Fear of childbirth and risk for birth complications in nulliparous women in the Danish National Birth Cohort

Abstract

Objectives  To examine the associations between fear of childbirth and emergency caesarean section and between fear of childbirth and dystocia or protracted labour and fetal distress.

Design  Prospective cohort study.

Setting  Danish National Birth Cohort.

Population  A total of 25 297 healthy nulliparous women in spontaneous labour with a single fetus in cephalic presentation at term following an uncomplicated pregnancy.

Methods  Data were collected during 1997–2003 from computer-assisted telephone interviews twice in pregnancy linked with national health registers.

Main outcome measures  Risk for emergency caesarean section of women who feared childbirth; risk for dystocia/protracted labour or fetal distress of women who feared childbirth.

Results  Fear of childbirth in early (16 weeks, 6 ± 29 days) and late (31 weeks, 4 ± 21 days) pregnancy was associated with emergency caesarean section: OR, 1.23 (1.05–1.47) and 1.32 (1.13–1.55), respectively. When fear of childbirth was expressed at both interviews, the OR was 1.43 (1.13–1.80). Women who feared childbirth had an increased risk for dystocia or protracted labour (OR, 1.33; 1.15–1.54), but not for fetal distress (OR, 0.94; 0.72–1.23).

Conclusions  Fear of childbirth during pregnancy was associated with dystocia and emergency caesarean section but not with fetal distress.

Introduction

For decades, it has been hypothesised that maternal anxiety or fear of childbirth can prolong labour.^1,2 In countries where women are left alone during labour, continuous support by a lay person can significantly reduce birth complications, including emergency caesarean section.³ This finding may indicate an association between a feeling of security and reduced anxiety and a reduced risk for emergency caesarean section.

Studies of the effects of maternal psychological distress and fear of childbirth have provided little information,^4,5 as most were small, in some women were asked about fear of childbirth retrospectively, adding a risk for recall bias, and in all studies a composite outcome, ‘adverse obstetric outcome’, was used, covering haemorrhage in pregnancy, caesarean section or problems with breastfeeding. Of the few studies that partly addressed these limitations,^6–9 only a case–control study within a prospective cohort study found a significantly increased risk for emergency caesarean section,⁷ whereas the cohort studies did not.^6,8,9 An association between general anxiety or depression and emergency caesarean section was recently examined, but the studies had divergent results.^7–13 None of these studies included only nulliparous women, although women with a previous caesarean section might more often fear childbirth^14–16 and have increased risks for complicated labour.^17,18

We tested the associations between fear of childbirth and emergency caesarean section, dystocia or protracted labour, and fetal distress in the absence of dystocia or protracted labour in a large cohort of healthy, nulliparous singleton pregnant women without pregnancy complications and spontaneous onset labour beyond 37 weeks of gestation. To minimise confounding further, we included only women with no medical or obstetric risk factors.

Methods

We obtained data from the Danish National Birth Cohort, a nationwide cohort of 100 000 pregnant women.¹⁹ The general practitioner approached the woman at the first antenatal visit which usually occurs between weeks 6 and 12. All general practitioners were asked to participate in recruitment but many women were not approached as about half of the practitioners chose not to participate, mainly because of dissatisfaction with the size of the payment. Only women fluent in Danish and women who expected to carry the pregnancy to term were included. Half of eligible and invited women chose to participate, and it is estimated that 31% of the pregnant women during the recruitment period 1997–2003 participated.²⁰ The women had to mail a signed consent to the secretary of ‘the Danish National Birth Cohort’. The women were interviewed by computer-assisted telephone interviews by trained interviewers twice during pregnancy. If participants were not reached contact was tried three additional times. For the present study, we used information from interview I, conducted on average at 16 weeks and 6 days (±29 days) of gestation, and from interview II, on average at 31 weeks, and 4 days (±21 days) of gestation. Data were linked to the Medical Birth Register²¹ and the National Patient Register²² by the unique personal identification number in the civil registration system. Data are entered by the midwives and obstetricians. The outcome used was emergency caesarean section during labour, the diagnosis of dystocia/protracted labour and suspected fetal distress at any time during labour. Fear of childbirth was measured from answers to the question: ‘Are you anxious about the course of the upcoming delivery?’ 1 ‘Not at all’, 2 ‘Yes, a little’ or 3 ‘Yes, a lot’. Only answer 3 was considered fear of childbirth. The question was asked at both interviews. Other items from interview I concerned education or job, students were assigned the level of expected education, persons unemployed for <6 months were assigned the level at the most recent job, demographic information, obstetric history, height, pre-pregnancy weight, smoking and physical activity; from interview II, information about anxiety or depression and weight gain during pregnancy was collected. Anxiety was assessed from answers to the question: ‘During your pregnancy, have you been anxious and afraid for no reason?’, and symptoms of depression were assessed by answers to the question: ‘Have you felt down and blue during your pregnancy?’ Diagnoses of medical conditions and pregnancy complications and mode of delivery were derived from the National Patient Register by ICD-10 codes. The Medical Birth Register was consulted for birthweight, head circumference, induction of labour and duration of pregnancy.

A total of 40 156 women had given birth to a live child. We excluded women with a medical condition (haematological diseases, endocrinological diseases, circulation diseases, respiratory diseases, neurological diseases, gastrointestinal diseases and nephrological diseases), substance abuse or malformed uterus or other conditions that might interfere with a vaginal delivery (n = 1513; 3.8%); women who had any pregnancy complication (n = 3698; 9.2%) and women whose fetus was diagnosed with or suspected of having any abnormality (n = 1577; 3.9%). We also excluded 1145 (2.9%) women who delivered preterm (<37 weeks); 1590 (4%) women with a breech or transverse lie; those delivered by a planned caesarean section, or an emergency caesarean section for causes other than birth complications and women whose labour was induced (2780, 6.9%); 783 (2%) women who in addition reported a medical condition or pregnancy complication at one of the interviews and 1612 (4%) women who did not respond at interview II. In total, 25 458 women remained in the cohort. For the 419 (1%) women with missing values for height or weight, for whom the pre-pregnancy body mass index (BMI) could not be calculated, we made a ‘missing’ category, as exclusion of these women could bias the results. Values for other variables were missing for 161 (0.4%) women, who were excluded, leaving 25 297 (63%) women for the analyses.

Statistical analyses

We conducted logistic regression analyses with emergency caesarean section as the outcome and fear of childbirth at interview I, II or both interviews as the dependent variable. The analyses were conducted as univariate and then adjusted for educational level, maternal age, height, pre-pregnancy BMI, residential area, size of residence and smoking status, which were possible confounders found to be associated with fear of childbirth in a previous study of the cohort,²³ and also for weight gain, birthweight, head circumference and duration of pregnancy, which were found to be important variables for emergency caesarean section in the same cohort.²⁴ We then performed logistic regression analyses with dystocia or protracted labour and suspected fetal distress as the outcomes, univariate and adjusted for the same variables as above mentioned. For all three outcomes we checked for interactions but there was none.

Results

Younger women with short education and either a high or a low BMI, who gained high weight during pregnancy, were short and smoked more often expressed fear of childbirth (Table 1). The emergency caesarean section rate was 9.1%. Table 2 shows that fear of childbirth was associated with emergency caesarean section in intended vaginal deliveries with spontaneous onset labour by univariate analysis and when adjusted for all potential confounders. The association was strongest for women who expressed fear at both interviews (adjusted OR, 1.43; 95% CI, 1.13–1.80), and the association was stronger when fear was reported at interview II (adjusted OR, 1.32; 95% CI, 1.13–1.55) than at interview I (adjusted OR, 1.23; 95% CI, 1.05–1.47). Fear of childbirth was associated with a diagnosis of dystocia (adjusted OR, 1.33; 95% CI, 1.15–1.54) but not with suspected fetal distress in the absence of dystocia or protracted labour (adjusted OR, 0.94; 95% CI, 0.72–1.23) (Table 3).

Table 1.   Characteristics of healthy nulliparous women in the Danish National Birth Cohort

Characteristics	Total	Expressed fear at interview I or II	Expressed fear at interviews I and II
No. (%) women	25.297	2144 (8.5%)	812 (3.2%)
Maternal age (years)
<20	479 (1.9)	91 (4.2)	33 (4.1)
20–24	5.083 (20.1)	481 (22.4)	184 (22.7)
25–29	13.267 (52.4)	1021 (47.6)	376 (46.3)
30–34	5.378 (21.3)	464 (21.6)	171 (21.1)
≥35	1.090 (4.3)	87 (4.1)	48 (5.9)
Educational level, years of education
University, ≥17	5.446 (21.5)	342 (16.0)	121 (14.9)
BA, 14–16	8.691 (34.4)	667 (31.1)	226 (27.8)
Vocational training, 11–13	9.168 (36.2)	890 (41.5)	368 (45.3)
Unskilled, 9–12	1.268 (5.0)	161 (7.5)	65 (8.0)
Unemployed >6 months or in job but unspecified educational level	724 (2.9)	84 (3.9)	32 (3.9)
Pre-pregnancy body mass index
<18.5	1.115 (4.4)	110 (5.1)	43 (5.3)
18.5–24.9	18.059 (71.4)	1451 (67.6)	557 (68.6)
25–29.9	4.219 (16.7)	388 (18.1)	135 (16.6)
≥30	1.485 (5.9)	150 (7.0)	55 (6.8)
Missing	419 (1.7)	45 (2.1)	22 (2.7)
Weight gain in pregnancy (kg/week)
≤0.4	19.275 (76.2)	1514 (70.6)	574 (70.7)
>0.4	6.022 (23.8)	630 (29.4)	238 (29.3)
Height (cm)
≤160	2.320 (9.2)	230 (10.7)	89 (11.0)
161–165	5.211 (20.6)	469 (21.9)	172 (21.2)
166–170	8.207 (32.4)	688 (32.1)	263 (32.4)
171–175	6.009 (23.8)	480 (22.4)	186 (22.9)
>175	3.550 (14.0)	277 (12.9)	102 (12.6)
Smoking status
Nonsmoker	18.465 (73.0)	1411 (65.8)	590 (60.3)
Occasionally or stopped	3.557 (14.1)	315 (14.7)	141 (17.4)
Smoker	3.275 (13.0)	418 (19.5)	181 (22.3)

Table 2.   Risk for emergency caesarean section versus vaginal birth in 25 297 women in the Danish National Birth Cohort in relation to fear of childbirth

Fear of childbirth	Total no. births	Emergency caesarean section (%)	Crude OR	Adjusted OR*
*Adjusted for age, height, pre-pregnancy body mass index, pregnancy weight gain, educational level, residential area, size of residence, smoking status, birthweight, head circumference, duration of pregnancy, symptoms of anxiety and depression.
Interview I
No fear	23.400	2081 (8.9)	1	1
Fear	1.897	209 (11.0)	1.27 (1.09–1.48)	1.23 (1.05–1.47)
Interview II
No fear	23.426	2074 (8.9)	1	1
Fear	1.871	216 (11.5)	1.34 (1.16–1.56)	1.32 (1.13–1.55)
Interviews I and II
No fear	22.341	1963 (8.8)	1	1
At I or II	2.144	229 (10.7)	1.24 (1.07–1.44)	1.20 (1.03–1.40)
At I and II	812	98 (12.1)	1.43 (1.15–1.77)	1.43 (1.13–1.80)

Table 3.   Risk for fetal distress or dystocia/protracted labour in relation to fear of childbirth in 25 297 women in the Danish National Birth Cohort

Fear of childbirth	No. (%) women with fetal distress*	Crude OR	Adjusted OR**
*Without dystocia/protracted labour. **Adjusted for age, height, pre-pregnancy body mass index, pregnancy weight gain, educational level, residential area, size of residence, smoking status, birthweight, head circumference, duration of pregnancy, symptoms of anxiety and depression.
No	1862 (8.3)	1	1
At interview I or II	177 (8.3)	1.01 (0.90–1.26)	0.99 (0.84–1.17)
At interviews I and II	65 (8.0)	1.09 (0.92–1.29)	0.94 (0.72–1.23)
Fear of childbirth	No. (%) women with dystocia/protracted labour	Crude OR	Adjusted OR**
No	9368 (41.9)	1	1
At interview I or II	968 (45.2)	1.15 (1.04–1.25)	1.11 (1.01–1.22)
At interviews I and II	403 (49.6)	1.37 (1.19–1.57)	1.33 (1.15–1.54)

Discussion

We found that fear of childbirth was associated with emergency caesarean section. The association was strongest for women who reported fear at the interview closest to the time of the birth and stronger for those who reported fear twice during pregnancy, indicating the robustness of our findings. These results are in accordance with those of a Swedish case–control study⁷ of 97 cases and 194 controls matched for age and parity in a prospective cohort, with an OR of 3.0 for emergency caesarean section for women who expressed fear of childbirth; however, the control and case groups differed with regard to a number of important confounding variables. A study of 396 women in the UK showed no association between fear of childbirth and emergency caesarean section.⁸ A Norwegian study⁹ of 1321 women and a Swedish cohort study⁶ of 2662 women showed nonsignificant associations between fear of childbirth and emergency caesarean section. All four studies included parous women, some of whom had previously had a caesarean section, and nulliparous women with medical and obstetric risk factors, and no adjustment was made for weight, height or weight gain, which are associated with fear of childbirth and with emergency caesarean section.^23,24 We would have expected the studies to have found stronger associations than ours, but the sample sizes of two of them were possibly too small to detect a true association.^6,9 This study was based on a cohort of Danish women. Possible cultural differences in the perception of fear could have influenced to different results between our and other populations.

Fear and anxiety activate a hormonal stress response in pregnant and labouring women,^25–28 which can result in dystocia or protracted labour.^25,29,30 It has been hypothesised that hyperactive contractions and fetal distress are a consequence of maternal anxiety,⁷ and some evidence exists that maternal anxiety influences fetal circulation, which could result in hypoxia.^27,31–34 Our finding that fear of childbirth was associated with a diagnosis of dystocia but not with suspected fetal distress in the absence of dystocia or protracted labour does not correspond to the meagre evidence that release of stress hormones can cause impaired fetal circulation.^27,31–34 Only one of these studies addressed fetal distress during labour.

One of the advantages of our study was the large sample size, which allowed us to include only nulliparous women and to exclude women with medical and obstetric risk factors that could act as confounders. We also eliminated the risk of type II errors, which may have accounted for the lack of association in other studies.^6,9 We adjusted for all potential confounders identified in previous studies of the same cohort. One limitation of the study is that fear of childbirth was measured from answers to a single question, which did not take into account different aspects of this fear; however, an imprecise measure would result in a weaker association than the true one. Preferably, fear of childbirth could have been measured closer to the birth or even at admission but unfortunately this was not possible. We believe that the emergency caesarean section rate of 9.1% in this study is comparable with the 10.8% emergency caesarean section rate in the total population of first-time mothers in Denmark with spontaneous onset labour at term with cephalic presentation 1997–2003 that in addition included births in women with pregnancy complications.¹⁷ Participants were more often nonsmokers, older than 25 years and with a BMI between 18.5 and 24.9.²⁰ A study of the possible selection bias because of this and a low participation rate of 31% found that for three associations between exposure—for which there was a selection into the cohort—and an outcome were tested and the associations were not affected.²⁰

We found previously that the women who dropped out of the study more often feared childbirth than those who continued.²³ If women with fear who dropped out had a higher risk of emergency caesarean section than those with fear who were included, the found association between fear of childbirth and emergency caesarean section in the study has been underestimated.

Having fear of childbirth in pregnancy was associated with an increased risk of emergency caesarean section and protracted labour/dystocia. Hodnett found that continuous support during labour could reduce the number of emergency caesarean section under certain circumstances,³ We did not test this in our cohort but as women who feel supported during labour benefit through positive birth experience and future psychologically wellbeing it is advisable to offer optimal support.^2,35,36

Conclusion

Fear of childbirth was associated with emergency caesarean section. We also found that fear of childbirth was associated with dystocia or protracted labour but not with fetal distress.

Conflicts of interests

None.

Contribution to authorship

ML planed the project, MH and CJ advised in the design, ML did the data management, the statistical analyses and drafted the paper. MH and CJ contributed to the writing of the paper.

Ethical approval

The Danish Data Protection Agency approved the project (No 2004-41-4747) and the Danish national birth cohort steering committee granted authorisation for the use of data from the cohort (No 2004–13).

Funding

ML was funded by The Health Foundation, The Danish Midwife Association, The Lundbeck Foundation, H:S Central Research Fund, The Aase and Ejnar Danielsen Foundation and the Linux Foundation. The Danish National Research Foundation has established the Danish Epidemiology Science Centre that initiated and created the Danish national birth cohort. The cohort is furthermore a result of a major grant from this Foundation. Additional support for The Danish National Birth Cohort is obtained from the Pharmacy Foundation, The Egmont Foundation, the March of Dimes Births Defects Foundation, the Augustinus Foundation and The Health Foundation.

Acknowledgements

We thank Visti Birk Larsen for help with data management.