6 months’ CA.

Footnotes

Author Contributions: JC contributed to conception and design contributed to acquisition, analysis, and interpretation; drafted manuscript; critically revised manuscript, gave final approval, and agreed to be accountable for all aspects of work ensuring integrity and accuracy. XS contributed to design, analysis, and interpretation; drafted the manuscript; critically revised the manuscript; gave final approval agrees to be accountable for all aspects of work ensuring integrity and accuracy. VP contributed to design and acquisition; drafted manuscript; critically revised the manuscript; gave final approval; and agreed to be accountable for all aspects of work ensuring integrity and accuracy. DH-D contributed to conception, design, and interpretation; drafted manuscript; critically revised the manuscript; gave final approval; and agreed to be accountable for all aspects of work ensuring integrity and accuracy.

Declaration of Conflicting Interests: The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding: The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The preparation of this artilce was supported by the National Institute of Child Health and Human Development (R21HD066186), National Institutes of Health.

References

  1. Achermann J. C. (2005). Development of the reproductive systems In Brook C. G. D., Clayton P. E., Brown R. S., Savage M. O. (Eds.), Clinical pediatric endocrinology (5th ed, pp. 153–170). Malden, MA: Blackwell Publishing. [Google Scholar]
  2. Austin M.-P., Hadzi-Pavlovic D., Leader L., Saint K., Parker G. (2005). Maternal trait anxiety, depression and life event stress in pregnancy: Relationships with infant temperament. Early Human Development, 81, 183–190. [DOI] [PubMed] [Google Scholar]
  3. Azurmendi A., Braza F., Garcia A., Braza P., Munoz J. M., Sanchez-Martin J. R. (2006). Aggression, dominance, and affiliation: Their relationships with androgen levels and intelligence in 5-year-old children. Hormones and Behavior, 50, 132–140. [DOI] [PubMed] [Google Scholar]
  4. Baibazarova E., van de Beek C., Cohen-Kettenis P. T., Buitelaar J., Shelton K. H., van Goozen S. H. (2013). Influence of prenatal maternal stress, maternal plasma cortisol and cortisol in the amniotic fluid on birth outcomes and child temperament at 3 months. Psychoneuroendocrinology, 38, 907–915. doi:10.1016/j.psyneuen.2012.09.015 [DOI] [PubMed] [Google Scholar]
  5. Baron-Cohen S., Knickmeyer R. C., Belmonte M. K. (2005). Sex differences in the brain: Implications for explaining autism. Science, 310, 819–823. [DOI] [PubMed] [Google Scholar]
  6. Bergman K., Sarkar P., O’Connor T. G., Modi N., Glover V. (2007). Maternal stress during pregnancy predicts cognitive ability and fearfulness in infancy. Journal of American Academy of Child and Adolescent Psychiatry, 46, 1454–1463. doi:10.1097/ chi.0b013e31814a62f6 [DOI] [PubMed] [Google Scholar]
  7. Blair C., Raver C. C., Granger D., Mills-Koonce R., Hibel L., & Family Life Project Key Investigators. (2011). Allostasis and allostatic load in the context of poverty in early childhood. Development and Psychopathology, 23, 845–857. doi:10.1017/S0954579411000344 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bouvattier C., Carel J. C., Lecointre C., David A., Sultan C., Bertrand A. M.…Chaussain J. L. (2002). Postnatal changes of T, LH, and FSH in 46, XY infants with mutations in the AR gene. Journal of Clinical Endocrinology & Metabolism, 87, 29–32. [DOI] [PubMed] [Google Scholar]
  9. Boyle C. A., Boulet S., Schieve L. A., Cohen R. A., Blumberg S. J., Yeargin-Allsopp M.…Kogan M. D. (2011). Trends in the prevalence of developmental disabilities in US children, 1997-2008. Pediatrics, 127, 1034–1042. doi:10.1542/peds.2010-2989 [DOI] [PubMed] [Google Scholar]
  10. Brand S. R., Engel S. M., Canfield R. L., Yehuda R. (2006). The effect of maternal PTSD following in utero trauma exposure on behavior and temperament in the 9-month-old infant. Annals of the New York Academy of Sciences, 1071, 454–458. doi:10.1196/ annals.1364.041 [DOI] [PubMed] [Google Scholar]
  11. Brazy J. E., Goldstein R. F., Oehler J. M., Gustafson K. E., Thompson R. J., Jr (1993). Nursery neurobiologic risk score: Levels of risk and relationships with nonmedical factors. Journal of Developmental & Behavioral Pediatrics, 14, 375–380. [PubMed] [Google Scholar]
  12. Brown C. M., Copeland K. A., Sucharew H., Kahn R. S. (2012). Social-emotional problems in preschool-aged children: Opportunities for prevention and early intervention. Archives of Pediatrics and Adolescent Medicine, 166, 926–932. doi:10.1001/archpediatrics.2012.793 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Buckwalter J. G., Stanczyk F. Z., McCleary C. A., Bluestein B. W., Buckwalter D. K., Rankin K. P.…Goodwin T. M. (1999). Pregnancy, the postpartum, and steroid hormones: Effects on cognition and mood. Psychoneuroendocrinology, 24, 69–84. [DOI] [PubMed] [Google Scholar]
  14. Cho J., Carlo W. A., Su X., McCormick K. L. (2012). Associations between salivary testosterone and cortisol levels and neonatal health and growth outcomes. Early Human Development, 88, 789–795. doi:10.1016/j.earlhumdev.2012.05.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Cho J., Holditch-Davis D., Miles M. S. (2008). Effects of maternal depressive symptoms and infant gender on the interactions between mothers and their medically at-risk infants. Journal of Obstetric, Gynecologic, and Neonatal Nursing, 37, 58–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Colangelo L. A., Craft L. L., Ouyang P., Liu K., Schreiner P. J., Michos E. D., Gapstur S. M. (2012). Association of sex hormones and sex hormone-binding globulin with depressive symptoms in postmenopausal women: The Multiethnic Study of Atherosclerosis. Menopause, 19, 877–885. doi:10.1097/gme.0b013e3182432de6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Corwin E. J., Guo Y., Pajer K., Lowe N., McCarthy D., Schmiege S.…Stafford B. (2013). Immune dysregulation and glucocorticoid resistance in minority and low income pregnant women. Psychoneuroendocrinology, 38, 1786–1796. doi:10.1016/j.psyneuen.2013.02.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. de Weerth C., Zijl R. H., Buitelaar J. K. (2003). Development of cortisol circadian rhythm in infancy. Early Human Development, 73, 39–52. [DOI] [PubMed] [Google Scholar]
  19. Denson T. F., Mehta P. H., Ho Tan D. (2013). Endogenous testosterone and cortisol jointly influence reactive aggression in women. Psychoneuroendocrinology, 38, 416–424. [DOI] [PubMed] [Google Scholar]
  20. Engle WA; American Academy of Pediatrics Committee on Fetus and Newborn. (2004). Age terminology during the perinatal period. Pediatrics, 114, 1362–1364. [DOI] [PubMed] [Google Scholar]
  21. Ferro M. A., Speechley K. N. (2013). Factor structure and longitudinal invariance of the Center for Epidemiological Studies Depression Scale (CES-D) in adult women: Application in a population-based sample of mothers of children with epilepsy. Archives of Women’s Mental Health, 16, 159–166. doi:10.1007/s00737-013-0331-5 [DOI] [PubMed] [Google Scholar]
  22. Field T. (2011). Prenatal depression effects on early development: A review. Infant Behavior and Development, 34, 1–14. doi:10.1016/j.infbeh.2010.09.008 [DOI] [PubMed] [Google Scholar]
  23. Forcada-Guex M., Pierrehumbert B., Borghini A., Moessinger A., Muller-Nix C. (2006). Early dyadic patterns of mother-infant interactions and outcomes of prematurity at 18 months. Pediatrics, 118, e107–e114. doi:10.1542/peds.2005-1145 [DOI] [PubMed] [Google Scholar]
  24. Geschwind N. G. A. M. (1987). Cerebral lateralization: Biological mechanisms, associations, and pathology. Cambridge, MA: MIT Press. [Google Scholar]
  25. Giltay E. J., Enter D., Zitman F. G., Penninx B. W., van Pelt J., Spinhoven P., Roelofs K. (2012). Salivary testosterone: Associations with depression, anxiety disorders, and antidepressant use in a large cohort study. Journal of Psychosomatic Research, 72, 205–213. doi:10.1016/j.jpsychores.2011.11.014 [DOI] [PubMed] [Google Scholar]
  26. Glynn L. M., Davis E. P., Schetter C. D., Chicz-DeMet A., Hobel C. J., Sandman C. A. (2007). Postnatal maternal cortisol levels predict temperament in healthy breastfed infants. Early Human Development, 83, 675–681. [DOI] [PubMed] [Google Scholar]
  27. Gray P. H., Edwards D. M., O’Callaghan M. J., Cuskelly M., Gibbons K. (2013). Parenting stress in mothers of very preterm infants—Influence of development, temperament and maternal depression. Early Human Development, 89, 625–629. [DOI] [PubMed] [Google Scholar]
  28. Grunau R. E., Haley D. W., Whitfield M. F., Weinberg J., Yu W., Thessen P. (2007). Altered basal cortisol levels at 3, 6, 8 and 18 months in infants born at extremely low gestational age. Journal of Pediatrics, 150, 151–156. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Gunnar M. R., Donzella B. (2002). Social regulation of the cortisol levels in early human development. Psychoneuroendocrinology, 27, 199–220. [DOI] [PubMed] [Google Scholar]
  30. Hedeker D., Gibbons R. D. (2006). Longitudinal data analysis. New York, NY: John Wiley. [Google Scholar]
  31. Kersting A., Dorsch M., Wesselmann U., Ludorff K., Witthaut J., Ohrmann P.…Arolt V. (2004). Maternal posttraumatic stress response after the birth of a very low-birth-weight infant. Journal of Psychosomatic Research, 57, 473–476. [DOI] [PubMed] [Google Scholar]
  32. Lagerberg D., Magnusson M. (2012). Infant gender and postpartum sadness in the light of region of birth and some other factors: A contribution to the knowledge of postpartum depression. Archives of Women’s Mental Health, 15, 121–130. [DOI] [PubMed] [Google Scholar]
  33. Lommatzsch M., Hornych K., Zingler C., Schuff-Werner P., Hoppner J., Virchow J. C. (2006). Maternal serum concentrations of BDNF and depression in the perinatal period. Psychoneuroendocrinology, 31, 388–394. doi:10.1016/j.psyneuen.2005.09.003 [DOI] [PubMed] [Google Scholar]
  34. Lutchmaya S., Baron-Cohen S., Raggatt P. (2002). Foetal testosterone and eye contact in 12-month-old human infants. Infant Behavior & Development, 25, 327–335. [Google Scholar]
  35. Muller-Nix C., Forcada-Guex M., Pierrehumbert B., Jaunin L., Borghini A., Ansermet F. (2004). Prematurity, maternal stress and mother-child interactions. Early Human Development, 79, 145–158. [DOI] [PubMed] [Google Scholar]
  36. Palmer F. B., Anand K. J., Graff J. C., Murphy L. E., Qu Y., Volgyi E.…Tylavsky F. A. (2013). Early adversity, socioemotional development, and stress in urban 1-year-old children. Journal of Pediatrics, 163, 1733–1739. doi:10.1016/j.jpeds.2013.08.030 [DOI] [PubMed] [Google Scholar]
  37. Radloff L. S. (1977). The CES-D scale: A self-report depression scale for research in the general population. Journal of Applied Psychological Measurement, 1, 385–401. [Google Scholar]
  38. Rosner W., Auchus R. J., Azziz R., Sluss P. M., Raff H. (2007). Position statement: Utility, limitations, and pitfalls in measuring testosterone: An Endocrine Society position statement. Journal of Clinical Endocrinology & Metabolism, 92, 405–413. [DOI] [PubMed] [Google Scholar]
  39. Saenz J., Alexander G. M. (2013). Postnatal testosterone levels and disorder relevant behavior in the second year of life. Biological Psychology, 94, 152–159. [DOI] [PubMed] [Google Scholar]
  40. Saleh el-S., El-Bahei W., Del El-Hadidy M. A., Zayed A. (2013). Predictors of postpartum depression in a sample of Egyptian women. Journal of Neurosychiatric Disease and Treatment, 9, 15–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Shafer A. B. (2006). Meta-analysis of the factor structures of four depression questionnaires: Beck, CES-D, Hamilton, and Zung. Journal of Clinical Psychology, 62, 123–146. [DOI] [PubMed] [Google Scholar]
  42. Sheldrick R. C., Neger E. N., Perrin E. C. (2012). Concerns about development, behavior, and learning among parents seeking pediatric care. Journal of Developmental & Behavioral Pediatrics, 33, 156–160. doi:10.1097/DBP.0b013e3182420f4a [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Squires J., Bricker D., Twombly L. (2002). The ages and stages questionnaires: Social-emotional. A parent-completed, child-monitoring system for social-emotional behaviors. Baltimore, MD: Brookes. [Google Scholar]
  44. van Bakel H. J. A., Riksen-Walraven J. M. (2004). Stress reactivity in 15-month-old infants: Links with infant temperament, cognitive competence, and attachment security. Developmental Psychobiology, 44, 157–167. doi:10.1002/dev.20001 [DOI] [PubMed] [Google Scholar]
  45. Wada K., Nagata C., Nakamura K., Iwasa S., Shiraki M., Shimizu H. (2012). Light exposure at night, sleep duration and sex hormone levels in pregnant Japanese women. Endocrine Journal, 59, 393–398. [DOI] [PubMed] [Google Scholar]
  46. Watterberg K. L., Gerdes J. S., Cook K. L. (2001). Impaired glucocorticoid synthesis in premature infants developing chronic lung disease. Pediatric Research, 50, 190–195. [DOI] [PubMed] [Google Scholar]
  47. Weinberg M. K., Olson K. L., Beeghly M., Tronick E. Z. (2006). Making up is hard to do, especially for mothers with high levels of depressive symptoms and their infant sons. Journal of Child Psychology and Psychiatry, 47, 670–683. [DOI] [PubMed] [Google Scholar]

Articles from Biological Research for Nursing are provided here courtesy of SAGE Publications

RESOURCES