Article Text

other Versions

Randomised controlled trial
Psychoeducation for pregnant women with fear of childbirth increases rates of spontaneous vaginal delivery, reduces caesarean rates and improves delivery experience
  1. Eileen K Hutton1,
  2. Wendy Hall2
  1. 1Department of Obstetrics and Gynecology, McMaster University, Hamilton, Ontario, Canada;
  2. 2University of British Columbia, School of Nursing, Vancouver British Columbia, Canada
  1. Correspondence to : Dr Eileen K Hutton, Department of Obstetrics and Gynecology, McMaster University, Midwifery Education Program, 1280 Main Street West, MDCL 2210, Hamilton ON, Canada L8S 4K1; huttone{at}

Statistics from

Commentary on: OpenUrlCrossRefPubMed

Implications for practice and research

  • The findings have the potential to increase care providers’ awareness about possible approaches to decreasing women's childbirth fear.

  • Prior to widespread implementation study findings should be replicated using a more methodologically sound approach that increases compliance with the intervention and avoids the Hawthorne effect, and takes anxiety into account.


Childbirth fear among the population of pregnant women in developed countries has been positively associated with poor birth outcomes and the increased likelihood of caesarean section.1 The rising proportion of women giving birth by caesarean section and the recognition of the increased risk of maternal morbidity associated with this mode of birth, has led to increased interest in interventions that may contribute to avoiding caesarean birth. Fearful women have reported a preference for caesarean birth, and being less positive about being pregnant and their approaching births.2


The researchers aimed to compare number of vaginal deliveries and delivery satisfaction among nulliparous Finnish women with fear of childbirth who were randomised to a psychoeducational intervention or to usual care. Over a 2-year period, 12 000 women were potentially screened for severe childbirth fear, and 371 meeting the criteria were randomised in a 1:2 ratio to a group therapy intervention (n=131) or usual care (n=240). The intervention comprised six educational sessions during pregnancy and one postbirth.


In the intention to treat analysis, the study found a significantly and clinically important reduction in the rate of caesarean birth in the intervention group. The authors also performed an analysis of those compliant with the intervention, compared with those who received additional therapy in the control group. The intention of this additional analysis was to consider those in the treatment group who took advantage of the treatment, compared with those who sought additional treatment among the control group.


This study represents an important step towards evaluating interventions with the potential to modify childbirth-related fear, lowering the likelihood of caesarean birth. The strength of this study is in the randomised controlled trial methodology.

There are several study limitations that may invalidate the conclusions. The authors were challenged to limit loss to follow-up and to ensure protocol compliance. The design does not account for the Hawthorne effect (or attention bias): the treatment group has six small-group counsellor sessions but the control group receives usual care, without additional small-group sessions. Additional attention regardless of content may account for differences between groups. The sample size was calculated with successful vaginal delivery as the outcome, but the outcome is reported as ‘spontaneous vaginal birth’. Eliminating vacuum extraction births strengthens between-group differences in the reported primary outcome; but this was not the prespecified primary outcome. The treatment group emergency caesarean rate seems low for nulliparous women (12.2%), and much lower than the 20% used for sample size calculations. Another limitation was that Rouhe and colleagues did not measure anxiety that has been highly associated with childbirth fear.3 Anxiety may have served as an effect modifier, limiting the intervention's impact. Claims that the findings can be generalised to all western countries do not acknowledge the lack of evidence linking caesarean section and fear of childbirth in studies from the UK and Canada.3 ,4

The finding that most undermines study confidence stems from the perprotocol analysis. Comparing the treatment group results by intention to treat and perprotocol reveals that outcomes for the 90 women who received the intervention are almost identical to those of the whole treatment group. Of the 90 women who were compliant with the intervention, 23.3% had a caesarean section; among the treatment group as a whole, which included women who were compliant with the treatment (n=90) and those who were not (n=41), only 22.9% had a caesarean section. If the treatment was truly effective one would expect that the caesarean births among those compliant with the treatment to be lower than among the whole group which includes nearly 30% who did not participate in the treatment at all. These findings create significant misgivings about effectiveness of the treatment and the study results.

View Abstract


  • Competing interests None.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.