UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Perinatal health epidemiology in multi-ethnic Amsterdam: psychobiological processes

de Wolf, G.

Link to publication

Citation for published version (APA):de Wolf, G. (2010). Perinatal health epidemiology in multi-ethnic Amsterdam: psychobiological processes.

General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.

Download date: 29 Jun 2020

Geertje GoedhartAnne C. Snijders

Arlette E. HesselinkMireille N. van Poppel

Gouke J. BonselTanja G.M. Vrijkotte

Chapter 5Maternal depressive symptoms in relation to perinatal mortality and morbidity: results from a large multi-ethnic cohort study.

Psychosomatic Medicine; in press.

64

Chapter 5 _________________________________________________________________________

Abstract

Objective This study explored (i) whether maternal depressive symptoms during pregnancy are associated with preterm birth (PTB), small-for-gestational-age (SGA), a low Apgar score and child loss, (ii) whether maternal smoking mediates the associations and (iii) whether the associations differ by ethnic background.

Methods Pregnant women in Amsterdam were approached during their first prenatal visit to par-ticipate in the Amsterdam Born Children and their Development (ABCD) study. They filled out a pregnancy questionnaire covering sociodemographic data, lifestyle and (psychosocial) health. Depressive symptoms were assessed with the CES-D scale. The baseline sample consisted of 8052 women; main ethnic groups were: Dutch, Creole, Turkish and Moroccan. Multiple logistic regres-sion analysis was performed.

Results The prevalence of perinatal outcomes was: 5.4% (PTB), 12.3% (SGA), 1.5% (low Apgar score) and 1.4% (child loss). The prevalence of high depressive symptomatology was 30.6%. After adjustment for maternal age, parity, education, ethnicity, pre-pregnancy BMI, hypertension, alcohol and drug use and a small mediation effect of maternal smoking, high vs. low levels of depressive symptoms were associated with SGA (OR 1.19, p = .02) and a low Apgar score (OR 1.74, p = .01), but not with PTB (OR 1.16, p = .18) and child loss (OR 1.28, p = .24). Stratified analyses by ethnic background showed a tendency towards higher risks, though insignificant, among Creole women.

Conclusions Several pathways may explain the detrimental effects of maternal depressive symp-tomatology on perinatal health outcomes, including a psychoendocrinological pathway involving the hormone cortisol or mediation effects by maternal risk behaviors. Further research should explore the underlying pathways, in particular among ethnic subgroups.

65

_______________________________________________ Maternal depression and perinatal outcomes

5

Introduction

While perinatal mortality rates decreased over the last decades due to improved perinatal health care, perinatal morbidity rates increased.1-4 This is not without long-term clinical, social and economical consequences; major indicators of perinatal morbidity, including small-for-gestational-age (SGA), preterm birth (PTB) and a low Apgar score, have been shown to be related to adverse health outcomes in adult life. PTB, for example, may result in cognitive, behavioral or emotional problems.4, 5 Babies born SGA are at an increased risk for cardiovascular disease, type 2 diabetes and psychomotor and intellectual impairment.6, 7 A low Apgar score is related to an increased risk for cerebral palsy, epilepsy, neurologic disability and low cognitive function.2, 8 This emphasizes the urgent need for prevention of not only perinatal mortality but also perinatal morbidity, by exploring and acting on their causal factors.

One of the causal factors for perinatal mortality and morbidity may be maternal depression during pregnancy. It is hypothesized that depression results in neuroendocrine alterations, including a hyper- or hypoactivity of the hypothalamic-pituitary-adrenal (HPA) axis.9-11 The hormonal end products of the HPA-axis, such as cortisol and norepinephrine, may in turn affect uterine artery blood flow, fetal development and growth, and parturition. Maternal depression may also indirectly affect perinatal health outcomes through mediation by maternal risk behaviors. Depressed women, for example, more often continue to smoke during pregnancy12 and in turn, smoking is an important risk factor for adverse perinatal health outcomes;13-16 hence, maternal smoking is a potential mediator.

It has been observed that fetuses of depressed women are more likely to have a reduced growth17-21 – resulting in SGA babies – and to be born preterm.17-20, 22-25 Other studies, however, did not observe an association between maternal depression and SGA or PTB.17, 21, 22, 26-28 One of the explanations for the inconsistent results is the limited comparability of the studies; they differ with respect to several methodological issues like the measurement of maternal depression (mental disorder21, 27 vs. depressive symptomatology20, 24, 28), the time window during pregnancy when maternal depression is measured (first24, second27 or third26 trimester), the study population (teenagers20, 21 vs. adults20, white vs. black25), and the study design (retrospective21 vs. prospective24). It is also hypothesized that offspring of depressed women are more likely to have an impaired general condition at birth, as measured by the Apgar score,29 or to die during the perinatal period, but this is – so far – hardly confirmed.21, 26, 28, 30, 31 As the global prevalence of mental illness, including prenatal depression, is rising,32 it becomes more and more important to study the adverse effects of maternal depression on perinatal mortality and morbidity in large prospective birth cohort studies.

Ethnic minority groups often have a high prevalence of both prenatal depression and adverse perinatal health outcomes. Although previous studies showed that ethnic disparities in perinatal health outcomes could only to a very small part be explained by levels of depressive symptoms,33-35 there might be interactive effects of depression with ethnic background. It has for example been suggested that, compared to white women, black women are more susceptible for stress-induced neuroendocrine and inflammatory pathways into adverse perinatal outcomes, possibly as a result of higher vascular (re)activity.35-38

66

Chapter 5 _________________________________________________________________________

As part of a prospective multi-ethnic cohort study among pregnant women in Amsterdam, we explored (i) whether maternal depressive symptoms during pregnancy are associated with four major perinatal health outcomes, i.e. PTB, SGA, a low Apgar score and child loss, (ii) whether maternal smoking during pregnancy mediates these associations and (iii) whether the associations are modified by ethnic background. As far as we know, this is the largest study in this specific research field.

Methods

The present study is part of the Amsterdam Born Children and their Development (ABCD) study. The ABCD study is a prospective community-based cohort study aimed at examining the relationship between maternal lifestyle and psychosocial conditions during pregnancy and the child’s health at birth as well as in later life. Approval of the study was obtained from the Central Committee on Research involving Human Subjects, the Medical Ethical Committees of participating hospitals and the Registration Committee of Amsterdam.

Setting and populationBetween January 2003 and March 2004, all pregnant women in Amsterdam were invited to enroll in the ABCD study at their first prenatal visit to participating obstetric care providers (around the 12th week of gestation). All approached women (n = 12 373) received a pregnancy questionnaire; 8266 women returned the questionnaire (response rate 67%). For the current study, women with multiple births were excluded. The baseline sample consisted of 8050 women who had complete data on the independent variable (CES-D score) and relevant covariates.

Data collectionTwo weeks after their first prenatal visit, a pregnancy questionnaire was sent to the women’s home address, to be returned by prepaid mail. The questionnaire covered sociodemographic data, lifestyle and (psychosocial) health. Reminders were sent 2 weeks after the initial mailing. Questionnaires were in Dutch and, depending on the woman’s country of birth, accompanied by an English, Turkish or Arabic copy. Information on pregnancy outcomes was obtained from two sources: (1) the Youth Health Care Registration at the Public Health Service in Amsterdam and (2) the Dutch Perinatal Registration (PRN) (individual linkage possible for 80% of our sample), which include the national obstetric databases for midwives and gynecologists, and the national neonatal database for pediatricians and neonatologists. Additional information on child loss was gathered from the mother and health care providers.

Maternal depressive symptomsThe level of maternal depressive symptoms during pregnancy was measured by self-report in the pregnancy questionnaire using the Center for Epidemiologic Studies Depression scale (CES-D). The CES-D is designed to determine depressive symptomatology in the week previous to the

67

_______________________________________________ Maternal depression and perinatal outcomes

5

acquisition of the questionnaire and does not aim to identify chronic or clinical depression.39 Nevertheless, scores on the CES-D scale correlate quite well with clinical assessments and a score of 16 or higher is commonly used to define high-risk groups or possible cases of clinical depression.24,

39, 40 In this study, we also applied the cut-off point of 16 to dichotomize the group into low and high depressive symptomatology. The CES-D is a 20-item scale which was found to have good validity, reliability and factor structure.39, 41 The scale has been translated into different languages; in this study, the validated Dutch version was used.42 Although the CES-D was initially designed for use in the general population, it has been demonstrated to be a valid and reliable measurement tool in different subgroups, including pregnant24, 43 and black39, 44, 45 women. Reliability analysis in our sample of pregnant women resulted in a Cronbach’s Alpha of 0.90 for the total sample of pregnant women; reliability was comparable among ethnic subgroups (Cronbach’s Alpha ranging from 0.89 to 0.91).

Perinatal health outcomesFour major perinatal health outcomes were explored, i.e. preterm birth (PTB), small-for-gestational-age (SGA), low Apgar score, and child loss. PTB was defined as a delivery between 24.0 and 36.6 weeks of completed gestation. Data on gestational duration were based on ultrasound or, when unavailable (<10%), timing of last menstrual period. Complete data for PTB analysis were available for 7550 women. Newborns were categorized as SGA when they had a birthweight below the 10th percentile for gestational age on the basis of gender- and parity-specific standards from the PRN. For SGA analysis, we excluded births below 24.0 weeks of gestation; finally, 7530 women had complete data for this analysis. The Apgar score is commonly used to measure the general condition of a neonate immediately after birth. The obstetric care provider measures the newborn’s condition at 1, 5 and 10 minutes after birth by paying attention to the appearance, pulse, grimace, activity and respiration of the baby.29 At every item, a score of 0, 1 or 2 points can be obtained, with a maximum total score of 10. A low Apgar score was defined as below 7. Only Apgar scores measured at 5 minutes after birth were analyzed, since this score is regarded as the best predictor of survival in infancy.46, 47 For this specific analysis, fetal deaths were excluded; 6390 women had complete data for the analysis on Apgar scores. Missing data for Apgar scores were mainly due to incomplete linkage with PRN data. Child loss included miscarriages (<22 weeks of gestation), fetal deaths (22 weeks of gestation – delivery) and early neonatal deaths (0 – 7 days after delivery). Non-spontaneous abortions (n = 14) and unknown child losses (n = 158) were excluded, leaving 7901 women with complete data for the child loss analysis.

CovariatesThe following covariates were measured in the pregnancy questionnaire and included in the analysis: maternal age, parity (0, ≥1), education (years of education after primary school), ethnicity [based on the country of birth of the participant’s mother: Dutch, Creole (Surinamese-Creole, Antillean/Aruban and Ghanaian), Turkish, Moroccan and other non-Dutch countries], maternal pre-pregnancy body mass index (BMI), hypertensive disorder (no/unknown, chronic hypertension

68

Chapter 5 _________________________________________________________________________

and pregnancy-induced hypertension), alcohol consumption (no/yes) and drug use (no/yes) during pregnancy. Maternal smoking during pregnancy (no/yes) was included as a potential mediator. Chronic and pregnancy-induced hypertension were defined combining self-reported data from the questionnaires and data from the PRN registration. Chronic hypertension was the case if (a) pre-existent hypertension was recorded in the PRN, or if (b) women reported high blood pressure and/or the use of medicines against high blood pressure before pregnancy or in the first 20 weeks of pregnancy. Pregnancy-induced hypertension was the case if (a) pregnancy-related hypertension, eclampsia or preeclampsia was recorded in the PRN, or if (b) women without pre-existent hypertension reported high blood pressure and/or the use of medicines against high blood pressure during pregnancy.

Statistical analysisDescriptive statistics were used to profile the sample characteristics according to the level of maternal depressive symptoms; differences were tested with ANOVA for continuous variables and chi-square tests for categorical variables. To examine the association between maternal depressive symptoms (high vs. low) and perinatal health outcomes, multiple logistic regression analyses were performed, adjusting for maternal age, parity, education, ethnicity, pre-pregnancy BMI, hypertension, alcohol and drug use (Model 1); maternal age, education and pre-pregnancy BMI were included as continuous variables. For SGA analysis, the quadratic terms for maternal age and BMI were also included, while parity was excluded (the definition of SGA already account for the parity effect).

To examine whether maternal smoking during pregnancy is a mediator in the associations between maternal depressive symptoms and perinatal health outcomes, the following conditions must hold: (1) the independent variable must affect the mediator, (2) the independent variable must affect the outcome variable, (3) the mediator must affect the outcome variable, (4) the effect of the independent variable on the outcome variable must be less when controlling for the effect of the mediator.48 Condition 1 and 3 were tested with descriptive statistics. Condition 2 and 4 were tested with logistic regression analysis by additionally entering maternal smoking to the adjusted model (Model 2).

To examine a possible interaction effect of maternal depressive symptoms with ethnic background, interaction terms were tested by logistic regression analysis.48 In addition, stratified analyses by the main ethnic groups (Dutch, Creole, Turkish and Moroccan) were performed for all four perinatal health outcomes. Data were analyzed using SPSS version 17.0 (SPSS Inc., Chicago, IL, USA). In all analyses, a p-value <0.05 was considered significant.

Results

Of the 8050 women in the baseline sample, 2465 (30.6%) reported high levels of depressive symptoms during pregnancy (Table 5.1). These women were significantly (p < .001) younger,

69

_______________________________________________ Maternal depression and perinatal outcomes

5

less often nulliparous and lower educated compared to the women who reported low levels of depressive symptoms. The average pre-pregnancy BMI (23.7 vs. 22.9 kg/m2) and the use of cigarettes (14.4% vs. 7.4%) and drugs (3.9% vs. 1.8%) during pregnancy was furthermore higher among these women. The use of alcohol during pregnancy was, however, higher among the women with low depressive symptomatology (23.0% vs. 16.4%). In our sample, 4270 women (53%) were of native Dutch origin, 640 women (8%) had a Creole background, 402 (5%) a Turkish background, 690 (8.5%) a Moroccan background and the final 2048 women (25.5%) had an origin in another non-Dutch country. While already 22% of the Dutch women reported high levels of depressive symptoms, this percentage was even higher among the Creole (42.8%), Turkish (55.7%), Moroccan (42.0%) and other non-Dutch group (35.6%).

In our sample, we observed a prevalence of 5.4% for PTB, 12.3% for SGA, 1.5% for a low Apgar score and 1.4% for child loss. Univariate logistic regression analysis showed that the prevalence of all four outcomes was significantly higher among the group of women with high levels of depressive symptoms (Table 5.2). After adjustment for relevant covariates (maternal age, parity, education, ethnicity, pre-pregnancy BMI, hypertension, alcohol and drug use; Model 1), the significant association of high vs. low depressive symptoms with PTB (OR 1.21, p = .10) and child loss (OR 1.32, p = .18) disappeared, while the associations with SGA (OR 1.25, p = .004) and a low Apgar score (OR 1.77, p = .01) remained statistically significant.

To examine whether maternal smoking during pregnancy mediates the association between maternal depressive symptoms and perinatal outcomes, we tested the conditions for mediation.48 Maternal depressive symptomatology was associated with maternal smoking prevalence during pregnancy (p < .001, Table 1). Maternal smoking was associated with PTB (p < .001) and SGA (p < .001), but not with a low Apgar score (p = .11) and child loss (p = .06) (results not shown). By adding maternal smoking to the adjusted logistic regression models, the effect of maternal depressive symptoms on the perinatal outcomes slightly decreased (Model 2, Table 5.2). All together, maternal smoking can be considered to be a partial mediator in the association between maternal depressive symptoms and both PTB and SGA. After adjustment for maternal smoking, the association of high vs. low depressive symptoms with SGA (OR 1.19, p = .02) and a low Apgar score (OR 1.74, p = .01) was statistically significant, while the association with PTB (OR 1.61, p = .18) and child loss (OR 1.28, p = .24) was insignificant.

The inclusion of an interaction term between depressive symptoms and ethnicity to the full logistic regression models resulted in insignificant interaction effects (p-values ranging from .367 to .903; results not shown). Stratified analyses by ethnic group, however, showed a tendency towards a stronger association of high vs. low levels of maternal depressive symptoms with all four perinatal health outcomes among the Creole group compared to the other ethnic groups (Table 5.3a-d).

70

Chapter 5 _________________________________________________________________________

Table 5.1 Sample characteristics (n = 8050) according to level of maternal depressive symptoms.

Maternal depressive symptoms

Low (n = 5585) High (n = 2465) pa

CES-D score (mean ± sd) 8.0 ± 4.0 23.4 ± 6.8

Maternal age (mean years ± sd) 31.3 ± 5.0 29.7 ± 5.7 < 0.001

Parity (% nulliparae) 58.2 50.9 < 0.001

Ethnicity (%)

Dutch (n = 4270) 77.8 22.2 < 0.001

Creole (n = 640) 57.2 42.8

Turkish (n = 402) 44.3 55.7

Moroccan (n = 690) 58.0 42.0

Other (n = 2048) 64.4 35.6

Education (mean years ± sd) 9.2 ± 3.9 7.5 ± 4.1 < 0.001

Maternal BMI (mean kg/m2 ± sd) 22.9 ± 3.8 23.7 ± 4.6 < 0.001

Hypertensive disorder (%)

No/unknown (n = 7145) 69.6 30.4 < 0.001

Chronic hypertension (n = 326) 54.0 46.0

Pregnancy-induced hypertension (n = 579) 74.8 25.2

Maternal smoking (% yes) 7.4 14.4 < 0.001

Alcohol use (% yes) 23.0 16.4 < 0.001

Drug use (% yes) 1.8 3.9 < 0.001aTested with ANOVA (continuous variables) or chi-square test (categorical variables).

Table 5.2 Associations between maternal depressive symptoms and perinatal health outcomes.Depressive symptoms

Perinatal outcomes

(%)

Crude model OR (95% CI)

Model 1a

OR (95% CI)Model 2b

OR (95% CI)

PTB (n = 7550) Low 5.0 1.00 1.00 1.00

High 6.3 1.26 (1.02 – 1.56) 1.21 (0.97 – 1.51) 1.16 (0.93 – 1.45)

SGA (n = 7530) Low 11.1 1.00 1.00 1.00

High 15.2 1.44 (1.25 – 1.67) 1.25 (1.07 – 1.45) 1.19 (1.02 – 1.39)

Low Apgar score (n = 6390) Low 1.2 1.00 1.00 1.00

High 2.1 1.73 (1.15 – 2.60) 1.77 (1.15 – 2.72) 1.74 (1.13 – 2.69)

Child loss (n = 7901) Low 1.2 1.00 1.00 1.00

High 1.8 1.48 (1.01 – 2.18) 1.32 (0.88 – 1.97) 1.28 (0.85 – 1.91)

a Adjusted for maternal age, parity, education, ethnicity, pre-pregnancy BMI, hypertension, alcohol and drug use; SGA analysis: additional inclusion of quadratic term for maternal age and pre-pregnancy BMI, exclusion of parity because SGA is by definition corrected for parity.

bAdditionally adjusted for maternal smoking during pregnancy (potential mediator).

71

_______________________________________________ Maternal depression and perinatal outcomes

5

Table 5.3a Association between maternal depressive symptoms and SGA, separated by ethnicity.Ethnicity Depressive

symptomsSGA (%) Crude model

OR (95% CI)Model 1a

OR (95% CI)Model 2b

OR (95% CI)Dutch (n = 4062) Low 9.0 1.00 1.00 1.00

High 11.6 1.33 (1.05 – 1.68) 1.23 (0.96 – 1.68) 1.14 (0.89 – 1.46)

Creole (n = 580) Low 17.6 1.00 1.00 1.00

High 25.0 1.56 (1.04 – 2.33) 1.47 (0.97– 2.24) 1.44 (0.94 – 2.20)

Turkish (n = 384) Low 11.0 1.00 1.00 1.00

High 12.8 1.19 (0.64 – 2.22) 1.29 (0.68 – 2.48) 1.21 (0.63 – 2.33)

Moroccan (n = 657) Low 12.8 1.00 1.00 1.00

High 15.3 1.23 (0.79 – 1.91) 1.28 (0.81 – 2.02) 1.26 (0.80 – 1.99)

aAdjusted for maternal age (linear & quadratic term), education, pre-pregnancy BMI (linear & quadratic term), hypertension, alcohol and drug use.bAdditionally adjusted for maternal smoking during pregnancy.

Table 5.3b Association between maternal depressive symptoms and PTB, separated by ethnicity.Ethnicity Depressive

symptomsPTB (%)

Crude model OR (95% CI)

Model 1a

OR (95% CI)Model 2b

OR (95% CI)Dutch (n = 4072) Low 5.0 1.00 1.00 1.00

High 5.2 1.04 (0.74 – 1.46) 1.01 (0.72 – 1.43) 0.95 (0.67 – 1.35)

Creole (n = 583) Low 8.2 1.00 1.00 1.00

High 12.3 1.57 (0.91 – 2.70) 1.44 (0.81– 2.54) 1.45 (0.81 – 2.57)

Turkish (n = 384) Low 4.0 1.00 1.00 1.00

High 5.7 1.43 (0.55 – 3.72) 1.38 (0.53 – 3.63) 1.38 (0.52 – 3.64)

Moroccan (n = 658) Low 4.2 1.00 1.00 1.00

High 3.6 0.86 (0.38 – 1.93) 0.78 (0.34 – 1.83) 0.81 (0.35 – 1.90)

aAdjusted for maternal age, parity, education, pre-pregnancy BMI, hypertension, alcohol and drug use.bAdditionally adjusted for maternal smoking during pregnancy.

Table 5.3c Association between maternal depressive symptoms and a low Apgar score, separated by ethnicity.Ethnicity Depressive

symptomsLow Apgar score (%)

Crude model OR (95% CI)

Model 1a

OR (95% CI)Model 2b

OR (95% CI)Dutch (n = 3611) Low 1.2 1.00 1.00 1.00

High 2.5 2.21 (1.26 – 3.87) 2.04 (1.14 – 3.63) 2.01 (1.12 – 3.60)

Creole (n = 476) Low 1.5 1.00 1.00 1.00

High 3.4 2.38 (0.69 – 8.25) 2.73 (0.76– 9.89) 2.72 (0.75 – 9.88)

Turkish (n = 289) Low 0.8 1.00 1.00 1.00

High 0.6 0.79 (0.05 – 12.82) 0.76 (0.04 – 13.94) 0.69 (0.03 – 14.39)

Moroccan (n = 458) Low 1.2 1.00 1.00 1.00

High 1.4 1.22 (0.24 – 6.09) 1.57 (0.30 – 8.28) 1.65 (0.31 – 8.78)

aAdjusted for maternal age, parity, education, pre-pregnancy BMI, hypertension, alcohol and drug use.bAdditionally adjusted for maternal smoking during pregnancy.

72

Chapter 5 _________________________________________________________________________

Table 5.3d Association between maternal depressive symptoms and child loss, separated by ethnicity.Ethnicity Depressive

symptomsChild

loss (%)Crude model OR (95% CI)

Model 1a

OR (95% CI)Model 2b

OR (95% CI)Dutch (n = 4236) Low 1.0 1.00 1.00 1.00

High 1.7 1.77 (0.97 – 3.24) 1.68 (0.91 – 3.12) 1.61 (0.87 – 3.00)

Creole (n = 615) Low 2.3 1.00 1.00 1.00

High 5.0 2.25 (0.92 – 5.51) 2.22 (0.89– 5.53) 2.14 (0.85 – 5.36)

Turkish (n = 393) Low 1.1 1.00 1.00 1.00

High 1.4 1.19 (0.20 – 7.23) 1.19 (0.19 – 7.44) 1.01 (0.15 – 6.61)

Moroccan (n = 682) Low 1.0 1.00 1.00 1.00

High 0.7 0.69 (0.13 – 3.82) 0.72 (0.13 – 4.04) 0.73 (0.13 – 4.08)

aAdjusted for maternal age, parity, education, pre-pregnancy BMI, hypertension, alcohol and drug use.bAdditionally adjusted for maternal smoking during pregnancy.

Discussion

The purpose of the present study was to explore whether maternal depressive symptoms during pregnancy are associated with major perinatal health outcomes. In general, higher rates of PTB, SGA, a low Apgar score and child loss were observed among women with high levels of depressive symptoms during pregnancy. After adjustment for covariates and a small mediation effect of maternal smoking, offspring of mothers with high depressive symptoms had a statistically significant higher risk for SGA and a low Apgar score. Although we did not find statistically significant interaction effects of depressive symptoms with ethnic background, a tendency for a higher risk was observed among women with a Creole background.

In contrast with previous smaller studies,21, 26-28 we observed a significant association between maternal depressive symptoms and a low Apgar score at 5 minutes after birth. The significant association we found with SGA was consistent with some previous studies,19, 20 but inconsistent with others.21, 27, 28 While several studies observed a higher PTB risk among depressed women,18-24 we could not prove this in our sample. This might be explained by the various definitions used for PTB. Although all studies defined PTB as a birth before 37.0 weeks of gestation, only one study23 restricted the PTB group by defining a minimum weeks of gestation. We excluded births before 24.0 weeks of gestation because they are mostly not viable. Including these births would have resulted in increased and significant odds ratios for the association between maternal depressive symptoms and PTB (OR 1.25, p = .03 after full adjustment), supporting previous studies. We, however, feel that including all births before 37.0 weeks of gestation results in overestimation of the PTB risk, partly reflecting a potential association between maternal depression and miscarriages/fetal death. The few studies that explored the association of maternal depression with child loss are hard to compare, because of different definitions of child loss. We did not observe a significant association with child loss, including both miscarriages and perinatal deaths, which was consistent with a previous study exploring stillbirths over the whole range of pregnancy.21 Several small studies exploring the

73

_______________________________________________ Maternal depression and perinatal outcomes

5

association with miscarriages, however, presented significant30, 49 as well as insignificant31 results. The number of child losses in our sample were too small to perform separate analyses for subtypes of child loss; moreover, the data on miscarriages were most likely incomplete.

Stratified analysis by ethnic background suggested that the association of high vs. low levels of maternal depressive symptoms with perinatal health outcomes was stronger among the Creole women compared to the other ethnic groups. A similar difference between African-American and white women was reported before by Orr and Miller.25 These differences are not likely to be explained by higher risk behaviors such as maternal smoking or a lack of prenatal health care visits, because these risk behaviors are also seen among the Turkish and Moroccan groups.33, 50, 51 A more likely explanation is the higher vascular activity and reactivity to stress among blacks.36, 37 Previous research exploring ethnic differences in vascular characteristics of pregnant women observed greater resistance artery contractility, higher vascular creatine kinase activity and lower nitric oxide bioavailability among blacks compared to whites, which may contribute to higher blood pressure levels and hypertensive disorders.52 In case of depression or stress, this may lead to exaggerated cardiovascular reactivity among blacks, implying a higher risk for adverse pregnancy outcomes.35

The comparison of studies on the detrimental effects of maternal depression during pregnancy is hampered by the use of various kinds of measurement tools to assess either maternal depressive disorders or depressive symptomatology. While most studies used a self-report questionnaire to assess the level of depressive symptoms,20, 22, 24, 28 some used a diagnostic interview to assess mental disorders based on DSM-IV criteria.21, 27 Observed associations seemed, however, independent of the measurement tool. The CES-D scale, which was used in this study to measure depressive symptomatology, was found to have good validity, internal consistency and factor structure in the general population39, 41 as well as in various subpopulations, including pregnant women24, 43 and different ethnic groups.24, 39, 44, 45 Nonetheless, it is questionable whether the commonly used cut-off point of a CES-D score of ≥16 to define high-risk groups can similarly be applied among pregnant women and certain ethnic groups, which tend to score higher on the somatic component of the CES-D scale.24, 41, 44, 45, 53, 54 To address this problem, several approaches can be followed, including the omission of the somatic subscale of the CES-D54 or by raising the cut-off point to the upper 10 percent.24 We explored both approaches by performing additional analyses; while we hypothesized that the association of maternal depressive symptoms with perinatal health outcomes would be stronger, the opposite appeared to be true (results not shown). This was also observed in a previous study by Kabir et al.54 which compared the psychometric properties and predictive capacity for perinatal outcomes of the full CES-D scale with the CES-D scale without the somatic component. Apparently, the full CES-D scale and customary cut-off point of 16 can appropriately be applied among pregnant women; whether this is also true for specific ethnic subgroups remains, however, unclear and needs to be further explored. In our sample, 30.6% of the women had a CES-D score of 16 or higher, meaning that they had elevated levels of depressive symptoms and that they were at high risk for clinical depression.24 This prevalence was even higher among the ethnic minority groups (42-55%). In the 2008 general health survey among the Amsterdam population, 46.6% of the women reported mild depressive symptoms and 8% severe depressive symptoms [measured

74

Chapter 5 _________________________________________________________________________

with the Kessler Psychological Distress scale (K10)].55 In the 2004 general health survey in Amsterdam, structured interviews were performed using the Composite International Diagnostic Interview (CIDI); the Turkish (14.9%) and Moroccan (6.6%) group were found to have a higher 1-month prevalence of depressive disorders compared to the Dutch (4.4%) and Surinamese (1.1%) group.56 These depression prevalences are, however, hard to compare as they are assessed using different psychometric measurement tools.

Several pathways may explain the association between maternal depressive symptomatology during pregnancy and offspring risk for SGA and a low Apgar score. First, maternal depression may have an indirect effect on the fetus through its promoting influence on other risk factors for adverse perinatal health outcomes. Examples of risk factors associated with both maternal depression and offspring health are poor nutrition,32, 57 hypertension,38, 58 use of cigarettes, alcohol and drugs,59, 60 and lack of prenatal health care visits.50 Most studies in this research field included smoking as a covariate in the statistical models though suggested that it would be more likely for smoking to be on the causal pathway, i.e. being a mediator.21, 23, 24, 28 Many people, including pregnant women, smoke to relieve depression, unhappiness or stress,12, 61 and maternal smoking, in turn, is an important risk factor for fetal growth and development.13, 16 We have shown through mediation analysis that smoking was a partial mediator in the association between maternal depressive symptoms and, in particular, PTB and SGA. As a result, reducing the prevalence of depression during pregnancy will not only directly improve perinatal health outcomes, but also indirectly through a reduction in the smoking prevalence. The use of antidepressants might also mediate an association between maternal depression and offspring outcome.19, 62, 63 In our sample, only 0,8% of the women reported the use of antidepressants during pregnancy. Antidepressant use was more often, but not exclusively, reported by women with high depressive symptomatology and by Dutch women. We did, however, not observe significant associations between antidepressant use and perinatal health outcomes (results not shown). Second, a direct pathway through which maternal depressive symptoms affect fetal development, growth and parturition involves the activity of the HPA-axis. Depression is found to be related to increased HPA activity which results, among others, in enhanced levels of the hormone cortisol.17 Maternal cortisol has the ability to cross the placenta. Normally, the placental barrier [including 11β-hydroxysteroid dehydrogenase type 2 (11β-HSD2)] limits fetal exposure to maternal cortisol.64 However, in the case of elevated maternal cortisol levels and/or placental 11β-HSD2 deficiency, increasing levels of maternal cortisol cross the placenta resulting in elevated fetal cortisol levels.9 This mechanism is supported by the observation that maternal cortisol levels are highly correlated to fetal cortisol levels.65 It is hypothesized that fetal exposure to excess cortisol levels may impair fetal development and growth and may trigger early parturition through several mechanisms, including dysregulation of the fetal autonomic nervous system activity,66 increased uterine artery vasoconstriction resulting in reduced uterine blood flow67 or increased vulnerability to infectious agents.68 Major strengths of the present study include the large sample size, the community-based sample and the prospective study design. Selective response was present in the ABCD study, with

75

_______________________________________________ Maternal depression and perinatal outcomes

5

a higher participation rate among Western women, but selection bias could not be proven.69 Even though the ethnic groups had reasonable sizes, ethnic minority groups were too small to prove a potential interaction effect with maternal depression on perinatal outcomes. Maternal depressive symptoms were measured during the first trimester of pregnancy, which is suggested to be the most critical window to affect fetal development.70 Nonetheless, a different level of depressive symptoms at the end of pregnancy might still increase or decrease the odds for adverse perinatal outcomes.17, 70, 71 We therefore recommend for future research to measure depressive symptoms at several time points during pregnancy.

In conclusion, in – to our knowledge – the largest study exploring the detrimental effects of maternal depressive symptomatology during pregnancy on perinatal mortality and morbidity, we observed a higher risk to be born SGA or with a low Apgar score for offspring of mothers who reported high levels of depressive symptoms during pregnancy. Although the effect sizes are moderate, the extremely high prevalence of high depressive symptoms during pregnancy in our sample makes it a major public health problem. Women should be screened for depressive symptomatology during early pregnancy or preferably, during the preconceptional period. They should subsequently be urged to follow a comprehensive intervention program which addresses both psychosocial well-being and risk behaviors like smoking. Further research should explore whether certain groups (e.g. black women) are more sensitive to the detrimental effects of high maternal depressive symptoms during pregnancy. Moreover, future research should explore the underlying mechanisms to explain the observed associations.

76

Chapter 5 _________________________________________________________________________

References

1. EURO-PERISTAT project with SCPE, EUROCAT, EURONEOSTAT. European perinatal health report. Better statistics for better health in pregnant women and their babies. Paris: EURO-PERISTAT; 2008. Retrieved from http://www.europeristat.com.

2. Thorngren-Jerneck K, Herbst A. Low 5-minute Apgar score: a population-based register study of 1 million term births. Obstet Gynecol 2001;98:65-70.

3. Romo A, Carceller R, Tobajas J. Intrauterine growth retardation (IUGR): epidemiology and etiology. Pediatr Endocrinol Rev 2009;6(Suppl 3):332-6.

4. Ramachandrappa A, Jain L. Health issues of the late preterm infant. Pediatr Clin North Am 2009;56:565-77.

5. McFarlin BL. Solving the puzzle of prematurity. Am J Nurs 2009;109:60-3.6. Barker DJ. Adult consequences of fetal growth restriction. Clin Obstet Gynecol 2006;49:270-83.7. Puga B, Puga PG, de Arriba A, Armendariz Y, Labarta JI, Longas AF. Psychomotor and intellectual

development (Neurocognitive Function) of children born small for gestational age (SGA). Transversal and longitudinal study. Pediatr Endocrinol Rev 2009;6(Suppl 3):358-70.

8. Ehrenstein V, Pedersen L, Grijota M, Nielsen GL, Rothman KJ, Sorensen HT. Association of Apgar score at five minutes with long-term neurologic disability and cognitive function in a prevalence study of Danish conscripts. BMC Pregnancy Childbirth 2009;9:14.

9. Seckl JR, Holmes MC. Mechanisms of disease: glucocorticoids, their placental metabolism and fetal ‘programming’ of adult pathophysiology. Nat Clin Pract Endocrinol Metab 2007;3:479-88.

10. Chida Y, Steptoe A. Cortisol awakening response and psychosocial factors: a systematic review and meta-analysis. Biol Psychol 2009;80:265-78.

11. Wadhwa PD, Dunkel-Schetter C, Chicz-DeMet A, Porto M, Sandman CA. Prenatal psychosocial factors and the neuroendocrine axis in human pregnancy. Psychosom Med 1996;58:432-46.

12. Borrelli B, Bock B, King T, Pinto B, Marcus BH. The impact of depression on smoking cessation in women. Am J Prev Med 1996;12:378-87.

13. Jaddoe VW, Troe EJ, Hofman A, Mackenbach JP, Moll HA, Steegers EA, et al. Active and passive maternal smoking during pregnancy and the risks of low birthweight and preterm birth: the Generation R Study. Paediatr Perinat Epidemiol 2008;22:162-71.

14. Salihu HM, Aliyu MH, Pierre-Louis BJ, Alexander GR. Levels of excess infant deaths attributable to maternal smoking during pregnancy in the United States. Matern Child Health J 2003;7:219-27.

15. Shah NR, Bracken MB. A systematic review and meta-analysis of prospective studies on the association between maternal cigarette smoking and preterm delivery. Am J Obstet Gynecol 2000;182:465-72.

16. Roza SJ, Verburg BO, Jaddoe VW, Hofman A, Mackenbach JP, Steegers EA, et al. Effects of maternal smoking in pregnancy on prenatal brain development. The Generation R Study. Eur J Neurosci 2007;25:611-7.

17. Alder J, Fink N, Bitzer J, Hosli I, Holzgreve W. Depression and anxiety during pregnancy: a risk factor for obstetric, fetal and neonatal outcome? A critical review of the literature. J Matern Fetal Neonatal Med 2007;20:189-209.

18. Field T, Diego M, Dieter J, Hernandez-Reif M, Schanberg S, Kuhn C, et al. Prenatal depression effects on the fetus and the newborn. Infant Behav Dev 2004;27:216-29.

19. Oberlander TF, Warburton W, Misri S, Aghajanian J, Hertzman C. Neonatal outcomes after prenatal exposure to selective serotonin reuptake inhibitor antidepressants and maternal depression using population-based linked health data. Arch Gen Psychiatry 2006;63:898-906.

20. Steer RA, Scholl TO, Hediger ML, Fischer RL. Self-reported depression and negative pregnancy outcomes. J Clin Epidemiol 1992;45:1093-9.

21. Ferri CP, Mitsuhiro SS, Barros MC, Chalem E, Guinsburg R, Patel V, et al. The impact of maternal experience of violence and common mental disorders on neonatal outcomes: a survey of adolescent mothers in Sao Paulo, Brazil. BMC Public Health 2007;7:209.

77

_______________________________________________ Maternal depression and perinatal outcomes

5

22. Kramer MS, Lydon J, Seguin L, Goulet L, Kahn SR, McNamara H, et al. Stress pathways to spontaneous preterm birth: the role of stressors, psychological distress, and stress hormones. Am J Epidemiol 2009;169:1319-26.

23. Jesse DE, Seaver W, Wallace DC. Maternal psychosocial risks predict preterm birth in a group of women from Appalachia. Midwifery 2003;19:191-202.

24. Orr ST, James SA, Blackmore PC. Maternal prenatal depressive symptoms and spontaneous preterm births among African-American women in Baltimore, Maryland. Am J Epidemiol 2002;156:797-802.

25. Orr ST, Miller CA. Maternal depressive symptoms and the risk of poor pregnancy outcome. Review of the literature and preliminary findings. Epidemiol Rev 1995;17:165-71.

26. Lundy BL, Jones NA, Field T, Nearing G, Davalos M, Pietro PA, et al. Prenatal depression effects on neonates. Infant Behav Dev 1999;22:119-29.

27. Andersson L, Sundstrom-Poromaa I, Wulff M, Astrom M, Bixo M. Neonatal outcome following maternal antenatal depression and anxiety: a population-based study. Am J Epidemiol 2004;159:872-81.

28. Berle JO, Mykletun A, Daltveit AK, Rasmussen S, Holsten F, Dahl AA. Neonatal outcomes in offspring of women with anxiety and depression during pregnancy. A linkage study from The Nord-Trondelag Health Study (HUNT) and Medical Birth Registry of Norway. Arch Womens Ment Health 2005;8:181-9.

29. Sykes GS, Molloy PM, Johnson P, Gu W, Ashworth F, Stirrat GM, et al. Do Apgar scores indicate asphyxia? Lancet 1982;1:494-6.

30. Nakano Y, Oshima M, Sugiura-Ogasawara M, Aoki K, Kitamura T, Furukawa TA. Psychosocial predictors of successful delivery after unexplained recurrent spontaneous abortions: a cohort study. Acta Psychiatr Scand 2004;109:440-6.

31. Nelson DB, McMahon K, Joffe M, Brensinger C. The effect of depressive symptoms and optimism on the risk of spontaneous abortion among innercity women. J Womens Health (Larchmt) 2003;12:569-76.

32. Leung BM, Kaplan BJ. Perinatal depression: prevalence, risks, and the nutrition link - A review of the literature. J Am Diet Assoc 2009;109:1566-75.

33. Goedhart G, van Eijsden M, van der Wal MF, Bonsel GJ. Ethnic differences in term birthweight: the role of constitutional and environmental factors. Paediatr Perinat Epidemiol 2008;22:360-8.

34. Goedhart G, van Eijsden M, van der Wal MF, Bonsel GJ. Ethnic differences in preterm birth and its subtypes: the effect of a cumulative risk profile. BJOG 2008;115:710-9.

35. Giscombe CL, Lobel M. Explaining disproportionately high rates of adverse birth outcomes among African Americans: the impact of stress, racism, and related factors in pregnancy. Psychol Bull 2005;131:662-83.

36. Barnes VA, Treiber FA, Musante L, Turner JR, Davis H, Strong WB. Ethnicity and socioeconomic status: impact on cardiovascular activity at rest and during stress in youth with a family history of hypertension. Ethn Dis 2000;10:4-16.

37. Dysart JM, Treiber FA, Pflieger K, Davis H, Strong WB. Ethnic differences in the myocardial and vascular reactivity to stress in normotensive girls. Am J Hypertens 1994;7:15-22.

38. Hogue CJ, Bremner JD. Stress model for research into preterm delivery among black women. Am J Obstet Gynecol 2005;192(5 Suppl):S47-55.

39. Radloff LS. The CES-D scale: a self-reported depression scale for research in the general population. Appl Psychol Meas 1977;1:385-401.

40. Weissman MM, Sholomskas D, Pottenger M, Prusoff BA, Locke BZ. Assessing depressive symptoms in five psychiatric populations: a validation study. Am J Epidemiol 1977;106:203-14.

41. Shafer AB. Meta-analysis of the factor structures of four depression questionnaires: Beck, CES-D, Hamilton, and Zung. J Clin Psychol 2006;62:123-46.

42. Bouma J, Ranchor AV, Sanderman R, van Sonderen E. The measurement of depressive symptoms with the CES-D: a manual. [In Dutch]. Groningen: Noordelijk Centrum voor Gezondheidsvraagstukken, Rijksuniversiteit Groningen; 1995.

78

Chapter 5 _________________________________________________________________________

43. Field T, Hernandez-Reif M, Diego M, Schanberg S, Kuhn C. Stability of mood states and biochemistry across pregnancy. Infant Behav Dev 2006;29:262-7.

44. Roberts RE. Reliability of the CES-D Scale in different ethnic contexts. Psychiatry Res 1980;2:125-34.45. Spijker J, van der Wurff FB, Poort EC, Smits CH, Verhoeff AP, Beekman AT. Depression in first generation

labour migrants in Western Europe: the utility of the Center for Epidemiologic Studies Depression Scale (CES-D). Int J Geriatr Psychiatry 2004;19:538-44.

46. Casey BM, McIntire DD, Leveno KJ. The continuing value of the Apgar score for the assessment of newborn infants. N Engl J Med 2001;344:467-71.

47. Drage JS, Kennedy C, Schwarz BK. The Apgar score as an index of neonatal mortality. A report from the collaborative study of cerebral palsy. Obstet Gynecol 1964;24:222-30.

48. Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol 1986;51:1173-82.

49. Sugiura-Ogasawara M, Furukawa TA, Nakano Y, Hori S, Aoki K, Kitamura T. Depression as a potential causal factor in subsequent miscarriage in recurrent spontaneous aborters. Hum Reprod 2002;17:2580-4.

50. Alderliesten ME, Vrijkotte TG, van der Wal MF, Bonsel GJ. Late start of antenatal care among ethnic minorities in a large cohort of pregnant women. BJOG 2007;114:1232-9.

51. Troe EJ, Raat H, Jaddoe VW, Hofman A, Steegers EA, Verhulst FC, et al. Smoking during pregnancy in ethnic populations: the Generation R study. Nicotine Tob Res 2008;10:1373-84.

52. Brewster LM, Taherzadeh Z, Volger S, Clark JF, Rolf T, Wolf H, et al. Ethnic differences in resistance artery contractility of normotensive pregnant women. Am J Physiol Heart Circ Physiol 2010. doi:10.1152/ajpheart.00919.2009

53. Simon GE, Goldberg DP, Von Korff M, Ustün TB. Understanding cross-national differences in depression prevalence. Psychol Med 2002;32:585-94.

54. Kabir K, Sheeder J, Stevens-Simon C. Depression, weight gain, and low birth weight adolescent delivery: do somatic symptoms strengthen or weaken the relationship? J Pediatr Adolesc Gynecol 2008;21:335-42.

55. GGD Amsterdam. This healthy is Amsterdam! Final report Amsterdam Health Monitor 2008. [In Dutch]. Amsterdam: GGD Amsterdam; 2009.

56. De Wit MA, Tuinebreijer WC, Dekker J, Beekman AJ, Gorissen WH, Schrier AC, et al. Depressive and anxiety disorders in different ethnic groups: a population based study among native Dutch, and Turkish, Moroccan and Surinamese migrants in Amsterdam. Soc Psychiatry Psychiatr Epidemiol 2008;43:905-12.

57. Ramakrishnan U, Manjrekar R, Rivera J, Gonzales-Cossio T, Martorell R. Micronutrients and pregnancy outcome: a review of the literature. Nutr Res 1999;19:103-59.

58. Lau TK, Pang MW, Sahota DS, Leung TN. Impact of hypertensive disorders of pregnancy at term on infant birth weight. Acta Obstet Gynecol Scand 2005;84:875-7.

59. Dew PC, Guillory VJ, Okah FA, Cai J, Hoff GL. The effect of health compromising behaviors on preterm births. Matern Child Health J 2007;11:227-33.

60. Hanna EZ, Faden VB, Dufour MC. The motivational correlates of drinking, smoking, and illicit drug use during pregnancy. J Subst Abuse 1994;6:155-67.

61. Goedhart G, van der Wal MF, Cuijpers P, Bonsel GJ. Psychosocial problems and continued smoking during pregnancy. Addict Behav 2009;34:403-6.

62. Yonkers KA, Wisner KL, Stewart DE, Oberlander TF, Dell DL, Stotland N, et al. The management of depression during pregnancy: a report from the American Psychiatric Association and the American College of Obstetricians and Gynecologists. Obstet Gynecol 2009;114:703-13.

63. Suri R, Altshuler L, Hellemann G, Burt VK, Aquino A, Mintz J. Effects of antenatal depression and antidepressant treatment on gestational age at birth and risk of preterm birth. Am J Psychiatry 2007;164:1206-13.

64. Mark PJ, Waddell BJ. P-glycoprotein restricts access of cortisol and dexamethasone to the glucocorticoid receptor in placental BeWo cells. Endocrinology 2006;147:5147-52.

79

_______________________________________________ Maternal depression and perinatal outcomes

5

65. Glover V, Teixeira J, Gitau R, Fisk NM. Mechanisms by which maternal mood in pregnancy may affect the fetus. Contemp Rev Obstet Gynecol 1999;155-60.

66. Diego MA, Field T, Hernandez-Reif M, Schanberg S, Kuhn C, Gonzalez-Quintero VH. Prenatal depression restricts fetal growth. Early Hum Dev 2009;85(1):65-70.

67. Field T, Diego M. Cortisol: the culprit prenatal stress variable. Int J Neurosci 2008;118:1181-205.68. Wadhwa PD, Culhane JF, Rauh V, Barve SS, Hogan V, Sandman CA, et al. Stress, infection and preterm

birth: a biobehavioural perspective. Paediatr Perinat Epidemiol 2001;15(Suppl 2):17-29.69. Tromp M, van Eijsden M, Ravelli ACJ, Bonsel GJ. Anonymous non-response analysis in the ABCD

cohort study enabled by probabilistic record linkage. Paediatr Perinat Epidemiol 2009;23:264-72.70. Sandman CA, Glynn L, Schetter CD, Wadhwa P, Garite T, Chicz-DeMet A, et al. Elevated maternal

cortisol early in pregnancy predicts third trimester levels of placental corticotropin releasing hormone (CRH): priming the placental clock. Peptides 2006;27:1457-63.

71. Roesch SC, Dunkel-Schetter C, Woo G, Hobel CJ. Modeling the types and timing of stress in pregnancy. Anxiety, Stress, and Coping 2004;17:87-102.