Birth Small Talk

Talking about birth

Assessing the safety of hospital birth

The Cochrane review comparing birth in hospital with birth at home has been updated recently. The update was well due, with the previous version published in 2012. The authors, Ole Olsen and Jette Clausen were also responsible for the 2012 version. While the evidence from randomised controlled trials has not changed, what has changed substantially is the way the authors looked at the question.

What is the intervention?

The first (1998) version of this review was titled “Home versus hospital birth”, with giving birth at home considered as the intervention and hospital birth as the control. While the title changed in 2012 to “Planned hospital birth versus planned home birth” the view that hospital birth was the intervention, and it was therefore the effectiveness and safety of hospital birth that was in question, not home birth, was not fully reflected in the body of the review. That shift in thinking has been clearly addressed by the authors in this version.

I think this is the correct framing. It mirrors discussions about intrapartum fetal heart rate monitoring where I often see arguments that we should keep using the common approach of CTG monitoring until such time as we have sound evidence that intermittent auscultation is clearly better. What impact would it have if instead we argued we should use intermittent auscultation for all women until we have evidence that CTG monitoring (and all the added bits and bobs that go along with it, like fetal spiral electrodes, central monitoring, and computer interpretation) had been conclusively proven to be the superior option? I suspect it would move the research agenda along at a far faster pace than has been the case to date.

Who should read this review and why?

If you ever have the need to write something (a university assignment, a research proposal, a policy document) to persuade someone of something, while using evidence to support your argument, then a close reading of the structure of this review will provide you with inspiration for how to go about it. Olen and Clausen step through the argument that defines leaving home and going to hospital as the first intervention in labour, and explore the common interventions that women typically encounter during their hospital birth (including the use of CTG monitoring) and the potential consequences of these. While they are far from the first authors to make this line of argument, they do it particularly well because of their strong use of research to support their logic line.

The other well laid out argument in this review is that, when it comes to interventions aiming to modify outcomes that are rare or longer term, randomised controlled trials may not be the best option. The authors calculated that to reliably show a 50% difference (if one exists) in maternal mortality, a randomised controlled trial about place of birth would need a sample size of one to two million women. This is simply unachievable.

As place of birth and intrapartum fetal heart rate monitoring share similar outcome sets (perinatal mortality and long term neurological damage for example), the argument in support of other research approaches applies well to both. Rather than abandoning the structure, rigour, and authority of the Cochrane review in the process of moving away from only considering evidence in randomised controlled trials, Olsen and Clausen argue for an expansion of what is considered valid evidence within a Cochrane review. They then use the background section of the review to provide that evidence for hospital birth.

And of course – if you are simply interested in keeping up to date with the evidence about the safety of hospital vs home birth, then the review is also worth reading.

What is the evidence about hospital birth from randomised controlled trials?

Only one randomised controlled trial was identified as meeting the selection criteria. This was a TINY trial, conducted in 1998, designed to assess the feasibility of a larger one. Eleven multiparous women, all considered low risk, were enrolled. Six women were allocated to hospital birth, with five expressing their disappointment at this allocation. Four of the five allocated to home birth gave birth at home. All women had vaginal births with good outcomes. The most important thing learned from the trial was that in an area where the home birth rate was low, women were prepared to have their intended place of birth randomly selected.

What is the evidence about hospital birth from observational studies?

Two recent and comprehensive systematic reviews formed the source of evidence for this section of the Cochrane review, drawing data from 16 original studies. The findings when comparing intended hospital birth with intended home birth for low risk women were:

  • No maternal deaths were reported in any trial
  • No difference in perinatal mortality
  • No difference in Apgar scores of less than seven at five minutes of age
  • For children born to nulliparous women, no difference in admission to neonatal intensive care
  • For children born to parous women, 20% higher rates of neonatal intensive care nursery admission with planned hospital birth
  • Significantly higher (at least 50% increase) rates of postpartum haemorrhage, infection, oxytocin augmentation, epidural analgesia, and episiotomy with planned hospital birth
  • Significantly higher (around 20% increase) rate of caesarean section with planned hospital birth
  • Use of continuous intrapartum CTG monitoring was three times higher, and the rate of use of a fetal spiral electrode doubled with planned hospital birth

In summary

Olsen and Clausen summarised these findings, saying “although the specific estimates varied between studies, the pattern of [hospital birth] doing significantly more harm than good was consistent” (p. 8). Given that the provision of home birth services in high income countries has been a contentious issue within maternity services and politically for the last century, I don’t expect that the new review will generate a sudden shift in availability. Like routine CTG use, the forces that maintain hospital birth are not primarily driven by evidence.


Olsen, O., & Clausen, J. (2023). Planned hospital birth compared with planned home birth for pregnant women at low risk of complications. Cochrane Database of Systematic Reviews, 3, CD000352. DOI: 10.1002/14651858.CD000352.pub3.

Categories: CTG, EFM, New research, Obstetrics, Perinatal mortality

Tags: , ,

3 replies

  1. “Like routine CTG use, the forces that maintain hospital birth are not primarily driven by evidence.“


    What hope do we have then!
    Brooke Jones
    Midwifery Student

    [Description: email_logo.png]


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s