Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Maternal Prenatal Positive Affect, Depressive and Anxiety Symptoms and Birth Outcomes: The PREDO Study

  • Anu-Katriina Pesonen ,

    Contributed equally to this work with: Anu-Katriina Pesonen, Marius Lahti (AKP); (KR)

    Affiliation Institute of Behavioral Sciences, University of Helsinki, Helsinki, Finland

  • Marius Lahti ,

    Contributed equally to this work with: Anu-Katriina Pesonen, Marius Lahti

    Affiliation Institute of Behavioral Sciences, University of Helsinki, Helsinki, Finland

  • Tiina Kuusinen,

    Affiliation Institute of Behavioral Sciences, University of Helsinki, Helsinki, Finland

  • Soile Tuovinen,

    Affiliation Institute of Behavioral Sciences, University of Helsinki, Helsinki, Finland

  • Pia Villa,

    Affiliation Department of Obstetrics and Gynecology, University of Helsinki, Helsinki, Finland

  • Esa Hämäläinen,

    Affiliations Department of Clinical Chemistry, University of Helsinki, Helsinki, Finland, HUSLAB, Helsinki University Central Hospital, Helsinki, Finland

  • Hannele Laivuori,

    Affiliations Medical Genetics, University of Helsinki and Helsinki University Hospital, Helsinki, Finland, Obstetrics and Gynecology, University of Helsinki and Helsinki University Hospital, Helsinki, Finland, Institute for Molecular Medicine Finland, University of Helsinki, Helsinki, Finland

  • Eero Kajantie,

    Affiliations National Institute for Health and Welfare, Helsinki, Finland, Children's Hospital, Helsinki University Central Hospital and University of Helsinki, Helsinki, Finland, Department of Obstetrics and Gynecology, Oulu University Hospital and University of Oulu, Oulu, Finland

  • Katri Räikkönen (AKP); (KR)

    Affiliation Institute of Behavioral Sciences, University of Helsinki, Helsinki, Finland



We investigated whether maternal prenatal emotions are associated with gestational length and birth weight in the large PREDO Study with multiple measurement points of emotions during gestation.


Altogether 3376 pregnant women self-assessed their positive affect (PA, Positive and Negative Affect Schedule) and depressive (Center for Epidemiologic Studies Depression Scale, CES-D) and anxiety (Spielberger State Anxiety Scale, STAI) symptoms up to 14 times during gestation. Birth characteristics were derived from the National Birth Register and from medical records.


One standard deviation (SD) unit higher PA during the third pregnancy trimester was associated with a 0.05 SD unit longer gestational length, whereas one SD unit higher CES-D and STAI scores during the third trimester were associated with 0.04–0.05 SD unit shorter gestational lengths (P-values ≤ 0.02), corresponding to only 0.1–0.2% of the variation in gestational length. Higher PA during the third trimester was associated with a significantly decreased risk for preterm (< 37 weeks) delivery (for each SD unit higher positive affect, odds ratio was 0.8-fold (P = 0.02). Mothers with preterm delivery showed a decline in PA and an increase in CES-D and STAI during eight weeks prior to delivery. Post-term birth (≥ 42 weeks), birth weight and fetal growth were not associated with maternal prenatal emotions.


This study with 14 measurements of maternal emotions during pregnancy show modest effects of prenatal emotions during the third pregnancy trimester, particularly in the weeks close to delivery, on gestational length. From the clinical perspective, the effects were negligible. No associations were detected between prenatal emotions and birth weight.


Mounting empirical evidence suggests that maternal emotional distress during pregnancy, including depressive symptoms, and state, trait and pregnancy-specific anxiety, may increase the risk of adverse birth outcomes, such as preterm birth (< 37 0/7 gestational weeks) or shorter length of gestation, intrauterine growth restriction (IUGR), and low (< 2.5 kg) or lower birth weight at delivery [17]. However, the literature is relatively inconsistent, and includes null associations as well [8]. The associations may also vary according to continuous or categorical treatment of the exposure and/or outcome variables, and the level of income in the country in which the sample was derived [1,4,5,7].

A small body of literature suggests that maternal positive predisposition or positive attitude towards pregnancy may, independently of emotional distress, decrease the risk of adverse birth outcomes. In a sample of 130 pregnant women, higher levels of self-esteem and optimism measured in gestational weeks 28–30 were associated with higher birth weight, but not with length of gestation [9]. Correspondingly, a higher level of dispositional optimism measured in gestational weeks 21–30 was associated with higher birth weight, but not with length of gestation [10], among medically high-risk sample of 129 pregnant women. A more recent study in 169 pregnant women demonstrated that more positive attitudes towards pregnancy, measured on average at gestational week 15, were associated with a longer length of gestation and a lower risk of preterm delivery. This study additionally showed that a steeper increase in maternal positive attitudes towards gestational weeks 19 and 30 was associated with a longer gestational length, but it was not associated with prematurity [11]. While these studies provide important insight into the potentially beneficial effects of positive characteristics, all of the existing studies have been conducted in relatively small and/or selected samples and none have covered the whole pregnancy. This has precluded testing whether the effects would vary according to the timing of the exposure. In addition, the measurement of positive affect has mostly referred to attitudes towards pregnancy or dispositional trait-like measures of the pregnant mother, and less to state-like fluctuations in current affect.

Our original contribution in this study is to explore the influence of positive affect and the role of the timing of exposures for the birth outcomes. To our knowledge, our study offers the most comprehensive view on maternal emotional state over the course of pregnancy reported thus far, with prospective assessments of emotions at two-week intervals from 12 weeks’ gestation onwards in a large data set. Therefore, the roles of negative affect and its timing were also relevant. Hence, we examined the hypothesis that higher positive and lower negative affect would predict longer length of gestation and higher birth weight in 3376 pregnant women. We also examined whether the associations would vary by the medical risk-status of mothers, since obstetric complications are associated with higher emotional distress during pregnancy [12].



The participants came from the Prediction and Prevention of Preeclampsia (PREDO) Study [1315]. The PREDO is composed of two subsamples. For the high-risk subsample, we recruited 973 singleton pregnant women with risk factors for preeclampsia and 110 singleton pregnant women without known risk factors as a comparison group at 12 0/7–13 6/7 gestational weeks as previously described [13,15]. For the community-based subsample, we recruited 3702 singleton pregnant women within the same pregnancy weeks. The recruitment took place between September 2005 and February 2010 when these women attended their first ultrasound scan in one of ten hospital maternity clinics participating in the PREDO. Of the 4785 recruited mothers, we had data on birth outcomes from the birth register and the two-week interval self-report of positive affect, depressive symptoms, and state anxiety during pregnancy for 3376 mothers (70.6% of those recruited). Participants of the study did not differ from mothers with missing data on prenatal emotions in subsample status or birth outcome (P-values > 0.29), but were 0.9 year older (95% Confidence Interval (CI) 0.6 to 1.2, P < 0.001), more often primiparous (40.7% vs. 33.9%, P < 0.001) and less often obese before pregnancy (body mass index (BMI) > 30 kg/m2) (13.3% vs. 17.1%, P = 0.002). The participants also less often smoked during pregnancy (3.3% vs. 9.3%, P < 0.001) and more often had tertiary education (59.9% vs. 46.4%, P < 0.001). The characteristics of the participants are presented in Table 1. The study protocol was approved by the Ethical Committee of Helsinki University Central Hospital, and all participants signed an informed consent.

Table 1. Characteristics of the participants of the Predo-Study.

Measurement of variables

Prenatal emotions.

The participants filled in well-validated questionnaires [1618] on prenatal emotions at two week intervals throughout pregnancy from 120/7 to 13 6/7 gestational weeks until delivery or until 38 0/7 to 39 6/7 gestational weeks. The questionnaires were filled in up to 14 times during pregnancy.

Positive affect: We used Positive Affect (PA) scale comprising 10 mood states derived from the International Positive and Negative Affect Schedule (PANAS) [18]. The participants were asked to rate the extent to which they currently feel each mood state on a scale from 1 (not at all) to 5 (very much).

Depressive symptoms: We used the Center for Epidemiologic Studies Depression Scale (CES-D) (16) for both continuous and categorical variables: CES-D scores were dichotomized at ≥16, which is the cut-off for individuals at risk for clinically significant depressive symptoms [16]. The participants were asked to rate the 20 questions covering the frequency of symptoms experienced during the preceding week on a scale of 0 (not at all, less than one day) to 3 (all the time / 5–7 days).

Anxiety symptoms: To measure state anxiety, we used the Spielberger State Anxiety Scale (STAI) [17].The participants were asked to rate the 20 items covering the extent to which they currently feel anxious on a scale from 1 (not at all) to 4 (very much so).

Birth outcomes.

Data on gestational length and birth weight came from the hospital birth records for the high-risk subsample and from the national birth register for the community-based subsample. We used the birth outcome variables as both continuous and categorical variables: preterm birth: ≤ 36 6/7 gestational weeks; term birth: 37 0/7-41 6/7 gestational weeks; post-term birth: ≥ 420/7gestational weeks; small for gestational age (SGA): birth weight for gestational age ≤ -2 standard deviations (SD); appropriate for gestational age (AGA): birth weight for gestational age > -2 SD—≤ 2SD; LGA: birth weight for gestational age ≥ 2 SDs, all according to Finnish national growth charts [19].

Confounding variables.

Maternal age (years), delivery mode (vaginal vs. Cesarean), parity (primiparous vs. multiparous), smoking (no/quit during first trimester/smoked throughout pregnancy) during pregnancy, and infant’s sex (girl vs. boy) were derived from the national birth register for both subsamples. For the medical high-risk sample, maternal pre-pregnancy BMI (underweight: BMI < 18.5 kg/m2; normal weight: BMI = 18.5 to 25 kg/m2; overweight; BMI = 25 to 30 kg/m2; obese: BMI ≥ 30 kg/m2, hypertensive pregnancy disorders (preeclampsia, gestational hypertension or chronic hypertension), gestational diabetes (hyperglycemia that first emerged or was first identified during pregnancy), and Type I diabetes were derived from patient case records and verified by a clinical jury. For the community-based subsample the data of pregnancy disorders and BMI came from the national birth register. In both subsamples maternal education (primary: < 10 years; secondary: 10–12 years; tertiary: > 12 years) prenatal use of antidepressant or other psychotropic medication (sedatives, barbiturates and anti-psychotics) and alcohol consumption during pregnancy (yes/no) were self-reported in a questionnaire given at the first ultrasound screening at gestational week 12 6/7-13 6/7. Social support was assessed in at two-week intervals (i.e. biweekly) in the questionnaire with a 65 mm long visual analog scale.

Statistical analyses

First, we used linear regression analyses to study associations between maternal positive affect, depressive symptoms and state anxiety with birth outcomes as continuous variables. Trimester-specific prenatal emotions were tested in separate models. CES-D and STAI scores were square-root- and logarithm-transformed to attain normality. All continuous variables were standardized to the mean of 0 and SD of 1 to facilitate interpretation of effect sizes. Hence, the unstandardized regression coefficients represent SD unit change per SD unit change. We calculated trimester mean scores of prenatal emotions (first trimester week 12 measurement; second trimester means of scores during weeks 14–26; third trimester week 28 to delivery or to week 38). The rationale of using trimester means in these analyses instead of the bi-weekly scores was based on the very high inter-correlations between the bi-weekly emotion scores (r range 0.41–0.80, all P-values < 0.001).

All analyses were adjusted for maternal age at delivery, parity, delivery mode, education and infant’s sex (Model 1) and further for maternal alcohol consumption and smoking during pregnancy, maternal pre-pregnancy body mass index, hypertensive pregnancy disorders, gestational and Type I diabetes, antidepressant and other psychotropic medication, and social support (Model 2). In case of significant associations, the explained variance (R2) was calculated for prenatal emotions by hierarchical multiple regression analyses. We used logistic regressions to examine whether maternal positive affect decreased and depressive symptoms and state anxiety increased the risk of preterm vs. term, post-term vs. term, SGA vs. AGA, and LGA vs. AGA births. Finally, to test whether the associations varied by the medical risk status of the sample, we added an interaction term ‘high-risk subsample vs. community subsample x trimester-specific PA/CES-D/STAI’ to the regression equations following the main effects. In these interaction analyses, the small comparison group recruited for the high-risk subsample was merged to the community-based subsample.


PA (range of Pearson r’s 0.57 to 0.83), CES-D (r’s 0.61 to 0.83) and STAI (r’s 0.55 to 0.79) scores were significantly correlated across the three trimesters (all P-values < 0.001). PA correlated negatively within the trimesters with STAI (r’s -0.51 to -0.66) and CES-D (r’s -0.50 to -0.62). STAI and CES-D correlated positively with each other within the trimesters (r’s 0.69 to 0.84), all P-values < 0.001. In terms of mean raw scores shown in Table 1, CES-D and STAI scores were significantly higher during the first and third trimesters compared to the second trimester (all P-values < 0.001), and significantly higher during the third trimester than during the first trimester (P-values ≤ 0.002). No mean score differences were found for PANAS (all P-values ≥ 0.26).

Birth outcomes as continuous

Table 2 shows the trimester-specific associations between PA, CES-D and STAI and birth outcomes. One SD unit increase in PA and one SD unit decrease in CES-D and STAI during the third trimester were associated with 0.05 SD unit longer and 0.04 and 0.05 SD unit shorter gestational length, respectively. These associations remained significant after we made adjustments for the additional confounders in Model 2. PA, CES-D, and STAI at the first and second trimesters were not associated with gestational length (P-values ≥ 0.17), and PA, CES-D, and STAI at any trimester were not associated with birth weight (P-values ≥ 0.16). In hierarchical multiple regression analyses, independent background variables (maternal age at delivery, parity, delivery mode, education, infant’s sex, maternal alcohol consumption and smoking during pregnancy, maternal pre-pregnancy BMI, hypertensive pregnancy disorders, gestational and Type I diabetes, antidepressant and other psychotropic medication, and social support) explained 8.7% of the variance in gestational length, whereas PA, CES-D or STAI during the third trimester explained only 0.2%; 0.1% and 0.2% of the variance, respectively (P-values ≤ 0.02 for the change).

Table 2. The associations between prenatal positive affect, depressive and anxiety symptoms with birth outcome.

Birth outcomes as categorical

PA during the third trimester was associated with a significantly decreased odds for preterm delivery (for each SD unit increase in PA, the Odds Ratio (OR) decreased by 0.8-fold (95% CI 0.7; 1.0, P = 0.016 in Model 1; OR = 0.8, 95% CI 0.6; 1.0, P = 0.025 in Model 2) (these analyses excluded two infants born before the third trimester). CES-D (OR = 1.2, 95% CI 1.0–1.4, P = 0.066 in Model 1, P = 0.12 in Model 2), and STAI (OR 1.2, 95% CI 1.0–1.4, P = 0.091 in Model 1; P = 0.26 in Model 2) scores during the third trimester, in turn, were higher in women with a preterm delivery, but these associations were not statistically significant. PA, CES-D, and STAI at the first and second trimesters were not associated with a preterm delivery (P-values ≥ 0.44) and PA, CES-D, and STAI at any trimester were not associated with a post-term delivery (P-values ≥ 0.20) or with SGA vs. AGA, or LGA vs. AGA status of the infant (P-values ≥ 0.087).

Post-hoc exploratory analyses further highlighted the significance of the trajectories of emotions experienced during pregnancy. Fig 1 shows the trajectories of standardized values of prenatal emotions in three groups divided by the gestational length (preterm, term, post-term). In comparison with mothers with a term delivery, mothers with a preterm delivery showed a decrease in PA (P = 0.002 for the difference between mean prenatal emotion score during the last eight weeks of pregnancy and the mean score derived from earlier measurement points during pregnancy) and an increase in both CES-D (P = 0.038) and STAI (P = 0.037) scores towards the end of pregnancy.

Fig 1.

Trajectories of maternal (A) positive affect (PANAS), (B) depressive symptoms (CES-D), and (C) anxiety symptoms (STAI) at two-week intervals according to preterm (≤ 36 6/7 gestational weeks), term (37 0/7-41 6/7 gestational weeks) and post-term delivery (≥ 42 0/7 gestational weeks).

Moderation by medical risk status of the sample

Medical risk status moderated significantly the associations between the third-trimester CES-D and gestational length (P for high-risk vs community sample x CES-D interaction = 0.012 in Model 1, P = 0.032 in Model 2, respectively). In the high-risk sample, one SD unit increase in the third trimester CES-D was associated with a 0.12 SD unit shorter gestational length (95% CI -0.19 to -0.04, P = 0.003 in Model 1; P = 0.003 in Model 2); in the community sample this association was not significant (P-values > 0.28).


We investigated among 3376 women whether prenatal PA and depressive and anxiety symptoms were associated with the likelihood of shorter gestation, preterm birth or lower birth weight. While the issue has been widely studied, the previous literature concentrates mostly on negative emotions and does not cover the effect of the emotions experienced throughout the entire pregnancy. With our data, we were able to overcome these issues with up to 14 measurement points during pregnancy, allowing a trimester-specific approach, with an adjustment for several major confounders. We found statistically significant, but clinically negligible associations between positive and negative emotions during the third trimester and gestational length, and no significant associations with birth weight. In addition, PA during the third trimester was associated with lower odds for a preterm birth.

PA was associated with longer gestational length and negative emotions with shorter gestational length. The general pattern of the results was in accordance with earlier studies [27], which have not, however, provided trimester-specific information with multiple measurement points, as done here. We found high correlations between the bi-weekly measurements of emotions. Although this may suggest that measurement of emotions once in each trimester would be sufficient for future studies, we would also like to point out that with increasing measurement points, there is a decrease in measurement error, resulting in more reliable estimates of prenatal emotions in each trimester. The observed associations held when the models were adjusted for an extensive list of covariates. However, prenatal emotions independently explained only 0.1–0.2% of the variation in gestational length, which is an extremely modest amount. This effect corresponds to 12 hours’ shorter or longer gestation for each SD unit change in prenatal emotions.

With regard to prematurity, higher PA during the third trimester was associated with a reduced risk for a premature delivery, in line with a recent study [11]. However, they assessed positive emotions only three times during the pregnancy. We also found a non-significant trend for a higher risk of prematurity in those scoring higher on CES-D and STAI during the third trimester, but these associations clearly declined in the fully adjusted models. If the mean value of depressive symptoms was ≥ 16 points in CES-D in the third trimester, the risk for a premature birth was 1.5-fold, but this association was not significant in either model. However, the magnitude of this effect is in accordance with two meta-analyses on the relation between prenatal depression and increased risk for prematurity [4,5]. Similar odds ratios were also reported in the meta-analysis on the effects of prenatal anxiety on the risk of preterm birth [2]. Importantly, when we examined the prenatal affect during eight weeks prior to the delivery, we found a significant decrease in PA and an increase in depressive and anxiety symptoms among mothers with a preterm delivery, relative to mothers with term deliveries, again highlighting the modest links between gestational length and prenatal emotions, particularly in the later stages of gestation. In contrast, emotions during the first and second trimesters were unrelated to gestational length.

Kramer [6] presented a large variety of different measures of acute and chronic stressors and psychological distress to mothers in the second trimester and reported that of all forms of stress and psychological distress, only pregnancy-specific anxiety was associated with spontaneous preterm delivery, a finding that has been described before [10,20,21]. The authors speculated that this result may also indicate a reversed causality since pregnancy-specific anxiety is also a mother’s perception of potential medical risks. However, these concepts may interact; pregnancy-specific and general anxiety influence each other over time, resulting in accumulating anxiety for some mothers during pregnancy [22].

We did not observe any associations of PA, prenatal anxiety, or depression with birth weight. These findings are in line with a meta-analysis that noted no associations between depression and birth weight [4] and with a study [8] focusing only on mothers with diagnosed depressive/or anxiety disorders, and observing no associations with birth weight. However, our findings are in contrast to other meta-analytic studies showing that maternal depression [5] and anxiety [2] during pregnancy predict an increased risk of low birth weight, to the Generation R Study reporting an association between anxiety symptoms at the second trimester and lower birth weight [23], and to studies showing that positive dispositions may be associated with larger birth weight [9,10].

The current study subject has been under active research, with several biological mechanisms potentially underlying associations between prenatal emotions and gestational length. Increased concentration of corticotrophin-releasing hormone (CRH) in early pregnancy plasma predicts preterm birth [2426]. Most of the CRH during pregnancy is produced by the placenta; this production is stimulated by circulating cortisol [2729]. With CRH stimulating cortisol secretion by the placenta and the fetal adrenal cortex, this has been suggested to create a positive feedback loop ultimately leading to delivery [2729]. However, the evidence for the associations between maternal depression during pregnancy and elevated CRH concentrations or cortisol levels is inconclusive including both positive [3033] and null findings [6]. The placenta plays a central role in regulating transfer of glucocorticoids to the fetus: The placental enzyme 11-beta hydroxysteroid dehydrogenase type 2 (HSD2) metabolizes most of active maternal cortisol to inactive cortisone, protecting the fetus from cortisol overexposure [34]. In a subsample of term births in the current PREDO Study we quantified placental mRNA levels of glucocorticoid (GR) and mineralocorticoid (MR) serotonin receptor genes as well as levels of 11-beta HSD2 and 11-beta hydroxysteroid dehydrogenase type 1 (HSD1) [13,14]. We found that maternal antenatal depression was associated with increased placental mRNA expression of both MR and GR, thus increasing placental glucocorticoid sensitivity, and suggesting the existence of a mechanism for increased fetal glucocorticoid exposure following maternal antenatal distress [14]. The associations were strongest for the third trimester depression scores. We also reported an association between higher GR mRNA levels and shorter gestation within the term range [13]. These results suggest that maternal antenatal mood is associated with placental glucocorticoid function, by both regenerating active glucocorticoids in placenta and increasing sensitivity to glucocorticoids.

The strengths of this study include the large well-characterized sample and the multiple repeated measurements of PA and anxiety and depressive symptoms, enabling us to reliably evaluate the effects of these factors on gestation length and birth size and to assess specific sensitivity periods. We were also able to assess possible confounding by multiple factors known to affect gestation length and birth size.

As to limitations, we cannot rule out that the results, especially with regard to prematurity, reflect reverse causality, i.e. prenatal emotions would be then generated as a response to the risk of premature birth. Mothers in Finland are followed intensively during pregnancy, with mothers being well aware of potential risks. Second, PA had a strong inverse correlation with negative emotions. We cannot say whether PA had an independent effect on gestational length or whether the effect could be reduced by a lack of negative emotions (depressive and anxiety symptoms). This is, however, a semantic question related to the core definition of positive emotions and their relation to depression and anxiety. Third, we did not have information about whether mothers with depression and anxiety symptoms received any psychosocial support. Finally, the participants may not be representative of the catchment areas of the study hospitals; these mothers had on average high educational attainment and smoking during pregnancy was rare. This would be expected to cause bias only if the association between prenatal emotions and pregnancy outcomes differs between participants and non-participants.

In summary, we used a large sample of pregnant women who rated their prenatal emotions bi-weekly to examine the relationships of PA and depressive and anxiety symptoms with gestational length and birth weight. Experience of PA during the third trimester was associated with longer gestation and lower odds of preterm delivery, whereas higher prenatal depressive and anxiety symptoms during the third trimester were associated with shorter gestation. The associations regarding depressive symptoms and gestational length were stronger among high-risk mothers. We also found that the course of emotions during the last eight weeks of pregnancy is different among women delivering preterm, as they displayed more negative emotions and less PA. Birth weight and fetal growth were not associated with prenatal emotions, and no effects on gestational length were observed for the first two trimesters. Our findings thus highlight the modest effects of both positive and negative emotions during the third pregnancy trimester on gestational length. While this subject has raised intense research interest in recent years, our findings indicate that the effects of prenatal emotions on birth characteristics are very small.

The clinical relevance may be in the message that prenatal negative emotions, including symptoms of depression and anxiety, may not play such a large role for pregnancy outcome per se. Of course, this has to be verified in further studies with as intensive measurements of prenatal emotions.

However, several studies show that prenatal negative emotions may be harmful to child development, being associated with increased psychopathology in the offspring, often independently of the maternal postnatal emotions [3538]. As maternal prenatal emotions are likely to persist to the postnatal period [39], there is however a risk for an accumulative risk for a less optimal child development through less adaptive parent-child interaction patterns. New evidence is also emerging on the links between maternal prenatal stress and emotions and epigenetic programming of genes regulating the HPA axis in the offspring [40,41]. We have shown that higher placental expression of genes regulating feto-placental glucocorticoid and serotonin exposure is likely to mediate partly the associations between higher maternal depressive symptoms during the third trimester of pregnancy and regulatory behavioral challenges of the infant [42]. Together with this evidence, our present findings suggest that the burden of maternal prenatal negative emotions to the child may be transferred otherwise than through the immediate birth outcome.

Author Contributions

Conceived and designed the experiments: AKP PV HL EH EK KR. Performed the experiments: AKP ML ST TK PV HL EH EK KR. Analyzed the data: AKP TK ST ML KR. Wrote the paper: AKP ML ST TK PV HL EH EK KR.


  1. 1. Bussieres EL, Tarabulsy GM, Pearson J, Tessier R, Forest JC, Giguére Y. Maternal prenatal stress and infant birth weight and gestational age: a meta-analysis of prospective studies. Developmental Rev. 2015;36: 1179–1199.
  2. 2. Ding XX, Wu YL, Xu SJ, Zhu RP, Jia XM, Zhang SF, et al. Maternal anxiety during pregnancy and adverse birth outcomes: a systematic review and meta-analysis of prospective cohort studies. J Affect Disord. 2014;159: 103–110. pmid:24679397
  3. 3. Dole N, Savitz DA, Hertz-Picciotto I, Siega-Riz AM, McMahon MJ, Buekens P. Maternal stress and preterm birth. Am J Epidemiol. 2003;157: 14–24. pmid:12505886
  4. 4. Grigoriadis S, VonderPorten EH, Mamisashvili L, Tomlinson G, Dennis CL, Koren G, et al. The impact of maternal depression during pregnancy on perinatal outcomes: a systematic review and meta-analysis. J Clin Psychiatry. 2013;74: e321–341. pmid:23656857
  5. 5. Grote NK, Bridge JA, Gavin AR, Melville JL, Iyengar S, Katon WJ. A meta-analysis of depression during pregnancy and the risk of preterm birth, low birth weight, and intrauterine growth restriction. Arch Gen Psychiatry. 2010;67: 1012–1024. pmid:20921117
  6. 6. 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: 1319–1326. pmid:19363098
  7. 7. Raisanen S, Lehto SM, Nielsen HS, Gissler M, Kramer MR, Heinonen S. Risk factors for and perinatal outcomes of major depression during pregnancy: a population-based analysis during 2002–2010 in Finland. BMJ Open. 2014;4: e004883. pmid:25398675
  8. 8. 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: 872–881. pmid:15105180
  9. 9. Rini CK, Dunkel-Schetter C, Wadhwa PD, Sandman CA. Psychological adaptation and birth outcomes: the role of personal resources, stress, and sociocultural context in pregnancy. Health Psychol. 1999;18: 333–345. pmid:10431934
  10. 10. Lobel M, DeVincent CJ, Kaminer A, Meyer BA. The impact of prenatal maternal stress and optimistic disposition on birth outcomes in medically high-risk women. Health Psychol. 2000;19: 544–553. pmid:11129357
  11. 11. Voellmin A, Entringer S, Moog N, Wadhwa PD, Buss C. Maternal positive affect over the course of pregnancy is associated with the length of gestation and reduced risk of preterm delivery. J Psychosom Res. 2013;75: 336–340. pmid:24119939
  12. 12. Molyneaux E, Poston L, Ashurst-Williams S, Howard LM. Obesity and mental disorders during pregnancy and postpartum: a systematic review and meta-analysis. Obstet Gynecol. 2014;123: 857–867. pmid:24785615
  13. 13. Raikkonen K, O'Reilly JR, Pesonen AK, Kajantie E, Villa P, Laivuori H, et al. Associations between maternal level of education and occupational status with placental glucocorticoid regeneration and sensitivity. Clin Endocrinol (Oxf). 2014;81: 175–182.
  14. 14. Reynolds RM, Pesonen AK, O'Reilly JR, Tuovinen S, Lahti M, Kajantie E, et al. Maternal depressive symptoms throughout pregnancy are associated with increased placental glucocorticoid sensitivity. Psychol Med. 2015: 1–8.
  15. 15. Villa PM, Kajantie E, Raikkonen K, Pesonen AK, Hamalainen E, Vainio M, et al. Aspirin in the prevention of pre-eclampsia in high-risk women: a randomised placebo-controlled PREDO Trial and a meta-analysis of randomised trials. BJOG. 2013;120: 64–74.
  16. 16. Radloff LS. The CES-D Scale A Self-Report Depression Scale for Research in the General Population. Applied Psychological Measurement. 1977;1: 385–401.
  17. 17. Spielberger CD. Manual for the State-Trait Anxiety Inventory. 1983.
  18. 18. Watson D, Clark LA, Tellegen A. Development and validation of brief measures of positive and negative affect: the PANAS scales. J Pers Soc Psychol. 1988;54: 1063–1070. pmid:3397865
  19. 19. Pihkala J, Hakala T, Voutilainen P, Raivio K. Characteristic of recent fetal growth curves in Finland. Duodecim. 1989;105: 1540–1546. pmid:2680445
  20. 20. Mancuso RA, Schetter CD, Rini CM, Roesch SC, Hobel CJ. Maternal prenatal anxiety and corticotropin-releasing hormone associated with timing of delivery. Psychosom Med. 2004;66: 762–769. pmid:15385704
  21. 21. Orr ST, Blazer DG, Orr CA. Maternal prenatal depressive symptoms, nicotine addiction, and smoking-related knowledge, attitudes, beliefs, and behaviors. Matern Child Health J. 2012;16: 973–978. pmid:21607614
  22. 22. Huizink AC, Menting B, Oosterman M, Verhage ML, Kunseler FC, Schuengel C. The interrelationship between pregnancy-specific anxiety and general anxiety across pregnancy: a longitudinal study. J Psychosom Obstet Gynaecol. 2014;35: 92–100. pmid:25093467
  23. 23. Henrichs J, Schenk JJ, Roza SJ, van den Berg MP, Schmidt HG, Steegers EA, et al. Maternal psychological distress and fetal growth trajectories: the Generation R Study. Psychol Med. 2010;40: 633–643. pmid:19656431
  24. 24. Hobel CJ, Dunkel-Schetter C, Roesch SC, Castro LC, Arora CP. Maternal plasma corticotropin-releasing hormone associated with stress at 20 weeks' gestation in pregnancies ending in preterm delivery. Am J Obstet Gynecol. 1999;180: S257–263. pmid:9914629
  25. 25. Lockwood CJ. Stress-associated preterm delivery: the role of corticotropin-releasing hormone. Am J Obstet Gynecol. 1999;180: S264–266. pmid:9914630
  26. 26. Sandman CA, Wadhwa PD, Chicz-DeMet A, Dunkel-Schetter C, Porto M. Maternal stress, HPA activity, and fetal/infant outcome. Ann N Y Acad Sci. 1997;814: 266–275. pmid:9160976
  27. 27. Moisiadis VG, Matthews SG. Glucocorticoids and fetal programming part 2: Mechanisms. Nat Rev Endocrinol. 2014;10: 403–411. pmid:24863383
  28. 28. Reynolds RM. Glucocorticoid excess and the developmental origins of disease: two decades of testing the hypothesis—2012 Curt Richter Award Winner. Psychoneuroendocrinology. 2013;38: 1–11. pmid:22998948
  29. 29. Wadhwa PD, Entringer S, Buss C, Lu MC. The contribution of maternal stress to preterm birth: issues and considerations. Clin Perinatol. 2011;38: 351–384. pmid:21890014
  30. 30. Field T, Diego M. Cortisol: the culprit prenatal stress variable. Int J Neurosci. 2008;118: 1181. pmid:18589921
  31. 31. Field T, Diego M, Hernandez-Reif M. Prenatal depression effects on the fetus and newborn: a review. Infant Behav Dev. 2006;29: 445–455. pmid:17138297
  32. 32. O'Keane V, Lightman S, Marsh M, Pawlby S, Papadopoulos AS, Taylor A, et al. Increased pituitary-adrenal activation and shortened gestation in a sample of depressed pregnant women: a pilot study. J Affect Disord. 2011;130: 300–305. pmid:21093926
  33. 33. Rich-Edwards JW, Mohllajee AP, Kleinman K, Hacker MR, Majzoub J, Wright RJ, et al. Elevated midpregnancy corticotropin-releasing hormone is associated with prenatal, but not postpartum, maternal depression. J Clin Endocrinol Metab. 2008;93: 1946–1951. pmid:18303075
  34. 34. Edwards CR, Benediktsson R, Lindsay RS, Seckl JR. Dysfunction of placental glucocorticoid barrier: link between fetal environment and adult hypertension? Lancet. 1993;341: 355–357. pmid:8094124
  35. 35. Korhonen M, Luoma I, Salmelin R, Tamminen T. A longitudinal study of maternal prenatal, postnatal and concurrent depressive symptoms and adolescent well-being. J Affect Disord. 2012;136: 680–692. pmid:22036793
  36. 36. O'Donnell KJ, Glover V, Barker ED, O'Connor TG. The persisting effect of maternal mood in pregnancy on childhood psychopathology. Dev Psychopathol. 2014;26: 393–403. pmid:24621564
  37. 37. Pearson RM, Evans J, Kounali D, Lewis G, Heron J, Ramchandani PG, et al. Maternal depression during pregnancy and the postnatal period: risks and possible mechanisms for offspring depression at age 18 years. JAMA Psychiatry. 2013;70: 1312–1319. pmid:24108418
  38. 38. Van den Bergh BR, Marcoen A. High antenatal maternal anxiety is related to ADHD symptoms, externalizing problems, and anxiety in 8- and 9-year-olds. Child Dev. 2004;75: 1085–1097. pmid:15260866
  39. 39. Van Batenburg-Eddes T, Brion MJ, Henrichs J, Jaddoe VW, Hofman A, Verhulst FC, et al. Parental depressive and anxiety symptoms during pregnancy and attention problems in children: a cross-cohort consistency study. J Child Psychol Psychiatry. 2013;54: 591–600. pmid:23215861
  40. 40. Kertes DA, Kamin HS, Hughes DA, Rodney NC, Bhatt S, Mulligan CJ. Prenatal Maternal Stress Predicts Methylation of Genes Regulating the Hypothalamic-Pituitary-Adrenocortical System in Mothers and Newborns in the Democratic Republic of Congo. Child Dev. 2016;87: 61–72. pmid:26822443
  41. 41. Palma-Gudiel H, Cordova-Palomera A, Eixarch E, Deuschle M, Fananas L. Maternal psychosocial stress during pregnancy alters the epigenetic signature of the glucocorticoid receptor gene promoter in their offspring: a meta-analysis. Epigenetics. 2015;10: 893–902. pmid:26327302
  42. 42. Raikkonen K, Pesonen AK, O'Reilly JR, Tuovinen S, Lahti M, Kajantie E, et al. Maternal depressive symptoms during pregnancy, placental expression of genes regulating glucocorticoid and serotonin function and infant regulatory behaviors. Psychol Med. 2015;45: 3217–3226. pmid:26096463