Childhood abuse and caesarean section among primiparous women in the Norwegian Mother and Child Cohort Study
M Lukasse, Oslo University Hospital, Rikshospitalet, Department of Obstetrics and Gynaecology, Sognsvannsveien 20, N-0027 Oslo, Norway. Emails Mirjam.firstname.lastname@example.org, Mirjam.Lukasse@Oslo-Universistetssykehus.no
Please cite this paper as: Lukasse M, Vangen S, Øian P, Schei B. Childhood abuse and caesarean section among primiparous women in the Norwegian Mother and Child Cohort Study. BJOG 2010;117:1153–1157.
We examined the association between a history of childhood abuse and caesarean section in the population-based Norwegian Mother and Child Cohort Study (MoBa). Our sample consisted of 26 923 primiparous women with singleton pregnancies at term. Of all women, 18.8% (5060) had experienced any childhood abuse, 14.3% (3856) reported emotional abuse, 5.2% (1413) reported physical abuse and 6.4% (1730) reported sexual abuse. The proportion of caesarean sections before labour was not affected by any childhood abuse. Any childhood abuse was associated with a slightly increased risk of caesarean sections during labour (adjusted odds ratio 1.16; 95% CI 1.03–1.30).
The observed increase in the caesarean section rates in many countries may have several causes, one of them being maternal request.1 A common reason for requesting birth by caesarean section is fear of childbirth.2 A clinical study of women referred to counselling because of their fear of childbirth and a request for planned caesarean birth, reported that 63% of these women had been subjected to abuse, without specifying the type or timing of the abuse.2
Reports on the duration of labour for childhood sexual abuse survivors present conflicting results depending partly on methodological aspects.3 Some suggest an association between a history of childhood sexual abuse and slow progress during labour.3 The mechanism behind this could be the impact of early life chronic stress, caused by childhood abuse, increasing stress proneness in adulthood. Stress may increase the production of stress hormones, such as adrenaline, noradrenaline and cortisol, which may affect uterine activity and cause slow progress. Slow progress in turn may lead to interventions during labour and, for some, to instrumental birth. A common indication for caesarean section during labour is failure to progress.1
Little research has been done on the possible association between a history of childhood abuse and caesarean section. We found only two studies that presented the proportion of caesarean section among women with a history of childhood abuse.4,5 These two studies reported contradicting results and neither of them included women representative for most European settings.
The aim of this study was to assess if there is an association between self-reported exposure to sexual, physical and emotional childhood abuse and birth by caesarean section.
Design and population
The Norwegian Mother and Child Cohort Study (MoBa) conducted by the Norwegian Institute of Public Health is a nationwide pregnancy cohort study started in 1999, with the aim of including 100 000 pregnant women by the end of 2009. The majority of all pregnant women in Norway are invited to participate, and the response rate is around 44%. Pregnant women are recruited through a postal invitation in connection with the routine ultrasound examination offered to all pregnant women in Norway at 17–18 weeks of gestation (http://www.fhi.no/morogbarn).
During pregnancy the women received three extensive questionnaires. Data from the questionnaires were linked to the Medical Birth Registry of Norway (MBRN), which keeps a record of all deliveries in Norway since 1967 (http://www.fhi.no/). Information in the registry is based on a standardised form completed by midwives shortly after delivery.
A woman’s previous birth greatly influences the mode of delivery for subsequent births. Because information on previous deliveries was not available to us, we only included primiparous women. The present study included 29 547 pregnancies and comprised primiparous women who returned both the baseline and the third questionnaire (at about 16–20 weeks and 30–34 weeks of gestation) and for whom data on delivery was available. We subsequently excluded 244 pregnancies of women who had not answered the questions on abuse. Multiple pregnancies, preterm birth (<37 weeks of gestation) and deliveries ≥42 weeks of gestation are associated with complications during labour and instrumental birth. To diminish the number of confounding variables we excluded 795 multiple pregnancies and 1585 women were excluded for not giving birth between 37 and 43 weeks of gestation. This left a total of 26 923 women for analyses. Participants in the study gave birth between January 2000 and December 2006. Informed consent was obtained from each participant. The Regional Committee for Medical Research Ethics and the Norwegian Data Inspectorate approved the study.
Caesarean section, coded as taking place before or during labour, was the outcome variable. Information about caesarean section was derived from the MBRN. The exposure variable of self-reported childhood abuse, a modified version of the Norvold Abuse Questionnaire (NorAq),6 came from the third MoBa questionnaire, at 30–34 weeks of gestation. Women were asked if they ever experienced any of the following: (1) has anyone over a longer period of time systematically tried to subdue, degrade or humiliate you? (mild emotional abuse); (2) has anyone threatened to hurt you or someone close to you? (moderate emotional abuse); (3) have you been subjected to physical abuse? (physical abuse); (4) have you been forced into sexual actions/acts? (sexual abuse). Women were given the opportunity to indicate if they experienced the abuse as a child (<18 years) and/or as an adult (>18 years) or never experienced the abuse (no, never). Women who answered yes to at least one of the four questions about childhood abuse (<18 years) were defined as having suffered from any childhood abuse. We considered that a preference to give birth by caesarean section could be a confounding factor in the association between childhood abuse and birth by elective caesarean section.1 This confounding variable was derived from the third questionnaire. Women who totally agreed with the statement ‘If I could choose, I would have a caesarean,’ were defined as preferring birth by caesarean. The following confounding factors for caesarean section were taken from the MBRN: suffering from any form of diabetes, pre-eclampsia, induction of labour, epidural analgesia, fetal distress and macrosomia. Macrosomia was defined as birthweight ≥4.5 kg. Socio-demographic characteristics and risk factors such as age, civil status, education, occupation, parity, height and pre-pregnancy weight were derived from the first questionnaire, at 16–20 weeks of gestation. BMI, and not height alone, was used as a possible confounding factor.
Frequency analyses were used to quantify the prevalence of the different types of childhood abuse. Cross-tabulation and Pearson’s chi-square test were used to study percentages and assess differences in demographic, medical and obstetric factors for women with physical, sexual and emotional abuse compared with women without childhood abuse.
The association of caesarean section before and during labour with the different types of childhood abuse, and combinations thereof, were analysed in multivariate logistic regression models. We adjusted for a priori selected potential confounding factors. All models included age, education, BMI, adult abuse, macrosomia and pre-eclampsia. We added preferring birth by caesarean section as a confounding factor in the models for caesarean section before labour. Epidural analgesia, induction and fetal distress were added in the models for caesarean section during labour. Women who gave birth by caesarean section before labour were excluded from the multivariate logistic regression analyses concerning caesarean section during labour. The reference group for all comparisons was women reporting no childhood abuse. All analyses were two-sided at α = 0.05. The statistical program SPSS version 15.0 was used to conduct all analyses.
The proportions of women reporting different types and combinations of childhood abuse are presented in Table 1. We found a substantial overlap between the different types of abuse (see Supplementary material: Figure S1). Twenty-nine percent of those reporting childhood abuse had been exposed to at least two types of abuse. Of those who experienced any childhood abuse 30.5% reported abuse as an adult. Women reporting childhood abuse were significantly younger, had completed less education, were less likely to be employed and fewer lived with their partner compared with women reporting no childhood abuse (see Supplementary material:Table S1). Obesity was significantly more frequent among women reporting childhood abuse (Table S1).
Table 1. Crude and adjusted odds ratios (OR) for caesarean section before and during labour according to childhood abuse category
|Any||5060 (18.8)||249||1.02 (0.88–1.17)||1.00 (0.86–1.17)||546||1.16 (1.05–1.28)||1.16 (1.03–1.30)|
|Mild emotional||3382 (12.6)||157||0.96 (0.80–1.13)||0.93 (0.77–1.12)||362||1.14 (1.01–1.29)||1.11 (0.97–1.28)|
|Severe emotional||1219 (4.5)||69||1.18 (0.92–1.51)||1.14 (0.86–1.52)||116||1.01 (0.83–1.23)||1.12 (0.89–1.40)|
|Any emotional||3856 (14.3)||186||1.00 (0.85–1.17)||0.96 (0.81–1.15)||402||1.11 (0.99–1.24)||1.11 (0.98–1.27)|
|Physical||1413 (5.2)||76||1.12 (0.88–1.42)||1.03 (0.79–1.35)||149||1.13 (0.95–1.35)||1.21 (0.99–1.48)|
|Sexual||1730 (6.4)||89||1.06 (0.85–1.33)||1.08 (0.85–1.38)||185||1.15 (0.98–1.35)||1.15 (0.95–1.38)|
|Physical and sexual||661 (2.5)||36||1.13 (0.80–1.60)||1.08 (0.74–1.59)||67||1.09 (0.84–1.40)||1.16 (0.86–1.56)|
|Physical and severe emotional||558 (2.2)||32||1.13 (0.79–1.62)||1.05 (0.70–1.58)||57||1.03 (0.78–1.36)||1.10 (0.80–1.51)|
|Sexual and severe emotional||401 (1.5)||28||1.47 (0.99–2.18)||1.34 (0.85–2.12)||35||0.94 (0.66–1.33)||0.90 (0.59–1.35)|
|Physical and sexual and severe emotional||289 (1.1)||19||1.38 (0.86–2.21)||1.26 (0.73–2.19)||30||1.13 (0.77–1.66)||1.16 (0.74–1.80)|
|Physical and sexual and any emotional||446 (1.6)||25||1.17 (0.78–1.75)||1.00 (0.63–1.61)||42||0.97 (0.71–1.34)||0.94 (0.65–1.36)|
The proportion with a preference for birth by caesarean section was 2.7% among women abused in childhood compared with 2.0% among women not exposed to childhood abuse (P = 0.001). The prevalence of pre-eclampsia was higher among women reporting childhood abuse, 5.0% compared with 4.2% among those not reporting childhood abuse (P = 0.013). More women abused in childhood were induced than those without childhood abuse, 15.2% compared with 13.7% (P = 0.006). There was no significant difference between women with and without a history of childhood abuse for the proportion augmented in labour, diagnosed with fetal distress, or giving birth to children with a birthweight ≥4.5 kg (macrosomia). Women abused in childhood received epidural analgesia more often (42.1%) than those not abused in childhood (38.5%, P < 0.001). The proportion of women having a spontaneous vaginal delivery was almost equal for the groups with and without childhood abuse (69%). However, the number of instrumental vaginal births, episiotomies and sphincter ruptures was significantly lower among women abused in childhood. None of the different categories of childhood abuse were significantly associated with a caesarean section before labour (Table 1). Two categories, any childhood abuse and mild emotional abuse, were associated with an increase of caesarean section during labour compared with women with no childhood abuse before adjustment for confounding factors (Table 1). After adjustment, only the category of any childhood abuse was significantly associated with an increase in caesarean section during labour (Table 1).
Our assumption that a history of childhood abuse is associated with birth by caesarean section was not confirmed for caesarean section before labour. The significant but small increase in the proportion of caesarean sections during labour did not affect the number of spontaneous vaginal births for women with any childhood abuse. Women with any childhood abuse were significantly less likely to have an instrumental vaginal delivery compared with women not abused in childhood.
Although women with a history of childhood abuse significantly more often expressed a preference for birth by caesarean section they did not give birth by caesarean sections before labour more often than women without a history of childhood abuse. This could be the result of counselling these women may have received during pregnancy.2 Many obstetric outpatient departments in Norway offer special services to women with fear of childbirth and a request for caesarean section.
Our results agree with a retrospective study of 400 out-of-hospital-planned births in the USA, which found that primiparous women abused in childhood had a significantly increased risk for caesarean section during labour compared with women with no history of childhood abuse.5 As in our study they included emotional, physical and sexual abuse, arguing that focusing exclusively on sexual abuse overlooks a substantial amount of damage to the psyche.5 Benedict et al.,4 investigating the association of sexual abuse only and selected pregnancy outcomes, found no significant difference between abused and comparison women for any labour and delivery characteristics. Over 70% of the women included in this study were African American and only a third of them were married or living with a partner. As in the study of Tallman and Hering5 the population included by Benedict et al.4 makes generalisation and comparison of the results difficult. The strength of our study is the large population-based sample.
A recent review on the impact of childhood sexual abuse on pregnancy and delivery found no studies indicating an increased risk for birth by caesarean section.3 When investigating the influence of childhood abuse on complications during pregnancy and labour, the focus has usually been on sexual abuse.3,4 This may be the reason why no significant associations have been found. If the association between childhood abuse and complications and interventions in childbirth is the result of neurobiological changes that cause altered reactions to stress, then all forms of childhood abuse should be investigated.
Our study showed a significant increase in caesarean sections during labour when all forms of childhood abuse were included. Besides the significant increase in caesarean sections during labour for women with any childhood abuse, our results are mainly negative (Table 1). A possible explanation could be that women who have suffered childhood abuse declined to participate in the study or answer the questions about abuse because of the social stigma associated with it, hence diluting the effect size. However, women when consenting to participate in the MoBa study would not have been aware that the study included questions on abuse as the study focuses on physical health before, during and after pregnancy. Of all the primiparous women who filled out questionnaires one and three, less than 1% were excluded from our sample for not answering the questions on abuse. This indicates that once women participate they also proceed to answer the questions about abuse. That still leaves the possibility of women under-reporting childhood abuse and so contributing to the mainly negative results in our study. As the study is anonymous and the childhood abuse is a matter of the past we deem it likely that women answered truthfully. Our study included no measure to assess if social desirability influenced women’s reporting of childhood abuse.
Participation in the MoBa study involved a considerable effort for women during pregnancy, the involvement of their partner and a commitment to providing information about their children for many years after birth. This may be one of the reasons why the response rate was low. A second possible reason is that pregnant women in Norway are often asked to participate in research studies. Third, the study offered no immediate benefits to the mother. It is certainly possible that some women who are burdened by their history of childhood abuse may not have the energy or desire to participate in the extensive MoBa study.
A limitation of our study is that the questions on abuse are not validated. The questions on emotional abuse are very similar to two of the three questions on emotional abuse used in the validated NorAq instrument.6 However, the questions on physical and sexual abuse are not. For each of the questions, whether the abuse is described or not, the abuse measured is subjective to the woman’s interpretation of both the questions and her own experiences. Substantially more women reported emotional childhood abuse than other kinds of abuse in our study. Quite likely it is easier to report emotional abuse than physical and sexual abuse. Another reason could be that there were two detailed questions describing emotional abuse compared with a single question assessing physical abuse and a single question asking about sexual abuse.
The use of the word abuse in the question about physical abuse may result in lower reporting of physical abuse compared with questions describing the abusive experience. It seems likely that only women who have experienced severe physical abuse will answer our question positively. If so, our results show the effect of severe physical abuse on women’s self-reported health in pregnancy. The original NorAq instrument6 has four questions on sexual abuse whereas our study has only one. The question used in our study allows women to define both ‘forced’ and ‘sexual actions/acts’. Some cases of sexual abuse will not be identified by this question. It is uncertain how this affects the observed associations.
Taking methodological considerations into account the prevalence of the emotional abuse compares well with other studies, whereas the prevalence of physical and sexual abuse is slightly low. A detailed comparison of our prevalences with other studies has been given in a previous publication.7
Women were asked to recall childhood experiences before age 18 years when they were on average 28 years old. The retrospective reporting of abuse could be subject to recall bias. Our study did not include questions about the onset, length of time and frequency of the abuse, nor did it include information about other adverse childhood exposures. The influence of these and other unknown factors on the results is unknown.
Further studies investigating the association between a history of childhood abuse and complication and interventions during childbirth are necessary. Large, prospective cohort studies in unselected populations, with a high response rate, using a validated instrument to measure adverse childhood experiences, are needed. However, most studies so far,3,4 ours included, suggest that a history of any childhood abuse is not significantly associated with an increased risk for caesarean section before labour. Although the number of caesarean sections during labour was significantly increased among women with any childhood abuse, it did not affect the number of spontaneous vaginal deliveries, because the number of instrumental vaginal births was significantly reduced. To experience a caesarean section during labour may be less traumatic emotionally, especially for childhood sexually abused women, than experiencing an instrumental vaginal birth. This study adds to the so far limited body of knowledge about the effect of a maternal history of childhood abuse on mode of delivery.
Disclosure of interests
Contribution to authorship
M Lukasse contributed to the idea and design of the study, the analysis and interpretation of data. S Vangen and P Øian contributed to the interpretation of data. B Schei: contributed to the idea and design of the study. All authors contributed to the interpretation of data, writing of the paper and approval of the final version.
Details of ethics approval
The study was approved by the Regional Committee for Ethics in Medical Research, date: 9 October 1998, reference number: S-97045/S-95113.
The Norwegian Mother and Child Cohort Study is supported by the Norwegian Ministry of Health, Oslo, NIH/NIEHS (grant no N01-ES-85433), NIH/NINDS (grant no.1 UO1 NS 047537-01), and the Norwegian Research Council/FUGE, Oslo (grant no. 151918/S10).
The first author is supported by The Norwegian Women’s Public Health Association, Oslo, for her PhD.
The efforts of the staff at the MoBa and the MBRN are greatly appreciated, and so is the contribution from all women participating in the study.