CCBYNC Open access

Perinatal and maternal outcomes by planned place of birth for healthy women with low risk pregnancies: the Birthplace in England national prospective cohort study

BMJ 2011; 343 doi: (Published 25 November 2011) Cite this as: BMJ 2011;343:d7400

Re: Perinatal and maternal outcomes by planned place of birth for healthy women with low risk pregnancies: the Birthplace in England national prospective cohort study

Dear Editor

The 'Birthplace' study in England is a detailed study producing a lot of interesting data and we commend the National Perinatal Epidemiology Unit for its work and for supporting new thinking surrounding the design of clinical services. Recently, the RCOG produced the blueprint 'High Quality Women’s Health Care' report which emphasised the importance of network arrangements for all aspects of women’s reproductive healthcare including maternity services.

This report supports our thesis on place of birth and emphasises the safety of childbirth in women with no pregnancy complications. We wish to highlight the following:

1. This is an English study on low-risk women, or as described, ‘women with no (known) complications in pregnancy’
2. The study distinguishes between nulliparous and multiparous women, and, indeed, the outcomes for these two groups are very different
3. The outcome for babies planned to be delivered at home to nulliparous women was less favourable than for other birth locations
4. Transfer for nulliparous women was up to 45% for mothers delivering at home or in midwifery units into obstetric units (OUs)
5. Outcomes for babies of multiparous women were no worse in any birth location
6. Maternal intervention rates were much higher in births planned for OUs with lower ‘normal births’ and spontaneous vertex deliveries.

Based on these findings, the RCOG advocates that first-time (nulliparous) mothers should be advised of the benefits of delivering in OUs or alongside midwifery units (AMU) unless geography prohibits. Subsequent deliveries in low-risk multiparous women can be conducted in any of the four locations (dependent upon geography, family choice and health economic considerations).

We have identified the following areas requiring more research:

Further work is needed to determine the cause of any adverse outcomes for mother. The intervention rate in OUs, even for this low-risk pregnancy group was high. One can only speculate on the causes of these differences (i.e. epidural analgesia uptake rates, differing protocols for the management of progress in labour and monitoring of the baby and differing experience of the healthcare professionals and the opportunity for consultant involvement). Initial intervention leads to further ‘clinical’ involvement and further intervention. This may have an impact on outcomes for both this and subsequent pregnancies both in human and financial terms. Equally, careful analysis of staffing ratios for both doctors and midwives by birthplace setting requires description. Home birth has a midwifery ratio of at least 1:1 and occasionally 1:2 midwives at the time of birth itself. This is very different from most hospital units which would be unable to demonstrate such ratios. Has the overall “busy-ness” of a unit an impact on maternal intervention with pressure on staff/beds/theatre time contributing to a need to achieve a delivery? Robust, prospectively collected information systems are essential to assist in such analysis. Research tells us that 1:1 care reduces the intervention rates and this could explain the differences and inform us of the improvements required. Improving midwifery numbers in OU may reduce the intervention rates rather than moving site of delivery.

More information is needed on the causes of adverse perinatal outcomes for the baby. The primary outcome analysis uses a composite aggregation and adjusts accordingly. However, in this low-risk population of mothers why were 20 of the 32 deaths in the home or FMU groups? Analysis of the secondary outcomes by parity and birth setting with appropriate adjustments for example of Apgar score <4, seizure rate and encephalopathy are necessary.

Finally, further work is needed into risk stratification and labelling. Why for example did 1 in 5 of the women in the OU group have complications at the onset of birth compared to the 5-7% of the other low-risk women?

We welcome the debate that such data will engender, to allow the focus on continuing improvements in maternity services within the United Kingdom. There requires wider realisation of the fact that preventative spend now to reduce intervention and improve outcomes will have a lasting impact for individual women but also for stretched NHS services both practically and financially for the future.

This study should not be used to declare what is bad, but what needs to be done to make it better.

Dr Anthony Falconer
President, Royal College of Obstetricians and Gynaecologists

Professor James Walker
Senior Vice President (International), Royal College of Obstetricians and Gynaecologists

David Richmond
Vice President (Standards), Royal College of Obstetricians and Gynaecologists

Professor Mark Kilby
President, British Maternal and Fetal Medicine Society

Competing interests: No competing interests

01 December 2011
Tony Falconer
27 Sussex Place, Regent's Park, London NW1 4RG