Hum. Reprod. Advance Access originally published online on December 3, 2008
Human Reproduction 2009 24(2):429-437; doi:10.1093/humrep/den418
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Rates of preterm birth following antenatal maternal exposure to severe life events: a population-based cohort study
1 Centre for Women’s Mental Health, University of Manchester, Manchester, UK 2 Biostatistics Group, University of Manchester, Manchester, UK 3 Maternal and Fetal Health Research Group, University of Manchester, Manchester, UK 4 National Centre for Register-based Research, University of Aarhus, Aarhus, Denmark 5 The Anu Research Center, Department of Obstetrics and Gynaecology, Cork University Maternity Hospital, Cork Ireland
6 Correspondence address. Tel: +353-214205031; Fax: +353-214205025; E-mail: a.khashan{at}ucc.ie
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionAuthors contributionFundingReferences |
|---|
BACKGROUND: Preterm birth and other pregnancy complications have been linked to maternal stress during pregnancy. We investigated the association between maternal exposure to severe life events and risk of preterm birth.
METHODS: Mothers of all singleton live births (n = 1.35 million births) in Denmark between 1 January 1979 and 31 December 2002 were linked to data on their children, parents, siblings and partners. We defined exposure as death or serious illness in close relatives in the first or second trimesters or in the 6 months before conception. Log-linear binomial regression was used to estimate the effect of exposure on preterm birth, very preterm birth and extremely preterm birth.
RESULTS: There were 58 626 (4.34%) preterm births (<37 weeks), 11 732 (0.87%) very preterm births and 3288 (0.24%) extremely preterm births in the study cohort. Severe life events in close relatives in the 6 months before conception increased the risk of preterm birth by 16% (relative risk, RR = 1.16, [95% CI: 1.08–1.23]). Severe life events in older children in the 6 months before conception increased the risk of preterm birth by 23% (RR = 1.23, [95% CI: 1.02–1.49]) and the risk of very preterm birth by 59% (RR = 1.59, [95% CI: 1.08–2.35]).
CONCLUSIONS: Our population-based cohort study suggests that maternal exposure to severe life events, particularly in the 6 months before pregnancy, may increase the risk of preterm and very preterm birth.
Key words: preterm birth/maternal stress/pregnancy/severe life events/obstetric complications
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAuthors contributionFundingReferences |
|---|
Preterm birth is a pregnancy complication with significant healthcare implications; in 2004, 12.8% of all births in the USA were preterm (National Centre for Health Statistics, 2006
) and recent figures showed that 1.3% of all deliveries in the UK (approximately 10 000 births p.a.) occur before 32 weeks’ gestation (Department of Health, 2007) when mortality and morbidity is most severe. In the UK, 75% of neonatal deaths, and the majority of neonatal intensive care admissions, are from these preterm babies (Lopez, 2001; Slattery and Morrison, 2002). In some industrialized countries, particularly the USA, the preterm birth rate has remained high over the past 20–30 years; indeed, there is a small upward trend (Goldenberg and Rouse, 1998; Tucker and McGuire, 2004; National Centre for Health Statistics, 2006). However, this increase in preterm births rate has been in part attributed to obstetric interventions such as induction of labour (Thompson et al., 2006).
Several psychosocial stressors, such as pregnancy-related anxiety (Orr et al., 2007), stressful life events (Wadhwa et al., 1993), distress (Rondo et al., 2003), death of husband (Cepick
and Mandys, 1989) and low self-esteem (Copper et al., 1996) or optimism (Rini et al., 1999), have been investigated as potential risk factors for preterm birth. Although the majority of researchers have reported a significant association between preterm delivery and maternal stress (Hedegaard et al., 1993; Wadhwa et al., 1993; Copper et al., 1996; Nordentoft et al., 1996; Lobel et al., 2000; Dole et al., 2003; Rondo et al., 2003), other studies have questioned the association (Cepick and Mandys, 1989; Pagel et al., 1990; Perkin et al., 1993; Peacock et al., 1995). There are several potential explanations for such conflicting conclusions, including poor definitions of stressors and inadequate sample size, which may have influenced the results of previous studies. Additionally, we have recently reported an association between maternal exposure to severe life events during or in the 6 months before pregnancy and reduced infant birthweight and increased risk of SGA (Khashan et al., 2008). Precht et al. (2007) reported an association between severe life events in the year preceding pregnancy and risk of SGA in babies born before 32 gestation weeks (Precht et al., 2007).
In this study, we used the Danish national registers to evaluate the association between antenatal maternal exposure to severe life events (death and serious illness) in close relatives and the risks of preterm birth (delivery at <37 weeks’ gestation), very preterm birth (delivery at <33 weeks’ gestation) and extremely preterm birth (delivery at <29 weeks’ gestation). We chose to define exposure to severe life events as death and serious illness in close relatives as these were events that could be measured objectively in the registers. We hypothesized that risk of preterm birth would be increased in mothers who were exposed to severe life events antenatally. We also explored the effect of timing of exposure during pregnancy and preconceptionally on risk of preterm birth.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAuthors contributionFundingReferences |
|---|
All women who gave birth to live singleton babies in Denmark between 1 January 1979 and 31 December 2002 (n = 1 439 122) were identified using the Medical Birth Register (Knudsen et al., 1998
). For the main analyses, mothers with a previous preterm baby were excluded because history of preterm birth is one of the main predictors of consequent preterm birth (Vause and Johnston, 2000). By means of the Civil Registration Number, we linked mothers to their close relatives, i.e. fathers, mothers, siblings, partners and older children (excluding stillbirths). We had access to information on older children who were born on or after 1 January 1973 only. Partner was defined as the legal father of the offspring. The Civil Registration Number is unique and enables linkage between all Danish national registers. To determine if and when any of the close relatives had died, we linked these relatives to the Civil Registration System (Pedersen et al., 2006). We also linked relatives to the National Hospital Register (Anderson et al., 1999), to identify any diagnoses of acute myocardial infarction, cancer or cerebrovascular accident. We conducted a separate analysis in which only mothers with history of preterm birth (previous baby only) were included. We defined maternal exposure as experience of a death or diagnosis, for the first time, of cancer (ICD-8 codes 140–207 and ICD-10 codes C00–C97), acute myocardial infarction (ICD-8 code 410 and ICD-10 codes I21 and I22) or cerebrovascular accident (ICD-8 codes 431, 433, 434 and ICD-10 codes I61, I63, I64) in one or more of the mother’s close relatives up to 6 months before conception or during the first two trimesters of pregnancy. The date of conception was calculated from the date of birth and the recorded gestational age at birth. We defined date of first maternal exposure as date of death of the relative or the date of their first admission to hospital which led to diagnosis of the illness. Date of first exposure was further classified into three periods: up to 6 months before pregnancy; first trimester (0–12 weeks) and second trimester (13–23 weeks). Mothers were considered exposed if they had links to all close relatives, and at least one relative died or was diagnosed with a relevant illness during the exposure period. They were considered unexposed if they had links to all close relatives and none of them was diagnosed with a relevant illness or died during the exposure period. If more than one exposure event occurred during the same pregnancy, priority for classification was given to the earliest event.
Mothers who were not living at the same address as their parents when the Danish Civil Registration System was established in 1968 had no links to them in this data source and subsequently had no links to their siblings either. A small proportion of women had missing links to their partners. The main source of this missing link to spouse is the death of the child shortly after birth (stillbirth or neonatal death). Therefore, it is anticipated that children of mothers who had no links to their partners would have very high risk of preterm birth, very preterm birth and extremely preterm birth. For the purpose of sensitivity analysis, subjects with missing exposure data were grouped in a separate category.
Gestational age was based on the last menstrual period (LMP), however, it was often corrected by ultrasound measurements, mainly for the most recent period (Pedersen et al., 2007). A recorded gestational age of <23 or >44 weeks was considered erroneous, and so these births were excluded from the study. We also excluded offspring of mothers who were exposed in the third trimester. Three measures of preterm birth were investigated in relation to maternal exposure to severe life events (Tucker and McGuire, 2004); (i) preterm birth, defined as a gestational age <37 weeks; (ii) very preterm birth, defined as a gestational age <33 weeks and (iii) extremely preterm birth, defined as any gestational age <29 weeks. The risk of preterm birth in the exposed group was defined as the number of preterm births divided by the total number of births in the exposed group. Similarly, very preterm and extremely preterm birth risks were defined as number of very preterm or extremely preterm births divided by the total number of births in each group.
Statistical analysis
Stata Software (Stata Reference Manual, 2003) was used for all statistical analyses. Log-linear binomial regression (Robbins et al., 2002) was used to estimate the relative risks (RRs) of preterm, very preterm and extremely preterm birth. When convergence was not achieved (a recognized problem with this model), log-linear Poisson regression with ‘robust’ variance estimation was used to generate the equivalent risk ratio (Zou, 2004). For each preterm birth measure, RRs were estimated in relation to exposure to: (i) severe life events (death and/or illness) in any relative; (ii) illness in any relative; (iii) death of any relative and (iv) severe life events in each relative separately. When we analysed severe life events in each parent separately, we excluded offspring of women who had no links to that parent or whose parent had died or emigrated from Denmark before the beginning of the exposure period. For the analyses of siblings and older children of the pregnant mother, we excluded offspring of women who had no siblings or older children, respectively. RRs were adjusted for maternal age, year of birth of the offspring, parity, maternal history of diabetes and maternal history of hypertension, acute myocardial infarction and renal disease (at the time of birth). In addition, analyses of severe life events in siblings were adjusted for number of siblings. In a separate model, we adjusted for maternal smoking as recorded during the first antenatal visit (available for the years 1991–1996 only) in addition to the other variables. To explore the potential effect of gestational age misclassification on the results, we restricted the data to those who were born between 1991 and 1996 when there was a variable indicating whether the women were certain about the LMP or not. Also, we analysed the data after excluding infants who were implausibly large for the recorded gestational age (if birthweight was >1500 g and gestational age <29 weeks or if birthweight was more than 2800 g and gestational age between 29 and 33 weeks) (Khashan et al., 2008).
The potential impact of exposure to severe life events during the third trimester was not studied because, in this type of analysis, the RRs would have been biased downwards due to an inbuilt tendency for a negative correlation between the probability of exposure between 24 weeks and birth and preterm birth.
The Danish Data Protection Agency and the Danish National Board of Health gave permission to conduct this study.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAuthors contributionFundingReferences |
|---|
During the study period (1979–2002), there were 1 439 122 singleton live births. After excluding those with missing gestational age (n = 47 197), history of preterm delivery (n = 28 324) or exposure during the third trimester (n = 11 990), the final cohort consisted of mothers of 1 351 611 babies of which 1 022 402 (75.6%) had exposure data. Mean gestational age of the cohort was 39.6 weeks. Table I presents the distribution of the three preterm birth measures, sex of baby, maternal age and history of diabetes, hypertension and renal disease and parity in relation to exposure status. During the study period, there were 58 626 preterm babies (4.34%), 11 732 very preterm babies (0.87%) and 3288 extremely preterm babies (0.24%). Mothers of 36 199 babies (2.65%) were exposed to severe life events during the exposure period (18 919 before pregnancy, 8586 in first trimester and 8694 in second trimester).
|
The exposed mothers appeared to be older, had higher parity and were more likely to have a history of diabetes and were slightly more likely to have a history of hypertension or renal disease (Table I). The RR of preterm birth was slightly elevated in relation to severe life events (adjusted RR = 1.09, [95% CI: 1.04–1.14]). However, the RR of preterm birth appeared to be higher when the severe life event took place in the 6 months before pregnancy (adjusted RR = 1.16, [95% CI: 1.08–1.23]) (Table II). RRs were the same whether the event was the death of a relative (adjusted RR = 1.16, [95% CI: 1.06–1.27]) or an illness in a relative (adjusted RR = 1.16, [95% CI: 1.07–1.27]) in the 6 months before conception (Table III). There was no significant association between risk of preterm birth and maternal exposure to any severe life event during the first (RR = 1.03) or second trimesters (RR = 1.01).
|
|
We also investigated the association between maternal exposure to severe life events and the risk of very and extremely preterm birth. The estimates do not support an association between maternal exposure to severe life events during the exposure period and risk of very or extremely preterm birth. Moreover, we restricted the cohort to the period 1991–1996 where smoking information was available. Analyses of this subgroup suggest that smoking was a risk factor for preterm birth. However, the RR of preterm birth in relation to severe life events in the 6 months before conception was not affected by adjustment for smoking, being 1.10 (95% CI: 0.98–1.24) without adjustment for maternal smoking and 1.09 (95% CI: 0.97–1.23) after adjustment. This analysis suggests therefore that, in the whole cohort, smoking was unlikely to have confounded the association between severe life events and risk of preterm birth. When the data were restricted to births between 1991 and 1996 (n = 384 472), it appeared that mothers of 84.7% of the offspring were ‘certain’ of their LMP, 11.6% were ‘uncertain’ and 3.7% had missing gestational age. However, 15.2% were uncertain when the data were restricted to very preterm birth. The RR of preterm birth in relation to severe life events in close relatives was 1.05 (95% CI: 0.97–1.15). When the data were restricted to those who were certain about their LMP, the RR was 1.04 (95% CI: 0.94–1.14). When we excluded infants with implausibly large birthweight for the recorded gestational age (n = 1194), the results did not change materially (data not shown).
Mothers of 24.4% of the offspring had missing links to at least one relative, mainly parents of the mothers. Mothers of 0.87% offspring had no links to their partners. Those women tended to be older and more likely to have given birth in the early periods of the cohort. Tables II and III show that women with missing relatives were at significantly higher risk of preterm birth, very preterm birth and extremely preterm birth.
Type of exposure
We estimated the RRs of preterm birth and very preterm birth in relation to severe life events in fathers, mothers, siblings, partners and older children individually. There was a significant association between risk of preterm birth and severe life events in fathers (1.07 [95% CI: 1.02–1.13]), but not in mothers, partners and siblings (Table IV). However the differences in risk between these categories were small, and events in fathers accounted for the majority of the serious events.
|
The association between maternal exposure to death or illness in an older child and risk of preterm birth and very preterm birth are shown in Table V; we were not able to estimate the RR of extremely preterm birth due to the small number of exposed cases. This comparison was restricted to mothers who had a previous baby (born in 1973 or later) which was full-term (n = 600 259). There were two reasons behind this restriction; firstly, any pregnant woman needs to have an older child in order to have the possibility of being exposed to severe life events in older child. In the restricted cohort, every woman was at risk of exposure to severe life events in older children. Secondly, in the whole data set, women who were having their first baby were at higher risk of preterm birth compared with those who had previous full-term baby (RR = 1.53, [95% CI: 1.45–1.59]).
|
The estimates suggest that the risk of preterm birth in mothers who experienced severe life events in older children in the 6 months before conception was elevated by 23% (adjusted RR = 1.23, [95% CI: 1.02–1.49]). The risks of preterm birth in relation to maternal exposure to severe life events in older children during the first (adjusted RR = 1.16, [95% CI: 0.60–2.25]) or second trimesters (adjusted RR = 1.02, [95% CI: 0.39–2.67]) were not significantly increased. There was a more than 3-fold increase in risk of very preterm birth in mothers who were exposed to severe life events in older children in the first trimester (adjusted RR = 3.67, [95% CI: 1.55–8.69]), but this estimate was based on only five exposed cases. Severe life events in older children in the 6 months before conception appeared to increase the risk of very preterm birth by 60% (adjusted RR = 1.59, [95% CI: 1.08–2.35]).
Severe life events in older children among mothers with history of preterm birth
We estimated the effect of maternal exposure to severe life events in older children on risk of preterm birth, very preterm birth and extremely preterm birth in the current pregnancy among women who had preterm delivery the last time they were pregnant. This analysis was planned at a latter stage as an attempt to replicate the previous results for exposure to severe life events in older children in mothers with no history of preterm birth. The logic of this extra analysis was that since children born preterm have an increased mortality, one might speculate that women with a tendency to have preterm delivery more often would experience bereavement of a child during subsequent pregnancies, thus introducing bias. Moreover, women who had previous preterm birth are at higher risk of preterm birth. Our analysis restricted to those with a history of preterm birth takes this possibility into account. The cohort consisted of 28 271 women whose previous baby was preterm, of which 27 465 were unexposed during the current pregnancy, and 796 were exposed to severe life events in older children (769 in the 6 months before conception, 17 in the first trimester and 10 in the second trimester). Mothers of 10 offspring were exposed in the third trimester but these were dropped from the statistical models, i.e. the final cohort was 28 261. There was a significant association between maternal exposure to severe life events in older children in the 6 months before conception and preterm birth (adjusted RR = 1.27, [95% CI: 1.12–1.44]), very preterm birth (adjusted RR = 2.30, [95% CI: 1.84–2.86]) and extremely preterm birth (adjusted RR = 2.75, [95% CI: 1.84–4.12]). The RRs of the outcome measures were not estimated in relation to events in the first or second trimesters due to sparse outcome data.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAuthors contributionFundingReferences |
|---|
This study found that antenatal maternal exposure to severe life events is associated with a modest increase in the risk of preterm birth. We found that maternal exposure to severe life events in close relatives during the 6 months before conception slightly increased the risk of preterm birth but not of very or extremely preterm birth. In the group where the older children had not been born preterm, maternal exposure to severe life events in older children in the 6 months before conception was associated with risk of preterm birth and very preterm birth. An especially high risk of very preterm birth was found in women who experienced severe life events in older children in the first trimester, although outcome data were sparse for this particular analysis. Among mothers whose previous baby was preterm, severe life events in the 6 months before conception were significantly linked with preterm, very and extremely preterm birth.
No significant association was observed following exposure to severe life events in mothers, siblings and partners, although there was a small, yet significant, association with severe life events in fathers. These severe life events may have evoked a lesser degree of maternal stress, as compared with severe life events in children. Such a finding would be consistent with the diagnostic criteria of the American Psychiatry Association; death of a child is classified as a catastrophic stressor while death of spouse is classified as an extreme stressor (American Psychiatry Association, 1987). This is also consistent with previous studies which reported significant associations between maternal exposures to severe life events in older children and risk of congenital malformations (Hansen et al., 2000) and SGA in very preterm babies (Precht et al., 2007).
Previous studies have investigated the effect of antenatal maternal stress on risk of preterm birth (<37 weeks); we aimed to investigate the effect on risk of preterm, very preterm and extremely preterm birth. We restricted the definition of severe life events to severe and objective events, which are rare, to secure a robust definition of severe life events. We believe such events would probably cause a high degree of stress in any person. However, we cannot rule out the possibility that less catastrophic and more common events (such as work related and financial problems) may also increase the risk of preterm birth. We were able to identify the timing of the event, during and before pregnancy.
Our findings are consistent with the majority of literature on preterm birth and maternal exposure to psychosocial stress. Dole et al. (2003) reported an increased risk of preterm birth among mothers who reported adverse life events, perceived racial discrimination and high levels of pregnancy-related anxiety (assessed between weeks 24 and 30). Copper et al. (1996) used a continuous measure of stress (4-point scale) and reported a 16% increase in risk of spontaneous preterm delivery, for each 1-point increase on the scale, among women who perceived their life during pregnancy to have been stressful (assessed between gestation weeks 25 and 29). Nordentoft et al. (1996) reported a 14% increase in risk of preterm birth in women who experienced stressful life events (assessed around 20 weeks’ gestation) during pregnancy (5-point continuous measure of stress). Other researchers (Pritchard and Teo, 1994) measured household strain and stress at the 20th and 30th week of pregnancy. The authors reported an increased risk of preterm birth among women who were classified as having high strain of the household role on both measurements. However, there was no significant relationship between stress and preterm birth. These studies did not investigate the effect of timing of exposure on risk of preterm birth; they assessed maternal stress only once during pregnancy.
Other investigators (Hedegaard et al., 1993; Rondo et al., 2003) reported more than 2-fold increased RR of preterm birth among mothers who were exposed to maternal distress (12-item General Health Questionnaire [GHQ]) during the third trimester. However, in contrast to the results of our study, they found no significant association between maternal distress during the early stages of pregnancy (at the 16th week) and risk of preterm birth.
The RRs of preterm and very preterm birth that we observed in women with history of preterm birth in relation to severe life events in older children were consistent with the results of a recent study (Dayan et al., 2002). In a similar population, Dayan et al. found that maternal anxiety and depression evoked an almost 5-fold increase in risk of spontaneous preterm labour.
In contrast, Peacock et al. (1995) found no significant association between anxiety score and preterm delivery. However, women who reported ‘trouble with nerves and depression’ had a 2-fold non-significant elevation in risk of preterm birth. Perkin et al. (1993) reported a non-significant association between depression and anxiety (GHQ) during pregnancy and risk of preterm birth.
Despite being the largest population-based epidemiological study to investigate the association between risk of preterm birth and antenatal maternal exposure to severe life events, there were several limitations to this study. We could not adjust for maternal socioeconomic status (SES) which is a potential confounder of the association between maternal stress and preterm delivery. It is plausible that low SES mothers have larger family sizes (Wagner et al., 1985) with higher rates of morbidity and mortality, and therefore are more likely to be exposed; for example McFadden et al. (2008) reported an increased risk of all-cause mortality, mortality from cardiovascular disease and mortality from cancer with decreasing social class, and Blakely et al. (2003) found an association between low SES and child mortality. Furthermore, risk of preterm birth is increased in low SES mothers (Steer and Flint, 1999). Recently, risk of preterm birth in the Danish and Norwegian populations was reported to have an inverse association with level of education (Thompson et al., 2006; Morgen et al., 2008). Morgen et al. (2008), however, found no association between risk of preterm birth and other SES indicators such as household income and parental occupation. Adjustment for family size in our study probably accounted for part of the potential confounding effect of SES but not all of it. Moreover, educational level is associated with the accuracy of gestational age (Howards et al., 2006) and the number of preterm births in the Medical Birth Register of Norway is expected to be considerably smaller than suggested, due to misclassification of gestational age (Gjessing et al., 1999). Kristensen et al. (1996) assessed the validity of gestational age in the Medical Birth Register and the National Hospital Register using 1982–1987 data. They reported an 87% agreement between the two registers when gestational age was measured in weeks. However, some research has suggested that misclassification of gestational age varies by gestational week (Gjessing et al., 1999; Haglund, 2007; Wingate et al., 2007). Thus, the present results could be confounded by level of education, and other indicators of SES, as these indicators are associated with risk of mortality and serious illness in the relatives and at the same time associated with risk of preterm birth. We were also unable to adjust for the potential confounding effect of maternal drug and alcohol intake.
Although the study was population-based, we were unable to investigate the effect of death of an older child on risk of extremely preterm birth due to sparse event data (a problem caused by the rarity of the exposures and outcomes). One further limitation was the fact that we could not investigate the effect of severe life events during the third trimester on the outcome measures.
Mothers of 24% of the offspring had missing links to at least one relative mainly parents of the pregnant mother. The RRs of preterm birth, very preterm birth and extremely preterm birth were significantly increased in mothers with missing relatives.
The highest risks were associated with missing partners; for example, the RR of extremely preterm birth was increased to 1.20 in relation to missing links to parents or siblings and increased to almost 13 in relation to missing partners. This increase was expected because the main source of missing links to partners was early death (stillbirth or neonatal death) of the infants and stillbirth and neonatal deaths are highly correlated with preterm birth (Tucker and McGuire, 2004). Missing links to partners were not expected to bias the results because they were grouped in a separate category and they represent a small percentage (<1%) of the cohort. Pregnant mothers had links to all their older children (who were born on or after 1 January 1973) therefore the issue of missing links should not affect the results related to death of an older child.
The effect of the timing of exposure that we observed in this study is consistent with the results of two recent studies which reported an association between severe life events in the 6 months before pregnancy and reduced birthweight (Khashan et al., 2008) and risk of SGA in very preterm babies (Precht et al., 2007). There are several potential mechanisms that may account for the association between maternal stress during the 6 months before conception or during the first trimester and a higher risk of preterm birth. It has previously been suggested that increased levels of epinephrine, as a consequence of stress, reduce blood flow to the fetus. It is plausible that severe stress may increase epinephrine levels which in turn precipitate preterm labour (Lobel et al., 1992). Maternal stress may also affect risk of preterm birth by increasing levels of pituitary hormones such as oxytocin. High levels of oxytocin later in pregnancy may cause preterm uterine contractions and trigger preterm delivery (Wadhwa et al., 1993). Moreover, it has been reported that the uterus of pregnant women who give birth to preterm babies appear to be more sensitive to oxytocin than women who give birth to full-term babies (Paarlberg et al., 1995). In addition, antenatal maternal stress may increase risk of preterm birth by changing maternal habits, such as increased consumption of tobacco and alcohol, poorer diet and use of drugs.
While the effect of severe life events on the risk of preterm birth was modest, the findings remain significant. Firstly, the etiology of preterm birth remains poorly understood and these findings suggest the possibility of several causative mechanisms worthy of further investigation. Moreover, while the effect is relatively modest, preterm birth remains the single largest contributor to perinatal mortality in the developed world. Consequently, even a moderate effect at a population level can contribute significantly to morbidity and mortality.
In conclusion, the present study suggests that severe life events may play a role in increasing the risk of preterm birth. It also suggests the timing of the exposure is crucial to the provocation of preterm birth as the effect is only present with pre-pregnancy and first trimester exposure. It is possible that the effects are mediated by an alteration in the uterine environment early in fetal/placental development.
|
Authors contribution |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAuthors contributionFundingReferences |
|---|
All the authors contributed to the study concept, design, analysis and interpretation of data. P.B.M. and M.G.P.: acquisition of data; A.S.K.: drafting of the manuscript; R.M., K.M.A., P.B.M., L.C.K., M.G.P., R.T.W. and P.N.B.: critical revision of the manuscript for important intellectual content; A.S.K., R.M. and M.G.P.: statistical analysis; All authors approved the submission of this manuscript. Carsten B. Pedersen, M.Sc., National Centre for Register Based Research, University of Aarhus, Denmark, assisted with the study design and analysis.
|
Funding |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAuthors contributionFundingReferences |
|---|
This study was funded by Tommy’s the Baby Charity (Philip N. Baker) and the Stanley Medical Research Institute (Preben Bo Mortensen).
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAuthors contributionFundingReferences |
|---|
American Psychiatry Association. Diagnostic criteria from DSM-III-R. (1987).
Anderson F, Madsen M, Jorgensen J, Mellemkjær L, Olsen JH. The Danish National Hospital Register. Dan Med Bull (1999) 46:263–268.[Web of Science][Medline]
Blakely T, Atkinson J, Kiro C, Blaiklock A, D'Souza A. Child mortality, socioeconomic position, and one-parent families: independent associations and variation by age and cause of death. Int J Epidemiol (2003) 32:410–418.
Cepick P, Mandys F. Reproductive outcome in women who lost their husbands in the course of pregnancy. Eur J Obstet Gynoecol Reprod Biol (1989) 30:137–140.
Copper RL, Goldenberg RL, Das A, Elder N, Swain M, Norman G, Ramsey R, Cotroneo P, Colins BA, Johnson F, et al. The preterm prediction study: maternal stress is associated with spontaneous preterm birth at less than thirty-five weeks’ gestation. Am J Obstet Gynecol (1996) 175:1286–1292.[CrossRef][Web of Science][Medline]
Dayan J, Creveuil C, Herlicoviez M, Herbel C, Baranger E, Savoye C, Thouin A. Role of anxiety and depression in the onset of spontaneous preterm labor. Am J Epidemiol (2002) 155:293–301.
Department of Health. NHS Maternity Statistics, 2005–2006. (2007) England.
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.
Gjessing HK, Skjærven R, Wilcox AJ. Errors in gestational age: evidence of bleeding early in pregnancy. Am J Public Health (1999) 89:213–218.
Goldenberg RL, Rouse DJ. Prevention of premature birth. N Engl J Med (1998) 339:313–320.
Haglund B. Birthweight distributions by gestational age: comparison of LMP-based and ultrasound-based estimates of gestational age using data from the Swedish Birth Registry. Paediatr Perinat Epidemiol (2007) 21(Suppl. 2):72–78.[CrossRef][Web of Science][Medline]
Hansen D, Lou HC, Olsen J. Serious life events and congenital malformations: A national study with complete follow-up. Lancet (2000) 356:875–880.[CrossRef][Web of Science][Medline]
Hedegaard M, Henriksen TB, Sabroe S, Secher NJ. Psychological distress in pregnancy and preterm delivery. BMJ (1993) 307:234–241.
Howards PP, Hertz-Picciotto I, Weinberg CR, Poole C. Misclassification of gestational age in the study of spontaneous abortion. Am J Epidemiol (2006) 164:1126–1136.
Khashan AS, McNamee R, Abel KM, et al. Reduced infant birthweight consequent upon maternal exposure to severe life events. Psychosom Med (2008) 70:688–694.
Knudsen LB, mag.scient.soc, Olsen J. The Danish Medical Birth Registry. Dan Med Bull (1998) 45:320–323.[Web of Science][Medline]
Kristensen J, Langhoff-Ross J, Skovgaard LT, Kristensen FB. Validation of the Medical Birth Registration. J Clin Epidemiol (1996) 49:893–897.[CrossRef][Web of Science][Medline]
Lobel M, Dunkel-Schetter C, Scrimshaw SCM. Prenatal maternal stress and prematurity: a prospective study of socio-economically disadvantaged women. Health Psychol (1992) 11:32–40.[CrossRef][Web of Science][Medline]
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.[CrossRef][Web of Science][Medline]
Lopez BA. Overview of current research in parturition. Exp Physiol (2001) 86:213–222.[Abstract]
McFadden E, Luben R, Wareham N, Bingham S, Khaw KT. Occupational social class, educational level, smoking and body mass index, and cause-specific mortality in men and women: a prospective study in the European prospective investigation of cancer and nutrition in Norfolk (EPIC-Norfolk) cohort. Eur J Epidemiol (2008) 23:511–522.[CrossRef][Medline]
Morgen CS, Bjørk C, Andersen PK, Mortensen LH, Andersen AMN. Socioeconomic position and the risk of preterm birth-a study within the Danish National Birth Cohort. Int J Epidemiol (2008) 37:1109–1120.
National Centre for Health Statistics. Births: preliminary data for 2006.
Nordentoft M, Lou HC, Hansen D, Nim J, Pryds O, Rubin P, Hemmingsen R. Intrauterine growth retardation and premature delivery: the influence of maternal smoking and psychosocial factors. Am J Public Health (1996) 86:347–354.
Orr ST, Reiter JP, Blazer DG, James SA. Maternal prenatal pregnancy-related anxiety and spontaneous preterm birth in Baltimore, Maryland. Psychosom Med (2007) 69:566–570.
Paarlberg KM, Vingerhoets AJJM, Passchier J, Dekker GA, Geijn HPV. Psychosocial factors and pregnancy outcome: a review with emphasis on methodological issues. J Psychosom Res (1995) 39:563–595.[CrossRef][Web of Science][Medline]
Pagel MD, Smilkstein G, Regen H, Montano D. Psychosocial influences on new born outcomes: a controlled prospective study. Soc Sci Med (1990) 30:597–604.[CrossRef][Web of Science][Medline]
Peacock JL, Bland JM, Anderson HR. Preterm delivery: effects of socioeconomic factors, psychological stress, smoking, alcohol and caffeine. BMJ (1995) 311:531–539.
Pedersen CB, Gøtzsche H, Møller JØ, Mortensen PB. The Danish Civil Registration System. Dan Med Bull (2006) 53:441–449.[Medline]
Pedersen CB, Sun Y, Vestergaard M, Olsen J, Basso O. Assessing fetal growth impairments based on family data as a tool for indentifying high-risk babies. An example with neonatal mortality. BMC Pregnancy Childbirth (2007) 7:28.[Medline]
Perkin MR, Bland JM, Peacock JL, Anderson HR. The effect of anxiety and depression during pregnancy on obstetric complications. Br J Obstet Gynecol (1993) 100:629–634.[Web of Science][Medline]
Precht DH, Andersen PK, Olsen J. Severe life events and impaired fetal growth: a nation-wide study with complete follow-up. Acta Obstet Gynecol Scand (2007) 86:266–275.[Medline]
Pritchard CW, Teo PYK. Preterm birth, low birthweight and the stressfulness of the household role for pregnant women. Soc Sci Med (1994) 38:89–96.[CrossRef][Web of Science][Medline]
Rini CK, Dunkel-Schetter C, Wadhwa PD, Sandman CA. Psychological adaptation and birth Outcomes: the role of personal resources, stress and socio-cultural context in pegnancy. Health Psychol (1999) 18:333–345.[CrossRef][Web of Science][Medline]
Robbins AS, Chao SY, Fonseca VP. What’s the relative risk? A method to directly estimate risk ratios in cohort studies of common outcomes. Ann Epidemiol (2002) 12:452–454.[CrossRef][Web of Science][Medline]
Rondo PHC, Ferreira RF, Nogueira F, Ribeiro MCN, Lobert H, Artes R. Maternal psychological stress and distress as predictors of low birthweight, prematurity and intrauterine growth retardation. Eur J Clin Nutr (2003) 57:266–272.[CrossRef][Web of Science][Medline]
Slattery MM, Morrison JJ. Preterm delivery. Lancet (2002) 360:1489–1497.[CrossRef][Web of Science][Medline]
Stata Reference Manual. Release 8.2. (2003) College station, TX: Stata Corp.
Steer P, Flint C. ABC of labour care: preterm labour and premature rupture of membranes. BMJ (1999) 318:1059–1062.
Thompson JMD, Irgens LM, Rasmussen S, Dalveit AK. Secular trends in socio-economic status and the implications for preterm birth. Paediatr Perinat Epidemiol (2006) 20:182–187.[CrossRef][Web of Science][Medline]
Tucker J, McGuire W. Epidemiology of preterm birth. BMJ (2004) 329:675–678.
Vause S, Johnston T. Management of preterm labour. Arch Dis Child Fetal Neonatal (2000) 83:79–85.
Wadhwa PD, Sandman CA, Porto M, Dunkel-Schetter C, Garite TJ. The association between prenatal stress and infant birthweight and gestational age at birth: a prospective investigation. Am J Obstet Gynecol (1993) 169:858–865.[Web of Science][Medline]
Wagner ME, Schubert HJP, Schubert DSP. Family size effects: a review. J Genet Psychol (1985) 146:65–78.[Web of Science][Medline]
Wingate MS, Alexander GR, Buekens P, Vahratian A. Comparison of gestational age classifications: date of last menstrual period vs. clinical estimate. Ann Epidemiol (2007) 17:425–430.[CrossRef][Web of Science][Medline]
Zou G. A modified Poisson regression approach to prospective studies with binary data. Am J Epidemiol (2004) 159:702–706.
Submitted on July 22, 2008; resubmitted on October 11, 2008; accepted on October 15, 2008.
CiteULike 
Connotea 
Del.icio.us What's this?
Related articles in Hum. Reprod.:
- Editor's Choice
- André Van Steirteghem
Hum. Reprod. 2009 24: 253.[Extract] [FREE Full Text]
This article has been cited by other articles:
|
|
 |
|
  M. R. Kramer and C. R. Hogue What Causes Racial Disparities in Very Preterm Birth? A Biosocial Perspective Epidemiol. Rev., November 1, 2009; 31(1): 84 - 98. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||