Skip to content

Childbirth Research Papers

lthough the state of pregnancy is both normal for, and often desired by, women in early adulthood, it is not uncommon to experience anxiety at the physical and emotional consequences of the gravid state. It is not the purpose of this paper to comment on psychiatric and psychological descriptions of the “normal” pregnant woman but concentrate on some pathological psychiatric conditions that may effect pregnant women. Anxiety and fear of pregnancy and childbirth are documented features of eating disorders, mood disorders, and pathological dread and avoidance of childbirth or tokophobia.

There are disorders in which anxiety is the primary symptom of a pathological mental state. First medically described by Marcé in the 18th century, the disorders of anxiety during pregnancy, especially those in the first trimester have profound features of turmoil.1 Paradoxically, some women with a pregravid anxiety disorder have a decrease of symptoms in the first trimester.2 Some level of anxiety may be protective and a portrait of the primiparous woman most likely to have a positive birth experience was proposed as anxious and realistically fearful, yet competent in her knowledge of labour and delivery.3


Although depressive illness is often perceived as a disorder of postnatal mothers it is, in fact, as common in pregnant women.4–7 Postnatal depression has become a focus of concern and the consequences to the child, mother, and family may include neglect of the child, family breakdown, self harm, and suicide. Research also suggests that the babies may develop emotional or behavioural problems or cognitive delays in childhood.8,9

In contrast, depression during pregnancy has been relatively neglected. Indeed, pregnancy has been thought to protect against depression. Studies of antenatal psychopathology have mostly examined antenatal mood as a predictor of postnatal depression.10

Watson et al found that in 23% of those with postnatal depression, this had started during pregnancy.11 Depressed mood during pregnancy has also been associated with poor attendance at antenatal clinic, substance misuse, low birth weight, and preterm delivery. Psychopathological symptoms during pregnancy have physiological consequences for the fetus and this may explain some of these effects.12

There is good research, dating back to that of Esquirol in 1818, describing all pregnancy related mental illness. He called this “puerperal insanity” and included recurrent prenatal depression.13 More recent studies, such as those by Menzies14 and Knauer,15 have reinforced the extent and severity of prenatal depression. Evans et al suggest that the proportion of women rating themselves as severely depressed was similar before and after childbirth.16 They suggest that postnatal depression is not a distinct syndrome. Their data suggest that childbirth is less likely than the events of pregnancy to be followed by depression in women who are not depressed. Furthermore, depressed women are more likely to improve after childbirth. These conclusions have been challenged.17

Ramrakha et al identified an increased probability of “risky sex” across a range of mental health diagnoses.18 Even clinical depression was associated with increased rates of “risky sex”, sexually transmitted diseases, and early sexual experience. With regard to sexual initiation, other questions arise, such as the role of sexual abuse—also a major precursor to mental health disorder. Risk taking may represent an indirect expression of anger or a mechanism for exerting control over one’s life. For a depressed young person, sexual activity and the associated risks may be seen as a diversion, a relief of tension or a salve of affection seeking—a self medication with sex.19

Depression is never trivial in mothers. The severity of the illness can have catastrophic consequences. It is a tragedy that some women die in the aftermath of childbirth. In fact, when all deaths in the year after delivery are taken into account, the Confidential Enquiry into Maternal Deaths showed that deaths from suicide are not only the leading cause of indirect death, but also the leading cause of maternal death overall.20 For some the fatal outcome may have been avoided.


Although the deliberate killing of a child by the mother has been described from ancient times, it is probably more reasonable to consider this act in two forms. These are the customary neonaticide of unwanted newborn babies and the killing of children for other reasons.

Neonaticide has often been focused on the differential killing of female children. This has led to a substantial shift in the number of reported male births to female births, figures sometimes exceeding 3:1. Although this is less common as contraception becomes more available, there are still areas of the world in which female babies are deliberately killed.

It is not the purpose of this article to look at the psychopathological causes for infanticide, nor the law surrounding it. However, the reader may wish to review the excellent chapter in Motherhood and Mental Health by Emeritus Professor Brockington (see additional reading).


Anorexia nervosa

Eating disorders or disordered eating behaviours frequently occur in the childbearing years. The defining psychopathology of anorexia nervosa is a phobic avoidance of normal body weight. There are comments in the literature, often repeated, that there is an “antagonism” between anorexia nervosa and reproduction, because of the cessation of menstruation in active cases of anorexia nervosa. Pregnancy is rare in active anorexia nervosa except where ovulation has been induced.21

There are several studies that look at the outcome of pregnancy after recovery from the disorder.22–24 There are also case reports of pregnancy occurring in women who had not yet recovered. At follow up, the babies born to these women have, as a group, abnormally low birth weights.25 A higher rate of stillbirths is reported.24 These women may have more premature births and caesarean sections. Research suggests no difference between those with active and remitted symptoms of anorexia nervosa as a variable.26

Bulimia nervosa

Bulimia nervosa is characterised by a loss of control over eating, with self perpetuating cycles of bingeing and purging. It is common27 and highly responsive to specific forms of psychotherapy.28 Women who suffer from this condition generally improve during pregnancy.29 However, there is some research to suggest that the babies of bulimic mothers are premature and underweight.30 Although bulimic symptoms may improve as the pregnancy progresses for the majority of women, over half had worse symptoms after pregnancy than before. Severe bulimic symptoms at conception, a history of anorexia nervosa, gestational diabetes, and unplanned pregnancy all predicted relapse. Postpartum depression occurred in one third of the group and was predicted by the presence of a lower body mass index at conception, a higher frequency of bingeing after delivery, and a higher weekly alcohol intake at conception. Two thirds of those with a history of anorexia nervosa reported postpartum depression.31 The findings of the studies published to date, while somewhat inconsistent, suggest that both anorexia nervosa and bulimia nervosa may negatively effect fetal outcome. Together the data suggest that having a past or current eating disorder may put a mother and her infant at risk for problems during pregnancy.32

Sexuality and eating disorders

Seminal descriptions of anorexia nervosa by Lasegue in 1873 cited difficulties in heterosexual relationships as aetiological.33 This notion has often been repeated but rarely substantiated. Studies of sexuality and anorexia nervosa do suggest an association with body mass index but have been limited by cross sectional34 or retrospective study designs.35 The latter study also suggested that sexual activity in anorexia nervosa does not necessarily reflect levels of sexual interest or enjoyment. This disparity between drive and behaviour resonates with the similar disparity between hunger and feeding behaviour in anorexia nervosa, demonstrating the core subversion of appetitive behaviour. Balakrishna and Crisp have suggested that anorexia nervosa represents an avoidance response to the sexual thrust of puberty,36 while Tuiten et al have argued that changes in sexuality are solely due to the hypogonadism of emaciation.37 There are apparent differences in psychosexual histories between patients with restrictive anorexia nervosa, purging anorexia nervosa, and bulimia nervosa. Purging anorectics report more past sexual history than restrictive anorectics,35 while women with bulimia are generally more sexually active and experienced than even women without an eating disorder, with greater lifetime numbers of sexual partners and higher rates of induced abortion.38 Oral contraceptives are inappropriate and unreliable in the presence of regular self induced vomiting.21 High levels of childhood sexual abuse are reported to be associated with eating disorders.21


It has historically been very difficult to accurately estimate the rate of miscarriage in comparison to completed pregnancies. Research has discussed the estimated rates between 1845 and 1933 before readily available termination.39 More recently, diagnosing pregnancy using radioimmunoassay showed that 43% of all pregnancies were lost by spontaneous abortion,40 a large percentage of these not recognised by the woman.

There are well documented psychological and psychiatric consequences to a miscarriage.41 Seven percent of all miscarriages occur in women who have recurrent spontaneous abortion,39 although more recently the likelihood of successful subsequent pregnancy is much better.42


The psychological and psychiatric effects of termination are complex and often related to the reason for that termination. In the large study of Ekblad the author came to a conclusion that was subsequently often repeated, that is, “The greater the psychiatric indication the greater the risk of adverse sequelae”.43 In addition, it should be noted that all the postpartum psychiatric complications may also occur after termination of pregnancy. Some women have intense and vivid fantasies of “killing the baby” for over a year.44 It has been suggested that “almost every woman who entertained the thought of obtaining a termination, however briefly, became depressed in pregnancy or after delivery”.45 In one study two women terminated a pregnancy because they were too terrified to endure a delivery.46 Termination may be requested by women who suffer from tokophobia, dearly want a baby, but are unable to understand their own aversion to parturition. In the absence of an empathic professional ear or relevant medical literature their only choice may be to terminate the pregnancy and live with the psychological impact of that decision.

Psychological maladjustments after abortion are significantly associated with a history of depression. It has been suggested that prior psychological state is equally predictive of subsequent depression among women with unintended pregnancies regardless of whether they abort or carry to term.47 Reardon and Cougle challenge this.48 They suggest that married women who carry an unplanned pregnancy to term were at less risk of depression than married women who aborted. Single women, however, had high depression scores independent of outcome.


Prevalence of fear of pregnancy and childbirth

Fear of childbirth is common and more intense in pregnant nulliparous women than in pregnant parous women.49,50 Over 20% of pregnant women report fear and 6% describe a fear that is disabling.51,52 Altogether 13% of non-gravid women report fear of childbirth sufficient to postpone or avoid pregnancy.53

It is well known that pregnancy may be a time of considerable anxiety with symptoms escalating in the third trimester.54 Women still suffer from the fear of death during delivery.55 When this specific anxiety or fear of death during parturition precedes pregnancy and is so intense that tokos (childbirth) is avoided whenever possible, this is a phobic state called “tokophobia”. Tokophobia may effect women from childhood into old age.

Prevalence of anxiety

In 1941 Sontag warned that a woman’s emotional state could effect her pregnancy.56 He stated “deeply disturbed maternal emotion produces a marked increase in activity of the fetus”. Many researchers have subsequently studied the causes and consequences of maternal anxiety and fear. Fear of pain is often reported as the reason for fearing delivery. Fearful women with free access to analgesic drugs during labour were still more likely to experience childbirth negatively and suffer severe emotional imbalance postpartum. Thus “fear of pain” in a Western society may be a culturally accepted way of expressing something more complex.57 More recently, pregnant women fearful of childbirth reported a lack of trust in the obstetric team, fear of their own incompetence and fear of dying.58 Other studies have suggested that the greatest fear was of delivering a physically damaged or congenitally malformed child.59,60 Women who have suffered childhood sexual abuse or rape fear the experience of childbirth will revisit the distress and helplessness of abuse. Women who have already suffered during childbirth are afraid of retraumatisation.53

Treatment studies for fear of childbirth

Women with dread of childbirth are vulnerable to increased surgical intervention and subsequent psychological complications. Studies investigating the alleviation of fear of childbirth date back to the 1920s.61 Psychoprophylaxis was investigated in the 1950s62 and more recently the benefits of hypnosis.63

A psychoprophylactic preparation course offered to pregnant women afraid of childbirth made no significant difference to obstetric outcome.64 Psychological outcomes were not investigated. Important research has been conducted in Sweden. Ryding, an obstetrician and psychotherapist, offered either counselling or short term psychotherapy to pregnant women demanding a caesarean section that the obstetrician thought unnecessary.65 At term, half these women chose vaginal delivery. Sjogren investigated 72 women with severe anxiety about childbirth.66 They were offered psychotherapy or extra obstetric support. Subsequently, some women chose a vaginal delivery. These women experienced the delivery as positively as a reference group. Tokophobic women who strongly desired a surgical delivery and were refused suffered greater psychological morbidity than those granted their chosen delivery method.53 The number of women demanding elective caesarean section because of tokophobia is not known. The National Sentinel Caesarean Section Audit was published in October 2001. The audit found that one in five (21.5%) of births in England and Wales were by caesarean section; 7% of these were at the mother’s request with no medical reason.67


Tokophobia has been classified as follows:

  • Primary—nulliparous.

  • Secondary—previous traumatic deliveries.

  • Secondary to depressive illness in pregnancy.

(1) Secondary tokophobia

Some women develop a dread and avoidance of childbirth after a traumatic obstetric event in a previous pregnancy. Most typically this is a “traumatic” delivery but it could also occur after an obstetrically normal delivery, a miscarriage, a stillbirth, or a termination of pregnancy. Some women feel so traumatised they avoid a further pregnancy even when a baby is desperately wanted.

Secondary tokophobia and the relation to post-traumatic stress disorder (PTSD)

PTSD is increasingly recognised as a consequence of childbirth and is associated with a pathological dread and avoidance of further pregnancy and childbirth experiences.68 In 1875 Savage alluded to “a startling and horrible dream” after childbirth, succeeded by a “melancholy stupor”.69 Knauer described depression after a severe labour with a resulting phobia for pregnancy in 1897.15 However, it was the seminal work by Bydlowski and Raoul-Duval in 197868 that documented the “nightmare” of childbirth.68 They described 10 cases of “névrose traumatique post-obstétricale” in women who endured long, painful deliveries. They stated:

“Parturition—especially the first—can, by its obligatory violence and confrontation with an imminent and lonely death put the mother under extreme stress”.

They suggested that the aftermath of delivery trauma was intensified in a subsequent pregnancy. These traumatised women avoided childbirth. Some suffered nightmares so terrifying that they were unable to sleep.70 The clinical picture and course of PTSD after childbirth was described in four cases in 1995.71 PTSD may follow deliveries that “need not have been abnormal from the clinicians perspective”.72 In extreme and tragic cases, women will terminate much wanted pregnancies because they are unable to deal with the idea of another delivery.53 Almost 2% of women have a “post-traumatic stress symptom profile” related to childbirth in the first year postpartum.73 Furthermore, one third of women suffer “serious post-traumatic intrusive stress reactions” in the two months after an emergency caesarean section.74 Consequently, post-delivery stress clinics have been developed.75

Secondary tokophobia and the “vicious cycle principle”

Childbirth impacts on a subsequent delivery. An emergency caesarean section or instrumental vaginal delivery increases fear of childbirth in a subsequent pregnancy.76 Additionally, women who suffer from fear of childbirth during pregnancy subsequently have an increased rate of emergency caesarean section or instrumental vaginal delivery.64,76,77

Not surprisingly, the mode of delivery influences women’s psychological state after the delivery. Ryding et al reported that women who had an emergency caesarean section or instrumental vaginal delivery suffered more negative psychological reactions after delivery than women experiencing elective caesarean section or normal vaginal delivery.78 There is a clear relationship between women’s expectations before delivery and experiences after. So, the expectation of being in control (both self control and control over what was done to one) is positively associated with achieving that aim and with higher satisfaction with that delivery.10,79

Therefore, the expectation a pregnant woman has about the anticipated delivery might be highly relevant for her experience of and behaviour during the delivery. That is, her appraisal before delivery might influence her appraisal after—“a vicious cycle principle”.80 For some women the fear lasts many years.49

(2) Primary tokophobia

When dread of childbirth predates the first conception, this is primary tokophobia. The dread of childbirth may start in adolescence or early adulthood. Although sexual relations may be normal, contraceptive use is often scrupulous.53 Pregnancy is avoided to prevent parturition. In some tragic cases, a woman is so terrified of childbirth, she will terminate a wanted pregnancy rather than go through childbirth. Some women will actively seek out an obstetrician who will perform an elective caesarean section before becoming pregnant for the first time. Some women never overcome their fear of childbirth and remain childless and some adopt. Many feel shame at their perceived inadequacy. Some women enter the menopause having never delivered a much desired baby and grieve this loss into old age.

Aetiology of primary tokophobia

The aetiology of primary tokophobia is likely to be multifaceted. Psychological and social considerations have been postulated.81

A. Social culture

Fear of childbirth may transmit over generations82 and this can produce a second generation effect of a mother’s own unresolved frightening experience.83 It is suggested that women’s reproductive adaptation is like their mother’s. This suggests a psychological heredity.83 This may be further highlighted for girls when their parents have a negative attitude towards sexuality.76 Women accurately recall details of childbirth 20 years later.84

B. Anxiety theories

A phobia is an avoidance response. It may be learnt through frightening experiences, vicariously by seeing other’s fearful responses, or through instruction.85 Zar86 used the work of Lazarus87 to investigate fear of childbirth. She suggested a pregnant woman’s expectations of the delivery are relevant to her experiences of and behaviour during delivery. Furthermore, the appraisal of the last delivery will include the anxiety associated with it and indicate the level of fear for a future delivery.

Fear of childbirth has been associated with anxiety proneness in general80 and may belong to the family of anxiety disorders.

C. Trauma and abuse

Baker and Duncan reported that 12% of women described being sexually abused before the age of 16.88 Psychological morbidity secondary to childhood sexual abuse may be immense and diverse with increased rates of sexual dysfunction,89 anorexia,90 and PTSD.91

A history of childhood sexual abuse could be associated with an aversion to gynaecological examinations including routine smears or obstetric care. The trauma of vaginal delivery, or even the contemplation of it, may cause a resurgence of distressing memories. This can lead to dread and avoidance of childbirth even when a woman wants a baby.53

(3) Tokophobia as a symptom of depression

Less commonly, prenatal depression may present with tokophobia. With this, there is no morbid dread of childbirth and this fear is in stark contrast to the woman’s prepregnancy and predepression beliefs.46


Psychological morbidity is common in women during their childbearing years. However, conditions such as PTSD and tokophobia may not have any obvious preceding cause and may be unrecognised. The outcome of all of the conditions considered in this paper is less good than the outcome for other mothers. There is a need for psychiatrists and obstetricians to work together to improve this under- recognised and under-treated group of women and babies.

Further reading

  • In addition we would like to recommend the reading of Motherhood and Mental Health. This is a comprehensive overview of this specialised area of psychiatry by Emeritus Professor Ian Brockington published by Oxford Medical Publications in 1996.

Learning points

  • Deaths from suicide in the year after delivery is the leading cause of maternal death overall.

  • When dread and avoidance of childbirth predates the first conception, this is primary tokophobia.

  • Secondary tokophobia may be associated with post-traumatic stress disorder, postnatal depression, bonding disorder with the baby, and avoidance of a subsequent pregnancy.

  • Women with anorexia nervosa have higher rates of premature births and caesarean section.

  • Women with bulimia nervosa often have symptom improvement during pregnancy. However, they often relapse in the postnatal period with more severe symptoms than before conception.


  1. Tokophobia:

  2. Can be found in men

  3. Can start in childhood

  4. Can continue after the menopause

  5. Is a cause of childlessness

  6. Depression:

  7. Is as common in pregnancy as the postnatal period

  8. May cause tokophobia

  9. If severe and postnatal will delay developmental milestones

  10. Is the second most common cause of death of new mothers

  11. Anorexia nervosa:

  12. Can cause infertility

  13. If treated successfully improves the obstetric outcome

  14. Is associated with increased caesarean section rate

  15. Is associated with tokophobia

  16. Bulimia nervosa:

  17. Is associated with reduced libido

  18. Improves during pregnancy

  19. Improves postnatally

  20. Protects against postnatal depression

  21. Post-traumatic stress disorder after childbirth:

  22. May cause primary tokophobia

  23. Was described in 1995 by Ballard et al

  24. Responds successfully to debriefing

  25. Is associated with postnatal depression

  26. Tokophobia may be associated with:

  27. An obstetrically normal delivery

  28. Relationship breakdown during pregnancy

  29. Request for general anaesthetic during delivery

  30. Request for termination


1. (A) T, (B) T, (C) T, (D) T; 2. (A) T, (B) T, (C) F, (D) F; 3. (A) T, (B) F, (C) T, (D) F; 4. (A) T, (B) T, (C) F, (D) F; 5. (A) F, (B) T, (C) F, (D) T; 6. (A) T, (B) F, (C) T, (D) T.


Pain and epidural use in normal childbirth

Most popular

  • 13 March, 2018
  • 13 March, 2018
  • 13 March, 2018

Recent Articles

  • 30 September, 2016

    The association between maternal self-report of depression and the risk of pre-eclampsia: outcome data from 960 participants in a retrospective case-control study

  • 30 September, 2016

    Building on success and learning from experience: recommendations from the UK Stern report

  • 30 September, 2016

    Information for authors, news and resources. Evidence Based Midwifery 14(3): 107

  • 30 September, 2016

    ‘We all think with the same brain’: midwives’ stories of normal birth in a community of practice

  • 29 September, 2016

    The development, implementation and evaluation of a leadership programme for midwives

  • 29 September, 2016

    The influence of a patriarchal culture on women’s reproductive decision-making: exploring the perceptions of 15 Nepali healthcare providers

  • 28 September, 2016

    Planning birth in and admission to a midwife-led unit: development of a GAIN evidence-based guideline

  • 30 June, 2016

    News and resources from EBM June 2016

  • 30 June, 2016

    The effect of intrapartum pethidine on breastfeeding: a scoping review

  • 30 June, 2016

    Determining the influencing factors of a caesarean section birth on breastfeeding

More »

26 August, 2009

Pain and epidural use in normal childbirth

This paper is based on the Zepherina Veitch Memorial lecture given at the RCM’s annual event held in Belfast on 18 June 2009. Evidence Based Midwifery: September 2009

Denis Walsh PhD, MA, RM, RGN.

Associate professor of midwifery, University of Nottingham, Nottingham City Hospital, Postgraduate Centre, Hucknall Road, Nottingham NG5 1PB England. Email:

This paper is based on the Zepherina Veitch Memorial lecture given at the RCM’s annual event held in Belfast on 18 June 2009.


With epidural rates doubling in the UK over the past 20 years, the impact on normal labour and birth is profound. Changes have also occurred in wider birthing milieu, such as the rise of a risk discourse, the diminishing of a ‘rites of passage’ meaning to birth, the growth of obstetric anaesthetic services and the advent of informed choice in maternity care policy. This paper discusses these issues and argues that inadequate service provision and an impoverished approach to labour pain rather than women’s preferences are contributing to the rise in epidurals. An elective epidural service in relation to low-risk women is challenged and a call made for an urgent debate on how maternity services and ultimately society should respond to these profound changes.

Key words: Epidural, Zepherina Veitch memorial lecture, normal birth, risk, pain, evidence-based midwifery


This paper discusses rising epidural rates in low-risk labour in the UK, proffering some reasons for this trend. The side-effects of epidurals are detailed and, from this, the suggestion is made that epidural use is incompatible with normal labour. It then argues that inadequate service provision is the main contributor to the rise. Contrasting pain paradigms are then outlined, based on Leap’s (2000) and later Leap and Anderson’s (2008) seminal research and writing. Their approach of ‘working is pain’ is critically examined to see if it offers a way forward for the current debate around labour pain.

An epidural epidemic?

Epidural rates have doubled in the UK from 17% in 1989 to 33% in 2007/08 (BirthChoiceUK, 2009). Though the reasons for this have never been investigated, it is likely that some of the following play a part in this change:

• Elective epidural provision is now almost universally available in consultant maternity hospitals in the UK. A survey in 2006 found only four out of 196 consultant maternity units did not provide this option (Jones et al, 2008). Obstetric anaesthetists now have their own association and their numbers have grown substantially over the past 20 years (Wee et al, 2002)

• Epidural provision has been available in some UK maternity units for nearly 30 years and hence, crosses two generations of childbearing women. Anecdotally, midwives say the mothers of the childbearing women now more commonly recommend epidurals to their daughters than they did a generation ago

• Celebrity birth stories and media portrayals of childbirth often include epidurals (Daily Mail, 2004)

• Over recent decades, there has been a loss of ‘rites of passage’ meaning to childbirth, so that pain and stress are viewed negatively (Leap and Anderson, 2008)

• A technorationalist society considers pain as either preventable or treatable (Lauritzen and Sachs, 2001)

• The pain relief paradigm is dominant in maternity services (Leap and Anderson, 2008)

• The movement to institutional birth (93% hospital verses 7% home and birth centres) reinforces medical solutions to clinical symptoms, such as pain (Walsh, 2007)

• Fragmented models of care and loss of continuity contributes to greater use of pharmacological agents in labour (Hodnett et al, 2007)

• Informed choice as an ethical imperative influences practitioners’ responses to maternal requests for pain relief in labour (Walsh, 2007)

• The risk discourse predisposes to childbirth intervention including the use of pain-relieving agents (Walsh, 2007).

Several of these factors work in tandem. ‘Technorationalist society’ (Lauritzen and Sachs, 2001) is shorthand for a society that equates all scientific advances with progress. In relation to pain, technology and drugs have either prevented pain from emerging or treated it effectively when it does. It is counter-cultural in such a society to see a purpose to pain, especially physical pain related to biological function, which is how traditional and indigenous societies have probably viewed childbirth over thousands of years (Kitzinger, 2000). Childbirth within indigenous societies studied by Jordan (1993) was viewed as a ‘rite of passage’, an anthropological phrase referring to growth milestones. Rites of passage are associated with movement from one level of maturity and responsibility to another (van Gennep, 1966) – in the context of childbirth from woman to mother. It commonly involves passing through an experience of challenge and uncertainty known as a luminal phase before re-integrating into the new role.

Allied to an antipathy to childbirth pain, is a risk discourse that carries within it several paradoxes. In the West, it has never been safer to have a baby if judged by maternal and perinatal mortality rates (Department of Health, 2007), yet it appears that many women have never been more frightened of the process. The relatively new diagnostic category of ‘tokophobia’ (morbid fear of labour) is testament to that (Hofberg and Brockington, 2000). Another paradox is a high degree of risk aversion, yet a willingness to embrace medical interventions like drugs and surgery that carry risks themselves. Risk aversion appears to operate quite selectively. Mixed messages co-exist like a public health message to avoid any form of drug pre-conceptually and prenatally, but accept an array of drugs during intrapartum care.

Discussion about epidurals is often linked to the broader discussion of medicalisation of childbirth, because epidural typifies the ‘cascade of intervention dynamic’ that contributes to medicalisation (Johanson et al, 2002). Public health concern has prompted the Department of Health in the UK to measure normal labour processes as well as normal birth outcomes (NHS Information Centre, 2008).. A working definition has now been agreed as to what constitutes a normal labour and birth and it excludes epidural or spinal anaesthetic, induction of labour, forceps or ventouse, caesarean section (CS) and episiotomy. The difficulty of reaching agreement across a multi-disciplinary group is reflected in the fact that it can include augmentation of labour, artificial rupture of membranes, the use of entonox, opioids, electronic fetal monitoring and an actively managed third stage of labour (Werkmeister et al, 2008). The fact that the Consensus Group had to compromise to accommodate the various positions of the stakeholders indicates how certain procedures and drugs are now considered normal in labour. The ubiquity of oxytocin augmentation and epidural is demonstrated in Mead and Kornbrot’s (2004) survey of low-risk women, when they found that rates varied between 48% and 76% (augmentation) and 29% and 62% (epidurals). Furthermore, Symon et al (2007) demonstrated the stark contrast in low-risk women’s self-reported intervention rates between midwifery-led units and consultant units: pain agents were 25% in the former and 77% in the latter.

Box 1. A story from a midwife typifying the current conflict around pain and labour

The midwife had taken over from her colleague who was looking after a woman having her second baby. The woman had been in the latent phase of labour, but had recently shown signs of her labour accelerating. In the short time it took to hand over, the woman had become very distressed. The midwife rapidly tried to develop a rapport with her and gave some advice about focusing on breathing during the contractions. This was not enough and she began using entonox within a short period. The contractions were long and intense and beginning to get
expulsive. As the midwife auscultated the fetal heart on the woman’s abdomen, she noted that the auscultation point was tracking down the abdomen to rest over the symphysis pubis. She recognised the familiar manifestation of transition, but by then the woman was shouting loudly ‘to go home’, ‘caesarean now’ and ‘get me an epidural’.

Her distress was greater on the bed so the midwife encouraged her to get up, though she was continuously monitored because of meconium-stained liquor. She coped a little better upright or on the floor but still vocalised her distress in no uncertain terms. The midwife was faced with a dilemma. She was sure the second stage of labour was imminent, but the recourse to an epidural would have calmed the woman and made monitoring the fetal heart easier as she would have been semi-recumbent on the bed. After another 15 minutes, the woman was bearing down strongly and birthed a healthy baby boy. Later, both an anaesthetist and another midwife suggested an epidural was wholly appropriate in this situation and a lively discussion ensued.

Side-effects of epidurals

When it comes to specific risks associated with a medical procedure, epidurals have many. These include:

• Increased length of first and second stage of labour (Anim-Somuah et al, 2009)
• Need for more oxytocin (Anim-Somuah et al, 2009)
• Increased incidence of malposition (Anim-Somuah et al, 2009)
• Increase in instrumental delivery (Anim-Somuah et al, 2009)
• Increase in third- and fourth-degree tears (Rortveit et al, 2003).

Maternal side-effects are:

• Reduction in mobility (MIDIRS, 2006)
• Can lead to an inability to pass urine (MIDIRS, 2006)
• Hypotension, headache (MIDIRS, 2006)
• Pyrexia (Yancey et al, 2001)
• Up to 30% of women get partial, but not complete relief (Simkin, 1989)
• Reduces breastfeeding rate on discharge from hospital (Wiklund et al, 2009).

Neonatal side-effects include tachycardia due to temperature rise, and they are more likely to be hypoglycaemic (Lieberman and O’Donoghue, 2002). It also diminishes breast-seeking and breastfeeding behaviours (Ransjo-Arvidson et al, 2001). All of these negatives have to be balanced with the fact that an epidural offers the following advantages:

• It is more effective than non-epidural analgesic methods (Anim-Somuah et al, 2009)
• It makes CS safer (May, 1994)
• It enables pain-free assisted vaginal birth
• It is valuable for protracted, induced or augmented labours (MIDIRS, 2006)
• It is useful for some women with tokophobia or post-traumatic stress disorder (Heinze and Sleigh, 2003).

Few childbirth professionals would argue against epidural availability and use in these or similar situations. Because of the side-effects outlined above, the position of this paper is that epidurals render labour non-physiological and therefore a pivotal point for discussion is its role in normal labour. A related question is to what extent ‘informed choice’ has become an ethical imperative, regardless of context and prior preferences (for example, as expressed in a birth plan). In the story above (see Box 1), the scenario of a multiparous woman requesting an epidural in late first stage of labour has been used as an exemplar for the application of informed choice. Many UK midwives would express the tension between responding to a woman’s request in this situation and knowing that this is a transient and challenging part of the labour that will soon pass. Midwives from other countries have expressed surprise when this scenario has been presented in workshops, concluding quite unequivocally that ‘an epidural is not appropriate’.

Epidural rise and the inadequacies of service provision

While recent media coverage reported the epidural debate as professionals urging women to accept labour pain (The Observer, 2009), the original intent of the Zepherina Veitch memorial lecture (RCM, 2009) was to highlight the failure of the maternity services to provide forms of care that are known to lower epidural rates. High-level evidence exists in three areas: one-to-one care in labour (Hodnett et al, 2009), the provision of a home-like birthing environment (Hodnett et al, 2009) and access to water immersion (Cluett et al, 2009). All three forms of care lower epidural rates in low-risk groups, especially in nulliparous women. Yet, surveys reveal these forms of care are not universally available to women in the UK (Healthcare Commission, 2008; Redshaw et al, 2007; Alderice et al, 1995). This is despite the fact that the effectiveness of the first two (one-to-one care and home-like birthing environments) has been known about for over ten years. Given the slowness of service provision to implement these forms of care, it is not surprising that women opt for epidurals. Fragmented care systems and clinical, austere labour rooms would appear to be risk factors in themselves for greater reliance on pain medication.

Research also points out that the best predictor of labour pain is maternal confidence (Leeman et al, 2003) and this opens up two further dimensions to an understanding of labour pain: the significance of relationally-mediated care as a conduit for building confidence and the importance of pre-existing expectations. In addition to the studies of one-to-one care in labour, randomised and non-randomised controlled studies of a specific organisational model called caseload, which guarantees a known carer for labour, consistently shows lower rates of epidural and other birth interventions (Benjamin et al, 2001; North Stafford, 2000). Hodnett (2002) discovered the overlap between pain perception, confidence and how women rate their childbirth experience when she undertook a systematic review of the role of pain in childbirth satisfaction. Her conclusion was that ‘the influences of pain and pain relief… on subsequent satisfaction are neither as obvious, as direct, nor as powerful as the influences of the attitudes and behaviours of the caregivers’ (Hodnett, 2002: 160).

Women reported positive birth experiences when they felt in control, when communication was effective and when power was shared in relation to decision-making (Carlton et al, 2005). Carlton et al (2005) showed that pain relief did not necessarily improve women’s childbirth experience and that such a request may indicate a need for emotional support. These findings are backed up by Kannan et al’s (2001) research, where most women requesting an epidural for pain reported being less satisfied with their childbirth experience, despite lower pain intensity.

The importance of prior expectations and beliefs was shown in Heinze and Sleigh’s (2003) study, exploring the differences between women who labour with or without an epidural. The epidural group had a higher fear of childbirth, were less aware of the side-effects, had an external locus of control for childbirth and a desire for passive compliance in the process. The non-epidural group had an internal locus of control and had less fear, believed that control came from within and wished to actively participate in the process.

Attitude to labour pain

Leap (2000) and Leap and Anderson (2008) have made a seminal contribution to the area of attitudes and beliefs about labour pain and have developed an explanatory theory called ‘working with pain’. Their theory is based on research into home-birth experiences of women and the midwives who attended them and, for this reason, may have less applicability to institutional birth settings. Nevertheless, their approach is at the very least theoretically generalisable to women in normal labour in other birth environments. They contrast ‘working with pain’ with ‘pain relief’ (Leap and Anderson, 2008). Table 1 summarises the main differences between these two approaches.

The ‘working with pain’ paradigm is predicated on labour physiology that requires pain to be present for the release of beta-endorphins, a naturally occurring opiate-like compound (Buckley, 2009). These hormones have analgesic and euphoric effects, similar to exogenous opiates and probably contribute to the dynamic behind oxytocin release so that it is neither understimulated nor overstimulated. Endorphin effects are also associated with altered states of consciousness, ecstatic experiences and excitation (Ribeiro et al, 2005). An altered state of consciousness may be important for a labouring woman in helping her behave instinctively, for example, to use bodily movements to assist in the descent of her baby (Buckley, 2009).

It is probable that it is these effects that are captured so beautifully in Hannah’s birth, a short DVD produced by Sheena Byrom (Byrom, 2006). What the DVD also captures are the empathic responses from Hannah’s birth companions. Moberg (2004) in her captivating book, The oxytocin factor suggests that this hormone is secreted in both men and women, especially during therapeutic touch. She highlights the necessity to bathe a birth setting in love, not fear. When this occurs, the synergy created is more than the sum of individual parts. Hence the centrality of empathic relationships to the birth process, the importance of an optimum environment and of minimising disturbance. All of this takes on an urgency in an institutional birth setting where some of these factors are already compromised.

The transforming power of labour

One of the key questions for maternity care stakeholders is what will happen to the narratives of transformation and growth in childbirth if normal labour pain is effectively removed by rising epidural rates. These are the countless number of personal testimonies that women share about an experience of growth and empowerment through childbirth. The vast majority of these are characterised by drug-free or low intervention labours, though not all (Thompson, 2007). Lundgren and Dahlberg (1998) found that women placed a meaning onto pain that assisted them in the transition to motherhood. Women in Callister et al’s (2003) cross-cultural study of pain perception viewed mastering pain as an integral part of a self-actualising experience and, for some, this increased their sense of self-efficacy. The most moving testimonies come from vulnerable women whose lives prior to birth had been blighted by abuse or disempowerment. Phrases like ‘my greatest achievement’ (Esposito, 1999), ‘I can do anything now’ (Spitzer, 1995) and ‘I feel so strong’ (Walsh, 2006) characterise these stories. In addition to these qualitative papers, randomised controlled trials of midwife-led care, where epidural provision is not available, show higher levels of satisfaction with the birth experience (Hatem et al, 2008). All these studies pose a profound challenge to the ‘pain relief’ paradigm.

There is a difficulty in debating the topic because it could imply criticism of women who choose or need intervention. Emerging evidence that normal birth primes the bonding areas of a mother’s brain better than CS birth adds to this perception (Swain et al, 2008). In recognition that CS birth may undermine birth physiology, obstetricians have been researching the so-called ‘natural caesarean’ (immediate skin to skin at birth, delayed cord-cutting) to see if normal physiology can be harnessed in this situation (Smith et al, 2008). The advent of the ‘mobile epidural’ illustrates how obstetric anaesthetists are trying to engage with labour physiology around movement and upright posture to accrue those benefits for women with epidurals.

These attempts to engage with childbirth physiology in the context of medical procedures that undermine it highlights how science struggles to mimic precisely what is natural. The complexities behind oxytocin secretion remind biomedicine that altering one variable (skin to skin in caesarean or bodily movement in epidural), laudable though those initiatives are, will struggle to reproduce the exact conditions for maximising birth physiology (Odent, 2001) – that probably requires a whole system approach (Downe and McCourt, 2008), examining environment, attitudes, beliefs, practices, and relationships, for example.

Elective epidural service

In the light of this discussion so far, a rationale certainly exists for questioning the appropriateness of an elective, ‘on-demand’ epidural service for women at low obstetric risk, especially if there is public health commitment to increasing the rate of normal labour and birth in the UK. However, given its embeddedness in maternity service provision, it would be a brave person who would take up such a position. This paper’s intent is to simply encourage debate about these issues.

An anecdote is told of a maternity care professional who used to refer to epidurals as ‘happidurals’. In the context of a fragmented model of care, with little continuity and patchy provision of one-to-one support in labour, in a clinical environment with little resemblance to home, it is understandable that epidurals are a welcome relief. But it is important not to confuse system failure with women’s preference. In fact throughout the UK in different birth settings, women are birthing entirely drug free, even with their first baby. This group can be found in midwifery-led units, birth centres and at home. Although this only represents a small minority of women, surveys suggest many more women would like these options to be available (Redshaw et al, 2007). First birth mothers’ stories of drug-free labours tend to remain hidden in small-scale birth settings because they are seldom told beyond these settings. Their testimonies are important for labour ward midwives, obstetricians and anaesthetists to hear, because they are routinely exposed to the opposite. Case reviews in maternity hospitals tend to be of complications and emergencies only.


The evidence is indisputable that epidurals undermine childbirth physiology. That rates are double what they were 20 years ago says more about the context of childbirth and childbirth professionals’ attitudes, than it does about the current generation of women’s ability to adjust to labour pain. In fact, there is considerable anecdotal evidence that women adapt their expectations to the service provision, so the rare consultant unit that does not have an elective epidural service has not seen a fall in bookings (A Musgrave, 2005: personal communication) and, birth centres remain popular with women where available. However, the vast majority of women anticipating a normal labour and birth enter a large maternity hospital where epidural provision is electively available. In this context, the impact on childbirth intervention rates is profound. Addressing this context requires a rethinking of pain paradigms, attention to birth environment, and a move to more relational models of care. Finally, there needs to be a robust debate about whether epidurals really serve the maternity services best by being an elective choice, especially in relation to normal labour and birth.


Alderdice F, Renfrew M, Marchant S. (1995) Labour and birth in water in England and Wales: survey report. BJM3(7): 375-82.

Anim-Somuah M, Smyth R, Howell C. (2009) Epidural versus non-epidural or no analgesia in labour. Cochrane Database Syst Reviews4: CD000331.

Benjamin Y, Walsh D, Taub N. (2001) A comparison of partnership caseload midwifery care with conventional team midwifery care: labour and birth outcomes. Midwifery17(3): 234-40.

BirthChoiceUK. (2009) Maternity statistics for England 2007-2008. See: (accessed 6 August 2009).

Buckley S. (2009) Sexuality in labour and birth: an intimate perspective: In: Walsh D, Downe S. (Eds.). Essential skills for intrapartum care. Elsevier: London.

Byrom S. (2006) Hannah’s birth DVD. Blackburn and Burnley NHS Trust.

Callister L, Khalaf I, Semenic S, Kartchner R, Vehvilainen-Julkunen K. (2003) The pain of childbirth: perceptions of culturally-diverse women. Pain Management Nursing4(4): 145-54.

Carlton T, Callister L, Stoneman E. (2005) Decision-making in laboring women: ethical issues for perinatal nurses. Journal of Perinatal and Neonatal Nursing19(2): 145-54.

Cluett ER, Nikodem VC, McCandlish RE, Burns EE. (2009) Immersion in water in pregnancy, labour and birth. Cochrane Database Syst Reviews2: CD000111.

Department of Health. (2007) Maternity matters: choice, access and continuity of care in a safe service. HMSO: London.

Daily Mail. (2004) Pain-free births putting lives at risk. See: (accessed 6 August 2009).

Downe S, McCourt C. (2008) From being to becoming: reconstructing childbirth knowledges: In: Downe S. (Ed.). Normal childbirth: evidence and debate. Churchill Livingstone: London.

Esposito N. (1999) Marginalised women’s comparisons of their hospital and free-standing birth centre experience: a contract of inner city birthing centres. Health Care for Women International20(2): 111-26.

Hatem M, Sandall J, Devane D, Soltani H, Gates S. (2008) Midwife-led versus other models of care for childbearing women. Cochrane Database Syst Reviews 4: CD004667.

Healthcare Commission. (2008) Towards better births: a review of maternity services in England. Healthcare Commission: London.

Heinze S, Sleigh M. (2003) Epidural or no epidural anaesthesia: relationships between beliefs about childbirth and pain control choices. Journal of Reproductive and Infant Psychology 21(4): 323-34.

Hodnett ED, Gates S, Hofmeyr GJ, Sakala C. (2007) Continuous support for women during childbirth. Cochrane Database Syst Reviews 2: CD003766.

Hodnett ED, Downe S, Edwards N, Walsh D. (2009) Home-like versus conventional birth settings. Cochrane Database Syst Reviews 1: CD000012.

Hodnett E. (2002) Pain and women’s satisfaction with the experience of childbirth: a systematic review. Am J of Obstet Gynecol 186: S160-72.

Hofberg K, Brockington I. (2000) Tokophobia: an unreasoning dread of childbirth. A series of 26 cases. British Journal of Psychiatry176: 83-5.

NHS Information Centre. (2008) NHS Maternity Statistics England 2006/07. See: (accessed 19 August 2009).

Johanson R, Newburn M, Macfarlane A. (2002) Has medicalisation of childbirth gone to far? British Medical Journal 321: 892-5.

Jones R, Swales H, Lyons G. (2008) A national survey of safe practice with epidural analgesia in obstetric units. Anaesthesia63: 516-9.

Jordan B. (1993) Birth in four cultures: a cross-cultural investigation of childbirth in Yucatan, Holland, Sweden and the US. Waveland Press: Prospect Heights.

Kannan S, Jamison R, Datta S. (2001) Maternal satisfaction and pain control in women electing natural childbirth. Regional Anaesthesia and Pain Medicine26(5): 468-72.

Kitzinger S. (2000) Rediscovering birth. Little, Brown: London.

Lauritzen S, Sachs L. (2001) Normality, risk and the future: implicit communication of threat in health surveillance. Sociology of Health and Illness23(4): 497-516.

Leap N. (2000) Pain in labour towards a midwifery perspective. MIDIRS Midwifery Digest 10(1): 49-53.

Leap N, Anderson P. (2008) The role of pain in normal birth and the empowerment of women: In: Downe S. (Ed.). Normal childbirth: evidence and debate (second edition). Churchill Livingstone: London.

Leeman I, Fontaine P, King V, Klein MC, Racliffe S. (2003) The nature and management of labour pain. Part 1: non-pharmacological pain relief. American Family Physician68(6): 1109-12.

Lieberman E, O’Donoghue C. (2002) Unintended effects of epidural analgesia during labour. Am J Obstet Gynecol 186: S31-68.

Lundgren I, Dahlberg K. (1998) Women’s experience of pain during childbirth. Midwifery14: 105-10.

May A. (1994) Epidurals for childbirth. Oxford Medical Publications: Oxford.

Mead MM, Kornbrot D. (2004) An intrapartum intervention scoring system for the comparison of maternity units’ intrapartum care of nulliparous women suitable for midwifery-led care. Midwifery20(1): 15-26.

MIDIRS. (2006) The use of epidural analgesia for women in labour. Informed choice for professionals. MIDIRS: Bristol.

Moberg K. (2004) The oxytocin factor. Perseus Books: New York.

North Staffordshire Changing Childbirth Research Team. (2000) A randomised study of midwifery caseload care and traditional ‘shared-care’. Midwifery16(4): 295-302.

The Observer. (2009) It’s good for women to suffer the pain of a natural childbirth, says medical chief. See: (accessed 7 August 2009).

Odent M. (2001) New reasons and new ways to study birth physiology. International Journal of Gynaecology and Obstetrics75: S39-S45.

Ransjo-Arvidson A, Matthiesen A, Lilja G. (2001) Maternal analgesia during labour disturbs newborn behaviour: effects on breastfeeding, temperature and crying. Birth23(3): 136-43.

Redshaw M, Rowe R, Hockley C, Brocklehurst P. (2007) Recorded delivery: a national survey of women’s experience of maternity care 2006. National Perinatal Epidemiology Unit. See: (accessed 7 August 2009).

Ribeiro S, Kennedy S, Smith R. (2005) Interface of physical and emotional stress regulation through the endogenous opioid system and mu-opioid receptors. Prog Neuropsychopharmacol Biol Psychiatry.29(8): 1264-80.

Rortveit G, Daltveit A, Hannestad Y, Hunskaar S. (2003) Vaginal delivery parameters and urinary incontinence: the Norwegian EPINCONT study. Am J Obstet Gynecol189: 1268-74.

RCM. (2009) RCM annual event 2009. See: (accessed 7 August 2009).

Simkin P. (1989) Non-pharmacological methods of pain relief during labour. In: Chalmers I, Enkin M, Kierse M. (Eds.). Effective care in pregnancy and childbirth. Oxford University Press: Oxford.

Smith J, Plaat F, Fisk N. (2008) The natural caesarean: a woman-centred technique. BJOG115(8): 1037-42.

Spitzer M. (1995) Birth centres: economy, safety and empowerment. Journal of Nurse-Midwifery40(4): 371-5.

Swain J, Tasgin E, Mayes L, Feldman R, Constable R, Leckman J. (2008) Maternal brain response to own baby cry is affected by caesarean section delivery. Child Psychology and Psychiatry49(10): 1042-52.

Symon AG, Paul J, Butchart M. (2007) Self-rated ‘no’ and ‘low’ risk pregnancy: a comparison of outcomes for women in obstetric-led and midwife-led units in England. Birth34(4): 323-30.

Thomson G. (2007) A hero’s tale of childbirth: an interpretive phenomenological study of traumatic and positive childbirth. Unpublished PhD thesis. University of Central Lancashire: Preston, Lancashire.

van Gennep A. (1966) The rites of passage. The University of Chicago: Chicago.

Walsh D. (2006) Improving maternity service. Small is beautiful: lessons for maternity services from a birth centre. Radcliffe Publishing: Oxford.

Walsh D. (2007) Evidence-based care for normal labour and birth: a guide for midwives. Routledge: London.

Wee M, Yentis S, Thomas P. (2002) Anaesthetists’ workload. Anaesthesia57(5): 484-500.

Werkmeister G, Jokinen M, Mahmood T, Newburn M. (2008) Making normal labour and birth a reality – developing a multi-disciplinary consensus. Midwifery24: 256-9.

Wiklund I, Norman M, Moberg K, Ransjo-Arvidson A, Andolf E. (2009) Epidural analgesia: breastfeeding success and related factors. Midwifery25(2): e31-e38.

Yancey M, Zhang J, Schwarz J. (2001) Labour epidural analgesia and intrapartum maternal hyperthermia. Obstetrics and Gynecology98(5): 763-70.


Printer-friendly version