Birth outcomes in primiparous women who were raped as adults: a matched controlled study

Authors


L Halvorsen, Department of Obstetrics and Gynaecology, University Hospital of North Norway N-9038 Tromsø, Norway. Email lotta.halvorsen@unn.no

Abstract

Please cite this paper as: Nerum H, Halvorsen L, Øian P, Sørlie T, Straume B, Blix E. Birth outcomes in primiparous women who were raped as adults: a matched controlled study. BJOG 2010;117:288–294.

Objective  To compare the duration of labour and the birth outcome in a group of primiparous women who had been raped after the age of 16, with a control group from the same birth cohort.

Design  Cohort study.

Setting  University Hospital of North Norway.

Sample  Fifty women raped as adults and 150 controls.

Methods  Data about birth outcomes in the first pregnancy were collected from the patient files and data concerning the assault were obtained in a subsequent pregnancy through consultations with the women who had been raped. Birth outcomes in the group of women who had been raped were compared with matched controls using a multivariable logistic regression model.

Main outcome measures  Caesarean section, operative vaginal delivery and duration of labour.

Results  During their first delivery, the women who had been raped had an increased risk for caesarean section (adjusted odds ratio 15.7, 95% CI 5.0–49.1) and for assisted vaginal delivery (adjusted odds ratio 13.1, 95% CI 4.9–34.5) when compared with controls. The group of women who had been raped had a longer second stage of labour than the control group (120 versus 55 minutes, < 0.01). They were more often single mothers, unemployed and smokers, and had a higher body mass index and more previous pregnancy terminations and miscarriages than the control group.

Conclusions  The women who had been raped had a longer second stage of labour, and an increased risk of caesarean section and operative vaginal delivery compared with controls from the general birth cohort. These findings indicate that the consequences for delivery for women who had been raped as adults could be specific and may warrant particular attention. The birth experience of women who had been raped should also be illuminated in future studies.

Introduction

Rape is one of the most traumatising and violent sexual offences a person can be subjected to. It leads to pervasive negative health effects and a great reduction in quality of life.1–6 Various health problems such as post-traumatic stress disorder, depression, anxiety, eating disorders, substance abuse, chronic pain such as headache, fibromyalgia, abdominal and pelvic pain and sexual dysfunction have been described following physical, emotional and sexual abuse.4,7–19 Pregnant women with a history of abuse often express their prior experience through fear of birth or through a desire to be delivered by caesarean section.20–22

Various potentially confounding factors make the interpretation of results difficult. The occurrence of sexual abuse is generally evaluated on the basis of subjective recall on the part of the victims. It is well known that denial and minimisation on the one hand, and exaggeration on the other hand, may occur.4 Studies based on different populations and different definitions of sexual assault indicate that 3–33% of women have been subjected to forced sexual activity or sexual assault.3,7,10,19,23–25 The few studies that have examined a possible association between sexual assault and birth outcome show equivocal results.14,15,18,21,24 The studies are based on different populations and inclusion criteria which make the findings difficult to compare. The studies make little distinction between the various forms of physical and sexual abuse, when in life the abuse occurred, or parity when the birth outcome is described.

The background for the current study was our clinical experience that women who were known to have been raped as adults relatively frequently had a prolonged second stage of labour and were more frequently delivered by forceps, ventouse or caesarean section. We are not aware of other studies that describe birth outcome in women subjected to rape in adulthood, without a known history of sexual abuse in childhood.

The aim of the study was to compare birth outcomes and course of labour during the first delivery of women who were raped after the age of 16, with a control group from the same birth cohort.

Material and methods

The women who had been raped as adults were recruited from a cohort of 808 pregnant women referred to counselling by the mental health team at the antenatal clinic at University Hospital of North Norway between 2000 and 2007. The women were referred in a subsequent pregnancy for various psychological burdens such as having a previous traumatic birth experience, fear of birth, depression or a maternal request for caesarean delivery. Inclusion criteria were that the woman had been raped at or above the age of 16 years (the age of sexual consent in Norway), and before giving birth the first time, and that her first birth took place at the University Hospital. Exclusion criteria were that the woman had been subject to sexual abuse before the age of 16 years, that the rape had occurred in connection with substance abuse or prostitution, or that her rape history was known to her birth attendants. Fifty women who had been raped were included and nine were excluded (Figure 1).

Figure 1.

 Flow chart of inclusion in the study.

The counselling approach, which has been described in detail previously,20,22,26 was based on the establishment of good rapport and trust, attention to the woman’s reproductive health such as age at sexual debut, previous sexual assault, close relationships and family network, and previous encounters with health-care personnel. All forms of abuse were noted with respect to form, timing, age, perpetrator and how the abuse had taken place. Previous birth experience was the starting point for dialogue and information gathering. This crisis-oriented approach to fear of birth is based on the notion that abuse and rape are shame-laden and unexpressed experiences that the individual woman will not herself bring up without prompting from the midwife.

Our experiences from the antenatal clinic indicated that the prevalence of both caesarean section and operative vaginal delivery among women who had been raped was about 40%, compared with 12–17% in primiparous women as a whole at this hospital. To show a statistically significant difference between 40 and 17%, we needed 38 women in each group (α = 0.05 with 90% power). The control women were drawn from the birth cohort of primiparous women at our hospital, matched for age, year of first childbirth and presentation (cephalic and breech). As there is no information about the extent to which the women in the control population had been subject to rape, three women were chosen as controls for each woman who had been raped: the two women fulfilling the inclusion criteria who had given birth just prior to the woman who had been raped and the first after that particular woman.

The first childbirth of each woman who had been raped was identified through the birth log on the ward. Other data were collected from the women’s labour records and the electronic journal system Partus (Clinsoft®; Clinical Software, Oslo, Norway), where the following variables were registered: age, marital status, education, employment status, smoking habits, pre-pregnancy body mass index (BMI), prenatal visits, number of previous miscarriages and pregnancy terminations. High obstetric risk was also noted and defined as having: a chronic somatic illness of significance to the pregnancy, or complications such as diabetes, pre-eclampsia, polyhydramnios, oligohydramnios, intrauterine growth restriction, gestation lasting more than 42 weeks, prelabour rupture of membrane without labour after more than 24 hours and meconium staining in the fluid; whether the labour began spontaneously or was induced was registered along with the use of oxytocin augmentation in labour, epidural analgesia, operative vaginal delivery, caesarean section, episiotomy and blood loss during birth.

The active part of the first stage was defined as starting at 4 cm dilation and lasting until dilation was complete. The second stage was defined as the period from complete dilation until the child was born. The baby’s birthweight, 5-minute Apgar score and admission to the neonatal intensive care unit were also registered.

Statistical analyses

The group of women who had been raped and the control women were compared by calculating the chi-square test or Mann–Whitney U test as indicated. To evaluate differences in risk of caesarean or operative vaginal delivery, a multinomial logistic regression analysis was performed, adjusted for variables that were significantly associated with mode of delivery in univariate analyses (age, marital status, employment status, BMI, previous abortions, obstetric risk level, oxytocin augmentation, epidural analgesia and birthweight). Risks are reported as odds ratio (OR) and 95% confidence intervals (95% CI). The data were analysed with SPSS 16.0 (SPSS Inc Chicago, IL, USA) and STATA 10.1 (StataCorp LP, College Station, TX, USA).

Results

In all, 50 women who had been subject to rape before giving birth to their first child and who fulfilled the inclusion criteria and 150 controls from the total birth cohort were included in the study (Figure 1). None of the 50 women gave birth as a result of rape, and all were referred to our clinic during a subsequent pregnancy. There was no indication from the records that any of the women had expressed fear of birth or requested caesarean section in their first pregnancies. Among control women, fear of birth was described in the records of two women.

The average age of the women at the time of the rape was 18.5 years (range 16–30 years) and for 27 of them (54%) the rape was their sexual debut. In 25 women (50%) the rapist was a stranger. Socio-demographic characteristics, prenatal visits and birth outcomes in the two groups are described in Table 1. The group of women who had been raped had a longer second stage of labour than the control group of women (120 versus 55 minutes, < 0.01), there was no significant difference in duration of the first stage of labour (290 versus 250 minutes, = 0.87) (Table 2).

Table 1.   Characteristics of 50 primiparous women, raped as adults compared with 150 control women matched for age, parity, fetal presentation and year of birth
FactorCases n = 50Controls n = 150χ2/MW U testP
  1. BMI, body mass index; MW U test, Mann–Whitney U test; NICU, neonatal intensive-care unit.

  2. Data are either number (%) or median (range).

  3. *n = 46 cases and 142 controls (elective caesarean excluded).

Single (unsupported) (%)16 (32)15 (10)13.858<0.01
Higher education (%)23 (46)77 (51)0.4270.5
Not employed (%)14 (28)11 (7)14.644<0.01
Smokers (%)18 (36)31 (21)4.7660.03
Pre-pregnant BMI23.1 (17–45)22.0 (17–37.5)1.2240.22
Pregnancy termination and miscarriage (%)22 (44)37 (25)6.7400.01
Prenatal visits12 (5–17)12 (2–25)0.3370.74
High obstetric risk (%)19 (38)40 (27)2.3160.13
Labour induced or augmented with oxytocin* (%)37 (80)79 (56)9.044<0.01
Epidural analgesia* (%)27 (59)45 (32)10.723<0.01
Caesarean section (%)20 (40)23 (15)13.519<0.01
Ventouse/forceps (%)20 (40)17 (11)20.439<0.01
Apgar score < 7/5 min (%)1 (2)3 (2)0.0001.0
Blood loss >500 ml (%)8 (16)20 (13)0.2210.64
Birthweight (g)3450 (896–4780)3510 (2170–4700)0.1230.90
NICU admission (%)9 (18)8 (5)7.7360.01
Table 2.   Course and outcome of labour in the women who had been raped as adults and the control women who gave birth vaginally to their first child
Vaginal birthsCases n = 30Controls n = 127χ2/MW U testP
  1. MW U test, Mann–Whitney U test.

  2. Data are given as number (%) or median (range).

Induced labour (%)5 (17)13 (10)0.9890.79
High obstetric risk (%)7 (23)27 (21)0.6010.80
Oxytocin augmentation (%)24 (80)70 (55)6.2530.01
Epidural analgesia (%)17 (57)39 (31)7.1260.01
Episiotomy (%)15 (50)43 (34)2.7140.10
Ventouse/forceps (%)20 (67)17 (13)38.246<0.01
Duration of first stage in minutes290 (90–720)250 (60–780)0.1650.87
Duration of second stage in minutes120 (10–420)55 (5–290)4.221<0.01

Birth outcome and indications for operative delivery are illustrated in Figure 2. The crude risks of operative vaginal delivery and caesarean section were 12.9 (5.2–32.3) and 9.6 (4.0–23.1), respectively. The results of the multinomial logistic regression analyses are given in Table 3. None of the adjustment variables in the model were significant for operative vaginal delivery, the risk being 13.1 (4.9–34.5) for the women who had been raped compared with controls. The corresponding risk for a caesarean section in women who had been raped was 15.7 (5.0–49.1). Maternal age and BMI added small contributions to the probability of a caesarean section, whereas increased obstetric risk substantially increased the caesarean section risk. Oxytocin augmentation lowered the likelihood of a caesarean section.

Figure 2.

 Flow chart for birth outcome and indication for operative delivery, 50 primiparous women with a history of rape in adulthood and 150 women matched for age, parity, year of first childbirth and fetal presentation.

Table 3.   Risk for operative vaginal delivery and caesarean section in 50 women who had been raped as adults and 150 women acting as controls
Women who had been raped versus control womenOperative vaginal delivery OR (95% CI)PCaesarean section OR (95% CI)P
  1. BMI, body mass index; OR, odds ratio.

  2. Multinomial logistic regression analysis adjusted for age, marital status, employment status, body mass index, previous pregnancy termination and miscarriage, obstetric risk level, oxytocin augmentation and epidural analgesia.

Unadjusted12.9 (5.2–32.3)<0.019.6 (4.0–23.1)<0.01
Final model
Adjusted13.1 (4.9–34.5)<0.0115.7 (5.0–49.1)<0.01
Age (years)1.06 (0.98–1.16)0.141.14 (1.04–1.25)<0.01
BMI (kg/m2)0.97 (0.86–1.08)0.551.15 (1.04–1.28)<0.01
High obstetric risk (high/low)1.36 (0.49–3.71)0.5511.93 (4.4–32.3)<0.01
Oxytocin augmentation (yes/no)1.74 (0.66–4.61)0.270.15 (0.05–0.43)<0.01

Discussion

In this study population, women who had experienced being raped in adulthood, before they gave birth for the first time, had a 13-fold increase in risk of caesarean delivery and a ten-fold increase in risk of operative vaginal delivery than a control group of women taken from the total birth cohort in our hospital. After adjustment for covariates the odds ratios increased to 15.7 and 13.1, respectively. The wide confidence limits reflect the small number of women, but the P-value of the corresponding tests leaves little doubt about the importance of the findings (Table 3). The duration of the second stage of labour was significantly longer in women who had been raped than in the control group of women (120 versus 55 minutes).

The longer second stage of labour in the women who were raped may be one explanation for the high proportion of operative vaginal deliveries, and the prolongation could be the result of a reactivation of the previous rape trauma. Reactivation of previous traumas during labour is described in the literature and in studies about women who were sexually abused as children.27,28 The violating, invasive nature of the rape trauma, with painful penetration of the woman’s vagina during the rape, and the baby’s descent into the vagina during the second stage involve the same anatomical structures, albeit in different ways, and may contribute to a reactivation of the trauma.

It is possible that procedures and measures that are part of routine birth care may unintentionally contribute both to this reactivation and to the prolonged second stage. Normal procedure at the hospital is to call an obstetrician after 1 hour of active pushing without progress. This most often leads to management involving more active intervention from then on, which may make it even more difficult for the woman to give birth. Swahnberg et al.29 have described how women undergoing gynaecological treatments who have a history of sexual abuse or assault risked having a new experience of assault in their encounters with the health services.

The study population consisted of a selected group of women who had been raped as adults before giving birth for the first time, and was without any known history of sexual abuse during childhood. They were selected from women referred with psychological burdens such as fear of birth and/or request for caesarean section in the absence of a medical indication referred in a subsequent pregnancy, and do not represent all women who had been raped who gave birth in the same period. The validity of the results must be interpreted in the context of the study population and cannot be generalised to hold for all women who have been raped. There may be limitations in the methods used to gain an overview of the abuse, and we do not have information about possible sexual abuse or assault among the women in the control group. However, abuse in the control group would probably have contributed to a lessening of the demonstrated differences between the groups.

The women who had been raped were more often single and without paid employment. More of them smoked, their pre-pregnant BMI was higher and they had more frequently had pregnancy termination or miscarriages than the controls, features that have been found in other studies.18,24 The model presented in Table 3 is a result of numerous analyses in which different possible confounders have been included such as birthweight, marital status, level of education, employment status, smoking habits, epidural analgesia, pregnancy terminations and miscarriages. The effect of being raped is consistent through all models.

Previous studies have not been able to document a corresponding effect on birth outcomes in women subjected to sexual assault as adults.14,15,18,21. Van der Hulst et al.18 examined the birth outcomes of low-risk pregnant women who had been subject to sexual abuse both in childhood and as adults and found that they had the same chance of having a normal birth as a control group who had not reported a history of sexual abuse. Tiwari et al.15 looked at intimate partner violence during pregnancy and the preceding year in a group of primiparous women and found no significant differences in birth outcome between those subjected to abuse and those who were not. Stenson et al.24 studied the prevalence of sexual abuse in both childhood and adulthood in an unselected pregnant population, and did not find differences in birth outcome. Heimstad et al.21 showed that half of the women who had been subjected to sexual or physical abuse as children had an uncomplicated vaginal birth, compared with 75% of those who had not had such an experience. One possible explanation for the high proportion of caesarean sections and operative vaginal deliveries in the present study may therefore be that our sample consists only of women with no known history of sexual abuse in their childhoods, who had been raped as adults before giving birth for the first time and in a subsequent pregnancy were referred to the mental health team for different psychological problems.

The findings of the study indicate that rape in adulthood can result in relatively specific consequences for the course and outcome of labour, and that these should be given special attention. The experience of rape is often associated with a deep sense of shame; it remains unspoken of within the woman’s surroundings and so often remains unprocessed emotionally. Experiences of abuse are difficult themes to bring up; health personnel often do not ask about these issues and so do not identify them, and may feel insufficiently prepared to approach them if the woman should bring them up on her own.8,30,31

It is necessary to try to replicate our findings in future studies in other settings. Such studies could focus on managing the second stage of labour in the best way for women who have been raped. Another interesting starting point for a study would be to analyse birth outcome when rape was the sexual debut. It may at the same time be useful to examine the way childhood rape, either as a single occurrence or repeated over time, can influence the course of labour. Future research should also focus more qualitatively on the woman’s experience of labour, and on what kind of care best meets her needs.

Conclusion

Rape in adulthood and labour outcome have not previously been associated. This study indicates that there is an increased risk of a prolonged second stage of labour, caesarean section and operative vaginal delivery in the first delivery in women with a previous history of rape in adulthood. The validity of the results must be interpreted in the context of the study population and cannot be generalised to hold for all women who have been raped. Obstetricians and midwives caring for women in pregnancy and birth should acknowledge prior rape as a relative obstetrical risk factor. The approach to the woman throughout labour should help the woman, and not contribute to re-traumatisation.

Disclosure of interest

None.

Contribution to authorship

HN and LH are equal first authors. PØ and TS have been co-authors. EB wrote the first draft of the Methods section and has contributed to the analyses and revised the draft of the manuscript. BS has performed the regression analyses.

Details of ethics approval

The study was approved by the Regional Committee for Medical and Health Research for North Norway and the Ombudsman for Privacy at the University Hospital of North Norway.

Funding

The study was financed by the North Norway Regional Authority Clinical Research fund.

Acknowledgements

None.

Ancillary