Dr RD Maimburg, Department of Epidemiology, School of Public Health, Aarhus University, Bartholin Allé 2, 8000 Aarhus C, Denmark. Email firstname.lastname@example.org
Please cite this paper as: Maimburg R, Væth M, Dürr J, Hvidman L, Olsen J. Randomised trial of structured antenatal training sessions to improve the birth process. BJOG 2010;117:921–928.
Objective To compare the birth process in nulliparous women enrolled in a structured antenatal training programme, ‘The Ready for Child’ programme, with women allocated to routine care.
Design A randomised controlled trial.
Setting A Danish university hospital.
Participants Thousand hundred and ninety-three nulliparous women, recruited before week 22 + 0.
Methods Compliance to the protocol was monitored by questionnaires sent to the women by email, and by data from the local birth cohort database. Data were analysed according to the ‘intention-to-treat’ principle.
Intervention Women were randomised to receive 9 hours of antenatal training or no formalised training. Of the 1193 women, 603 were randomised to the intervention group and 590 were allocated to the reference group.
Main outcome measures Cervix dilatation on arrival at the maternity ward, use of pain relief and medical interventions during the birth process, and the women’s birth experience.
Results Women who attended the ‘Ready for Child’ programme arrived at the maternity ward in active labour more often than the reference group [relative risk (RR) 1.45, 95% confidence interval (95% CI) 1.26–1.65, P < 0.01], and they used less epidural analgesia during labour (RR 0.84, 95% CI 0.73–0.97, P < 0.01) but not less pain relief overall (RR 0.99, 95% CI 0.94–1.04, P < 0.72). Medical interventions and the women’s self-reported birth experiences were similar in the two groups. We found no adverse effects of the intervention.
Conclusion Attending the ‘Ready for Child’ programme may help women to cope better with the birth process. Adverse effects are few, if any.
For years antenatal training sessions have been part of the recommended antenatal care for pregnant women and their partners in Denmark and many other countries. Some countries even offer pre-pregnancy health education. The birth of the first child is a challenging event, and being prepared for it may help parents to cope with the many concomitant situations. Antenatal training sessions provide information on pregnancy, the birth process, infant care and early parenthood and may give parents more confidence.1,2 In recent years, antenatal training sessions have been phased out in most maternity wards in Denmark to save money, although nothing is known about the consequences of this, perhaps because evidence of the potential benefit of participating in antenatal training programmes is limited.3–10 Previous studies have been difficult to interpret, mainly because of limitations in study design and small sample sizes.10 The effect of an antenatal training session on the birth process remains largely unknown, but those who attend antenatal training programmes are described as highly motivated and well-educated women,4,11,12 making it difficult to interpret observational studies. We therefore conducted a randomised controlled trial to estimate whether attendance at antenatal training sessions during pregnancy had any effect on the birth process.
The women were enrolled from 10 + 0 to 21 + 6 weeks of gestation. Eligibility criteria included nulliparous women registered at the Aarhus Midwifery Clinic, older than 18 years of age at enrolment, singleton pregnancy, and the ability to speak and understand Danish. From May 2006 to May 2007 we sent an invitation to all pregnant women that registered for antenatal care and who fulfilled the inclusion criteria (n = 1980).
The women were invited to participate shortly after their first visit if this was scheduled before 22 weeks of gestation. Baseline data were collected when the woman accepted the invitation and before randomisation. Oral and written information was provided and the woman was randomised to either the intervention group or the reference group if she signed and returned the informed consent form.
Randomisation was assigned by a staff midwife using a computer-assisted voice response system. All citizens in Denmark have a unique personal identification number, which is stored in the Civil Registration System. The randomisation programme was set up to confirm that the unique personal identification number was correct by linking to the National Civil Registration System. Furthermore, the system ensured that the woman was only randomised once. A voice synthesiser spoke the registered data-input and regimen assignment. The randomisation programme used an algorithm generated by a data manager. The women were assigned with an average ratio of 1:1.
The intervention group received the ‘Ready for Child’ programme developed for this study. The programme comprised three modules, each lasting 3 hours. The training sessions were attended between 30 and 35 weeks of gestation, and the woman’s partner was also invited to participate. The content of the birth module included lectures on and discussion of labour onset, the birth process, the attending father, pain relief, birth interventions, fear of childbirth, and a film on giving birth. The newborn module included lectures on and discussions of how to care for the newborn, breastfeeding, childhood diseases, vaccination, and equipment and safety for the child in everyday life. The parent module included lectures on and discussions of transition to parenthood, maternity leave, sexual relations, conflicts in the parental relationship, the role of the grandparents, family and friends, and postnatal depression.
Four midwives, of varying seniority within the midwifery field, were chosen as instructors. The instructors attended a 3-day preparation course, 1 day for each module. They all adhered to a detailed teaching manual describing the content of the educational programme. The antenatal training programme was evaluated by the parents at the end of the course sessions to ensure that they had received instruction in all the subjects included in the programme. The reference group received standard care offered by the antenatal clinic, which did not include any antenatal training programmes, but different antenatal training programmes were provided by other stakeholders, mainly organised by relaxation therapists. Most of these programmes had a visiting midwife providing approximately 3 hours of lessons on the birth process.
Data were collected through questionnaires sent to the women’s email addresses or by regular mail. Questionnaires were sent out in weeks 24 and 36 of gestation and again 6 weeks and 1 year postpartum. Data on the birth process were collected and validated by trained midwives before being entered into the local birth cohort database. Data on cervix dilatation at the time the women arrived for delivery (the contact which ended with admission for birth) were collected specifically for this study. The midwife who examined the women on arrival registered information on cervix dilatation and the Delivery Fear Scale (DFS).13 The midwife read the ten statements in the DFS aloud to the woman, who then answered with a number between 1 and 5 (modified from 10 points in the original scale). Endpoints were marked with 1 (do not agree at all) and 5 (agree totally). The midwives in the maternity ward were blinded to the women’s randomisation status, and the women attending the study were told not to reveal their randomisation status to the midwife.
The study was approved by The Scientific Ethical Committee for Aarhus County and the Danish Data Protection Agency. The study was also reported to ClinicalTrials.gov.
The hypotheses of this study were the following.
The women attending the ‘Ready for Child’ programme would be empowered throughout the birth process leading to a later arrival at the maternity ward, measured by cervical dilatation in centimetres. They would use less pain relief and receive less medical intervention during the birth process. Their experience of the birth process would be more positive than for those who did not receive the intervention.
The main outcome measures were cervix dilatation (measured in centimetres) and the women’s ability to cope with fear (measured on DFS) on arrival, the use of different kinds of pain relief and common medical interventions during the birth process, and the women′s own birth experience (measured on a five-point Likert scale).
We estimated that 80% of the nulliparous women would arrive at the maternity ward with a cervix dilatation <3 cm and the sample size calculation was based on the assumption that no more than 70% of the women who attended antenatal classes would arrive with a cervical dilatation <3 cm. A total of 585 needed to be randomised to detect this decrease from 80% to 70%, using a statistical significance level of 5% and having 80% power to detect this difference.
Existing data indicated that 70% of the nulliparous women received pain relief during labour, and the sample size calculation was based on the assumption that no more than 60% of the women who attended antenatal classes would use pain relief during labour. A total of 712 women needed to be randomised to detect a decrease from 70% to 60% in use of pain relief during labour, using a statistical significance level of 5% and a power of 80%.
Incorporating a potential 15% lost to follow-up, the trial was set to include a total of approximately 980 women, which we expected to reach within 1 year of recruitment.
Data were analysed according to the ‘intention-to-treat’ principle and the results are reported in accordance with CONSORT requirements.14 We compared the intervention and reference groups at baseline (Table 1) for maternal age, body mass index, smoking, partner status, educational level and psychiatric history.
Table 1. Baseline characteristics of the women enrolled in the study (n = 1138)
Intervention mean (SD)
Reference mean (SD)
Data do not sum up to total number due to non responses.
Body mass index (kg/m2)
In 20 weeks of pregnancy
In relationship with partner
Living together with partner
Continuous data (cervix dilatation, DFS, newborn characteristics and birth experience) were assessed using Mann–Whitney U test. The analysis for cervix dilatation was made after excluding women who delivered by scheduled caesarean section (n = 50). Categorical data (medical data on the birth process) were analysed using the chi-square test.
All reported P values are two-sided, and the level of statistical significance was 5%.
Data management and statistical analysis were performed using stata Statistical Software, version 9 (STATA-Corp, College Station, TX, USA, 2006).
During the study period, 2295 nulliparous women were referred to antenatal care, of which 1193 were randomised. A total of 603 women were allocated to the ‘Ready for Child’ programme, while 590 women were randomised to a normal antenatal care regime.
The women’s baseline characteristics (Table 1) were collected in the first questionnaire, sent to the women in week 24 of gestation, 95% (1138/1193) of the women filled out the whole questionnaire. Information on compliance was obtained from the second questionnaire sent to the women in week 35 of gestation, completed by 89% (1058/1193) of the women.
In the intervention group, 72% (435/603) received the full allocated intervention, 85% attended the delivery session, 80% the newborn session, and 79% the parenthood session. Most of their partners participated in these sessions, respectively 72%, 66%, 67%. In the intervention group, a total of 4% attended other kinds of antenatal training sessions provided by stakeholders, for the reference group this figure was 45%.
In the intervention group, 16 women were lost to follow-up and 15 women were lost to follow-up in the reference group, giving a follow-up rate until the birth of the child of 97% (1162/1193) in both groups (Figure 1).
Information on the birth process was collected from an ongoing local birth registration. Information on cervix dilatation was available for 99% (1157/1162) and scores for DFS on arrival for birth were available for 62% (718/1162) of the women. The response rates for the DFS were 56% for the reference-group and 67% for the intervention-group (data not shown).
Information on the women’s own birth experience was obtained in the third questionnaire, sent to the women 6 weeks postpartum, and 89% responded (1067/1193) (Figure 2). Most women (98%) answered their questionnaires by email.
The women’s characteristics of potential importance for the birth process were similar for the two groups at baseline (Table 1). Women who had attended the antenatal training programme were more often in active labour on arrival for birth at the maternity ward than the reference group (Table 2). On arrival for birth, the two groups had similar ability to cope with fear in early labour based on their statement on the DFS. No differences were found in the two groups’ single items scores or sum scores on the DFS. Also, no differences were found between the intervention group and the reference group in numbers of outpatient’s visits or of telephone calls to the maternity ward in the time period from first contraction to the admission that ended with birth of the child. This result was also found when the time period was extended to the last month before the birth of the child (data not shown).
Table 2. Relative risks for obstetrics outcome according to the intervention (n = 1162)
Intervention (n = 587) n (%)
Reference (n = 575) n (%)
*P < 0.05.
**Induced labour excluded.
Cervix >3 cm on arrival at maternity ward
Cervix >3 cm on arrival at maternity ward**
Puncture of membranes
Use of pain relief
Use of non-pharmacological pain relief
Intracutaneous sterile water injection
Use of pharmacological pain relief
Nitrous oxide/oxygen (N2O/O2)
Pudendal nerve block
Other (primary halcion, codeine, paracetamol)
No differences in overall use of pain relief were found between the intervention group and the reference group (Table 2). Stratification of pain relief into pharmacological and non-pharmacological pain relief revealed a tendency towards less use of pharmacological pain relief in the intervention group. Further stratification into different types of pain relief showed that the women who attended the antenatal training used epidural analgesia less than the reference group (Table 2).
We found no significant differences in birth experience as reported by the women 6 weeks postpartum. Also, no differences in medical intervention or complications during the birth process were found, except for induced labour on the indication prolonged pregnancy (relative risk 0.72, 95% confidence interval 0.53–0.96, P = 0.03).
We found no adverse effects of attending the ‘Ready for Child’ programme for the women or for their newborns. Newborn outcomes such as growth, Apgar scores and biochemical status are presented in Table 3.
Table 3. Relative risks for newborn outcome according to the intervention (n = 1162)
Intervention group (n = 575) n (%)
Reference group (n = 587) n (%)
Birth weight <2500 g
Gestational age <37 weeks
Apgar score <7 at 1 minute
Apgar score <7 at 5 minutes
Standard base excess <12 mmol/l
The difference in cervix dilatation was statistically significant, even when a Bonferonni correction was applied to allow for the fact that the study had five primary outcomes. The other differences identified in the analysis were less pronounced when corrected for multiple comparisons.
Nulliparous women who attended antenatal training during pregnancy more often arrived at the maternity ward in active labour than the reference group but did not experience more fear of labour. There was no difference in the overall use of pain relief except for the use of epidural analgesia during labour, but it did not affect the women’s birth experience. No adverse effects were found for the women or their newborns as a result of attending the antenatal training programme. We found no differences in potential confounders at baseline between the intervention group and the reference group and, apart from the training programme, all women received the same antenatal care. A high compliance was present for the intervention group but around 45% in the reference group received some, but not comparable, antenatal training offered by stakeholders, mainly relaxation therapists. If these private programmes are of value, we underestimate the effect of our intervention in the intention-to-treat analysis. The high compliance for the women and their partners in the intervention group and the high rate of women in the reference group seeking antenatal training by stakeholders show that antenatal training is an attractive health promotion programme, which creates great opportunities also in a general public-health perspective.
Response rates on data collected from questionnaires were 89% or more, and medical birth data was obtained for 97%.
Nearly 50% more women in the intervention group arrived at the maternity ward in active labour compared with those in the reference group. The differences in cervix dilatation between the intervention and the reference group was 0.9 cm, equivalent to approximately 2 hours of labour for nulliparous women in the early phase of labour.15 We found no differences in the women’s ability to cope with fear on arrival at the maternity ward despite the differences in cervix dilatation. This may indicate that the women who attended antenatal training were more prepared and better able to cope in early labour at home.
A better ability to cope with labour may also explain the less frequent use of epidural analgesia in the intervention group. It is well-known that women who receive epidural analgesia experience more fear, but not more pain, before the administration of epidural analgesia,14 and they remain in labour for a longer time. A more advanced cervical dilatation when arriving at the maternity ward has been found to be associated with fewer obstetric interventions such as epidural analgesia.16 The later arrival observed in the intervention group in this study does not seem to be a threat to safety. This assumption is based on the results from the DFS, the information obtained from the birth process, and the newborns’ own outcomes. However, this study is underpowered to detect small differences. How antenatal training reduces the number of inductions in prolonged pregnancies is unclear. Our study size is not large enough to determine whether those attending antenatal training are better able to wait for the spontaneous onset of labour, which in turn could reduce the number of inductions for prolonged pregnancy.
We found no differences between the two groups concerning their reported birth experiences. Nine percent of the women in our study experienced a ‘very bad’ or ‘bad’ birth experience. This provides limited room for improvement and makes it difficult to detect any differences between the two groups. It is well-known that the birth experience largely depends on continuous support throughout the birth process17 and a positive attitude from the attending midwife and partner,18 conditions that were equally provided for both groups. Fifty-two percent of the women who experienced a ‘very bad’ or ‘bad’ delivery had an instrumental delivery or a caesarean section.
This study is to our knowledge, the largest randomised controlled trial to estimate the potential effect of antenatal training in preparing for the birth process.
Women attending the ‘Ready for child’ programme coped better in early labour, and used less epidural analgesia during labour. These results did not affect the women’s overall birth experience or the health of their newborn. Our results on the use of analgesia are similar to those published by Scott and Rose3 more than 30 years ago, but are in conflict with the results published by Fabian et al.8 An antenatal training programme offered to pregnant women by midwives is an important low-technology health promotion tool, and may be performed at low cost in most settings. Whether these results also apply to multiparous women requires additional studies.
Disclosure of interest
Contribution to authorship
RDM, MV and JO designed and initiated the trial. RDM co-ordinated the trial in collaboration with MV, JD, LH and JO. RDM analysed the data, and all authors participated in the interpretation of the results. RDM wrote the paper.
Details of ethics approval
This study was approved by The Regional Science Ethics Committees nr. 20060030 in Denmark and the Danish Data Protection Agency nr. 2006-41-6122.
This study is supported by grants from The Egmont Foundation, The Health Insurance Foundation, The National Board of Health, The Augustinus Foundation and The Danish Midwifery Association.
We are indebted to all the women and their families who took part in the project, to the midwives involved in the study, and to the staff at the maternity ward and at the Perinatal Epidemiology Research Unit at The University Hospital of Aarhus, Skejby, for their kindness in collecting the birth data. We are grateful to Klaas Wijma for his kindness in allowing us to use the Delivery Fear Scale. The results reported in this paper are related to the birth process. Endpoints related to the period after the birth will be considered in a future publication.
1Escott D, Spiby H, Slade P, Fraser RB. The range of coping strategies women use to manage pain and anxiety prior to and during first experience of labour. Midwifery2004;20:144–56.
11Redman S, Oak S, Booth P, Jensen J, Saxton A. Evaluation of an antenatal education programme: characteristics of attenders, changes in knowledge and satisfaction of participants. Aust N Z J Obstet Gynaecol1991;31:310–6.
Journal club for obstetricians, midwives and obstetric nurses
1 Background: Describe the evidence with regards to the advantages and disadvantages of antenatal classes, including but not limited to maternal satisfaction, physical outcomes, cost-effectiveness of birth care, and organisational measures of quality and safety.
2 Methods: The relevant section of an article should describe the methods used in sufficient detail to allow both readers to appraise the study and researchers to conduct similar studies. Comment on the description of the intervention; compare with similar interventions (classes) in your area.
Many women were invited to participate but refused or did not answer. Would it be useful to know more about their demographics and history? Why?
Define ‘intention-to-treat analysis’, and discuss its strengths and weaknesses.
This study addressed nulliparous women only. Describe the implications for methodology and interpretation of including parous women as well.
3 Results and implications: Discuss the differences in process and outcome of care between women in the two groups, with reference to the balance between medicalisation and safety.
A large number of women in the reference group (no classes) sought alternative antenatal training to compensate for not having classes. Discuss the implications for the analysis of the findings and for policies in your unit/area.
Compare the findings of this paper with previous studies.1,2 Are there any differences in the approach or the results, and why?
Discuss the implications of the findings for other pregnant women, researchers, healthcare professionals, healthcare managers, and policy makers. ▮
1Bergström M, Kieler H, Waldenström U. Effects of natural childbirth preparation versus standard antenatal education on epidural rates, experience of childbirth and parental stress in mothers and fathers: a randomised controlled multicentre trial. BJOG2009;116:1167–76.