AIMS Journal, 2025, Vol 37, No 4
By The AIMS Campaigns Team
In September 2025, the National Maternity and Perinatal Audit (NMPA) belatedly published their ‘State of the Nation Report’ for 2023. This covers births in England, Wales and Scotland, but not Northern Ireland. On Thursday 13th November they published a ‘snapshot report’ on induction of labour. This uses the same data but goes into greater depth about maternal characteristics and outcomes in pregnancies where labour was induced.
In this briefing note, we highlight these reports’ key findings and raise some queries/ suggestions for further exploration.
The full report includes the audit’s key findings and recommendations. It includes a section of the ‘key results at a glance’ which is worth reading - a good effort has been made to ensure that it is easy to read.
The report includes data for:
520 643 births across 120 Trusts in England
45 316 across 14 Boards in Scotland
26 635 across 7 Boards in Wales
Total 592 594
The report, meant to be annual, has repeatedly been delayed due to “[u]nforeseen delays in the supply of this data from NHS England”. The most recent report before this one was for births in 2017/18, published in 2022. NMPA now plans to fill in results for the intervening years, but hasn’t said when this will happen. The gap in the data means that it isn’t possible to look at recent trends.
More up-to-date maternity statistics and the historic trends for some measures are available in the national reports for England , Wales and Scotland .However, these do not include all the measures that NMPA reports. See our previous Birth Activists Briefing: The latest UK Maternity Statistics: where to find them and what they can tell you for more about what the different reports include.
A set of summary results tables is available as a separate document. You can also explore the Trust/Board level data for each measure included in the State of the Nation report (site level data for 2023 is not yet available). You can also compare the figures from different Trusts and Boards for each measure. This provides some more detailed breakdowns of various indicators, for example by parity (number of previous births) or gestation (weeks of pregnancy at birth). It compares these with the national average and shows whether the figure is higher or lower than the expected range.
Note that data for individual Trusts and Boards are not the actual figures. NMPA applies a statistical ‘case-mix adjustment’ ” in order to make meaningful and fair comparisons between trusts/boards with different patient populations”. This adjusts for factors outside the control of a maternity service such as the mix of ages in their population. This gives a better measure of the performance of the different services.
There are some variations between the three nations, but overall the picture is similar. So for this briefing we are focusing on the overall figures for Great Britain.
Ethnic mix - Three quarters (73%) of those who gave birth in 2023 in Great Britain were white, 15% Asian, 6% Black, 3% mixed and 3% other.
Socio-economic status a quarter (25%) were from the most deprived section of the population (defined as the most deprived 20% of areas by Index of Multiple Deprivation) and 22% from the next most deprived.
The report notes that compared to the total population, those that gave birth were more likely to be from ethnic minorities and the more deprived groups.
Age groups - The majority (61%) were aged between 25 and 34, a fifth (20%) between 35 and 39, and just under 5% over 40.
BMI - About two-fifths (42%) were in what is classed as the ‘healthy range’ for BMI (18.5 - 24.5 kg/m2 ). A fifth (22%) would have been classed as obese (over 30 kg/m2 ) and 4% as severely obese (over 40 kg/m2 ) using the BMI approach. A small number (just under 3%) would have been considered underweight (under 18.5 kg/m2).
As the increasing medicalisation of birth is often attributed to birthing women being older or having a higher BMI than in the past it would be interesting to see how these rates have changed over time. Unfortunately they were not included in previous NMPA reports.
First and subsequent births - Just under half were first births (45.8%) and 31% followed a previous caesarean.
Medical conditions - Pre-eclampsia affected about 3% of pregnancies, placental disorders 1.4% and ‘amniotic fluid abnormalities’ 3.5%. 12% of the women who gave birth in this period had pre-existing diabetes and just under 1% pre-existing hypertension (high blood pressure).
The report does not include data about the type of birthplace (e.g. Freestanding Birth Centre, Alongside Birth Centre, Homebirth or obstetric unit (OU)). This means that the impact of reduced access to out of OU birthing options is not visible in the data.
Induction - The average induction rate was 33.9%. This is similar to the latest figures in the national data sets for England, Scotland and Wales.
However, the range between different maternity care providers was very wide. The lowest rate reported was 12% and the highest was 46.8%. As that may include a small number of very high or very low figures, NMPA also reports the interquartile range. This is the spread of the middle half of the results. Even using this the range is wide: 29.6 - 39%.
For first pregnancies the induction rate was 39.7% and for those who had birthed before it was 28.9%.
The snapshot audit of induction discussed below gives more detail of how induction rates and outcomes such as mode of birth vary between different groups of women.
Preterm birth (birth before 37 weeks of pregnancy) accounted for 6.3% of the total. Of these, 43.9% occurred spontaneously and 56.1% were iatrogenic (due to medical intervention). Again there was a wide spread of both spontaneous and iatrogenic pre-term birth rates. The authors comment that this may be partly because different units offer different levels of neonatal care, but they don’t think that can be the whole explanation. The proportion of pre-term births that followed a spontaneous onset of labour was much higher in Scotland (73.0%) than in England (40.9%) or Wales (46.3%). It would be interesting to understand the reasons for iatrogenic pre-term birth and why practice seems to be so different in Scotland.
Mode of birth - Compared with the previous report for 2018/19 the vaginal birth rate has fallen significantly (49.4% versus 60% previously.) Both planned and unplanned caesareans were up. Planned caesarean accounted for 16.4% of births versus 12.1% in 2018/19. Unplanned caesareans accounted for 23.1% versus 15.5%.
Caesarean rates are also reported for three of the ‘Robson Groups’. These are categories based on things like the number of previous births, whether the baby is head down and whether labour started spontaneously.
For Robson Group 1 (first pregnancy, single head down baby, after 37 weeks of pregnancy and spontaneous labour) the caesarean rate was 18%.
For Robson Group 2 (as Group 1 but with induction or planned caesarean) the rate was 56.0%. Frustratingly, because they are grouped together, we cannot tell for sure how many of these were planned caesareans and how many were unplanned caesareans following an induction. However, the detailed data shows that the planned caesarean rate for all first babies at term is only 10.8%. Also, the snapshot audit of induction of labour (see below) showed an unplanned caesarean rate for first-time mothers having an induction of over 40%. This suggests that the high caesarean rate for Robson Group 2 must be largely driven by inductions that end in unplanned caesareans. (However, note that the measure of induction in the State of the Nation report is based on single births between 37+0 and 42+6 weeks gestation, but the snapshot report includes births from 24+0 weeks. This means that the figures in the two reports cannot be compared directly.)
These are the average figures but there was variation between providers for Robson Groups 1 and 2.
Over 80% of Robson Group 5 (second or more pregnancy, single head down baby, after 37 weeks of pregnancy, one or more previous caesarean births), had another caesarean birth.
Vaginal Births After Caesarean (VBACs) are in decline. The report says that only about a quarter (25.8%) of those‘eligible for a VBAC tried for one in 2023 compared with two-fifths (40.0%) in 2016/17, and 38.0% in 2018/19. It’s not clear how they are defining ‘eligible’. Is that all women with a previous caesarean or only those considered ‘suitable’ for VBAC by their doctor?
The VBAC success rate has also fallen, from 60.7% in 2018/19 to 52.4% in 2023. As a result only 14.2% of those who had a previous caesarean had a VBAC. The range is from 11.9% to 16.7%. The authors comment that “the variation may be reflective of counselling, consideration of obstetric history and individual informed decision-making.” AIMS wonders how women are being counselled about the risks and benefits of their options for birth after a caesarean. Are carers becoming more risk averse and more likely to recommend a planned caesarean, or switch to a caesarean during a VBAC labour?
Perineal damage - Episiotomies were given in almost a quarter of vaginal births (24.4%). This seems very high compared to the vaginal births that were assisted with instruments (11.1%). It implies that about one in eight spontaneous vaginal births without instruments involved an episiotomy.
The rates of 3rd or 4th degree tears “has remained similar to previous reports” at just over 3%. This suggests that the OASI (Obstetric Anal Sphincter Injury) care bundle, designed to reduce obstetric anal sphincter injuries, has not had a noticeable impact. The bundle was piloted in 16 units in 2017/18, with an evaluation published in 2020. It has since been rolled out elsewhere, despite limited evidence of benefit and a number of concerns AIMS Commentary: the OASI care bundle debate | AIMS
Other - Severe blood loss (over 1.5 litres) affected about 3% of women, and 3% were readmitted to hospital within 42 days of the birth.
Three measures reflecting the care of newborn babies are reported.
Baby’s condition at birth - 1.45% of babies had an Apgar score (an assessment of wellbeing) under seven at five minutes of age, showing that the vast majority of babies are born in good condition. The percentage of babies with an Apgar score of under seven at five minutes of age was slightly higher after induction - see below.
Skin-to-skin contact & breastfeeding - In England, just under three-quarters of babies (73.4%) had skin-to-skin contact within the first hour after birth. Slightly fewer (72.6%) had any breastmilk at their first feed but there was insufficient data to report breastfeeding rates at discharge. Figures for skin-to-skin contact are not available for Scotland or Wales, but breastfeeding initiation rates were lower than in England: Scotland 63.1% and Wales 65.8%.
The authors make a number of recommendations including:
National and local level initiatives and campaigns should be targeted at improving rates of timely pregnancy booking. It’s good to see that the stakeholders they say should develop these initiatives include “women and birthing people and their families”.
Government health departments should incorporate the impact of the changing trends in maternity care and outcomes when reviewing and planning maternity services.
Maternity care commissioners and maternity networks should use the evidence of variation in care processes and outcomes identified in this report, and the results for their local populations, when working with their constituent units to identify opportunities for improvements in service provision and clinical practice.
There are also recommendations about reviewing the data definitions and the descriptions of care processes and outcomes, and optimising data quality.
As explained above, this is based on the same data as the ‘State of the Nation’ report, again with case-mix adjustment, but includes a deeper dive into maternal characteristics and outcomes for mothers who had their labours induced. It covers all mothers expecting a single baby who underwent induction as long as the labour was expected to result in the birth of a live baby.
Confusingly, the authors define induction of labour as “the process of artificially starting labour by the softening and opening of the cervix and/or breaking the amniotic membranes” - seeming to exclude the use of oxytocin - but then list oxytocin infusion as one of the methods used for induction. They do not mention the use of membrane sweeps.
We already knew about the wide variation in induction rates from previous ‘State of the Nation’ reports as well as the latest one. This more detailed report shows that there is also a huge spread in the rates of unplanned caesarean births. Induction rates varied from 16% to 42% with a mean of 32%. Unplanned caesareans following induction varied from 18% to 45% with a mean of 30%, and 40% of Trusts and Boards had unplanned caesarean rates following induction that were higher or lower than the expected range.
It would be very interesting to know whether there is a correlation between the induction rate at a unit and its rate of unplanned caesareans. It should not be difficult for NMPA to carry out such an analysis.
An audit such as this can only tell us what was happening, not why, but as the data has been adjusted for clinical and demographic characteristics, it seems likely that at least in part the variation reflects differences in policy and practice. One of the recommendations of the report is that “Maternity care commissioners should undertake a structured review to identify the drivers of practice variation in IOL care within their networks, such as clinical culture, local policies and protocols and clinical leadership, to target a reduction in unwarranted variation in IOL care processes and outcomes.” AIMS feels that, whilst this may be helpful, if carried out with sufficiently robust challenge of local policies and practices, we would like to see a large-scale research project to examine when the offer of induction is and is not appropriate. We note that the second question on the Royal College of Midwives list of the top ten research priorities Research Prioritisation Project - Royal College of Midwives was “What are the appropriate reasons for Induction of labour? What are the short and long term maternal and baby outcomes associated with it? How should this be communicated to women and birthing parents and their informed consent gained?” This shows how important this issue is to both women and midwives.
The report notes that “IOL was recorded as unsuccessful for 6% of women and birthing people”. However it’s hard to know how accurate this is since they also note the lack of a standardised definition, and that there could be problems with data coding and completeness. The authors note that “the literature agrees that concluding an IOL has been unsuccessful should not occur before oxytocin is administered” but in clinical practice it is often recorded as unsuccessful before this. They also note that a ‘successful’ induction could still be followed by an unplanned caesarean later in labour. Would women class this as a success?
AIMS thinks that ideally the data should reflect what proportion of labours start, and what proportion progress to a vaginal birth after a) cervical ripening alone b) breaking of the waters (with or without cervical ripening) and c) after oxytocin administration.
The report includes a chart showing how a mother’s chances of having an unplanned caesarean after an induction vary according to her age, ethnic group, level of deprivation (as measured by the Index of Multiple Deprivation), pregnancy history and gestational age at birth. It is not easy to read what the mean figures on these charts are and it’s frustrating that despite a suggestion from AIMS at the draft stage, NMPA have not included the mean rates in the text. The figures below are estimated by eye.
First time mothers have a very high chance of an unplanned caesarean (over 40%) but for those who had birthed before without a caesarean the rate was only just over 10%. In contrast, almost half of mothers being induced after a previous caesarean had an unplanned caesarean.
Of particular concern is the finding of yet another disparity in outcomes for different ethnic groups. Compared to white mothers, those from all other ethnic groups were more likely to have an unplanned caesarean, and the rate was significantly higher for Black mothers (almost 40%) and Asian mothers.
Caesarean rates were also higher for mothers who gave birth before 37 or after 41 weeks of pregnancy, and increased with increasing age of the mother - as high as 40% for those aged 40 to 44, and almost 50% for those aged over 45.
It is clear that a mother’s characteristics make a difference to her chances of having an unplanned caesarean following induction of labour. Unfortunately, the report lacks detail that might help us understand the reasons for this. Are there underlying differences between the groups or is there a tendency for doctors to intervene more readily in what they regard as ‘higher risk’ labours?
Another of the report’s recommendations is for future versions of the maternity datasets to support recording of other factors “such as gestational age at induction, maternal decision-making, the indication, method(s) and duration of induction.” To this AIMS would add the need for research on how factors such as the reason for the induction, the gestation at which it is done and the method of cervical ripening used affect the chances of failed induction, fetal distress and unplanned caesarean birth, for both first-time mothers and those that have birthed before.
The authors quote the received wisdom that “Evidence from a systematic review, meta-analysis and randomised trial demonstrates that IOL is associated with a decreased likelihood of a caesarean birth, and does not increase the chance of giving birth with the use of instruments.”
AIMS questions whether the findings of randomised controlled trials are a good guide to what happens in the real world (See our Birth Information webpage Induction and the chance of a caesarean: what’s the evidence?) and the NMPA’s own data suggests otherwise. In this report the unplanned caesarean rate for first-time mothers having an induction is over 40% but the State of the Nation report says that for first-time mothers with a single baby in spontaneous labor (Robson Group 1) it was 18% - so it’s more than twice as high after induction. It’s frustrating that, despite AIMS’ suggestion, NMPA have not included a comparison of outcomes in spontaneous and induced labours, though they presumably have the data to do this.
The report also explores how maternal characteristics affect the baby’s wellbeing, as measured by the chance of a baby having an Apgar score under seven at five minutes after birth.
In the State of the Nation report this was the case for 1.45% (about 15 in 1000) of all babies. In the Induction snapshot the average figure was 1.59% (16 in 1000). As with the induction and unplanned caesarean rates there was wide variation between Trusts and Boards. The rate was much higher in Scotland (2.53%) than in England (1.48%) with Wales in between (1.9%).
Babies born to Black mothers were more likely than those born to white mothers to have an Apgar score of under seven at five minutes following induction (about 2.5%). However, those from other ethnic groups were less likely to do so.
The report does not examine how these rates vary between babies born by caesarean or vaginally following induction.
The authors note that a number of key questions remain unanswered due to data being lacking or of poor quality. “These questions include:
How often do women and birthing people decline IOL?
What are the most common indications for induction of labour?
Which methods of induction are most frequently used?
What is the average duration from commencing induction to being transferred to a labour suite?
What proportion of IOL are unsuccessful?”
AIMS would also like to see the data being available to answer these questions, to which we would add:
How often are women pressured into accepting an unwanted induction?
How many mothers opt for a planned caesarean rather than an induction?
How do factors such as the reason for the induction, the gestation at which it is done and the method of cervical ripening used affect the chances of a) failed induction b) unplanned caesarean birth and c) low Apgar score, for both first-time mothers and those that have birthed before?
How do the outcomes of failed induction, unplanned caesarean and low Apgar scores influence women’s satisfaction with their experience?
Is there any correlation between the length of induction and/or the indication for induction and the chances of a caesarean or of a low Apgar score?
Is there any correlation between failed inductions/unplanned caesareans and low Apgar scores?
We think that NMPA already has the data to answer further questions, and will ask them to look at further analysis to investigate:
Do units with a higher induction rate also have a higher rate of unplanned caesareans following induction, and vice versa?
How do outcomes such as unplanned caesarean rates and low Apgar scores differ between spontaneous and induced labours, for first time mothers and for those who have birthed before, with or without a previous caesarean?
How do Apgar scores at five minutes differ between babies born vaginally and those born by caesarean following an induction?
The NMPA ‘State of the Nation’ report contains information on a range of important measures, including a number that are not available in other maternity statistics publications. The authors have made commendable efforts to present the key figures in an accessible format, and the whole report is well worth a read.
It is unfortunate that the data is not as up-to-date as we would wish. This is the ‘State of the Nation’ as it was two years ago and things may well have changed. For a number of measures, more recent Trust level data is available for England in the National Maternity Dashboard
It's also unfortunate that gaps in publication make it harder to understand the trends. Hopefully the team will be able to fill these gaps before too long.
It is somewhat confusing that we have a variety of sources reporting national maternity statistics, that they include different measures, and that they are published with different timings and presented in different ways. There must surely be scope for streamlining this information and creating a single clear, accessible and up-to-date source for all the statistics that service providers, service users and birth activists need.
Meanwhile, AIMS has identified some specific queries which we will raise with the NMPA in our role as members of their Clinical Review Group.
The Induction of Labour Snapshot Audit adds some valuable detail to our picture of the variations in rates of induction and the outcomes for mothers and babies following induction. The authors note the lack of data and/or poor data quality that leaves some questions unanswered. AIMS hopes that their call for more extensive data to be collected will lead to action. Meanwhile, we feel that there is more that could usefully be done with the current data.
Key points worth noting include:
These reports confirm the finding in other data sets that around a third of labours in Great Britain are being induced. The rate is even higher for first births, at around 40%.
There is wide variation between providers in their induction rates, and to a lesser extent in their caesarean rates for Robson Groups 1 and 2.
There is also wide variation in unplanned caesarean births and of babies with an Apgar score under seven at five minutes following induction of labour. Rates are particularly high for Black mothers.
There is wide variation in both spontaneous and iatrogenic pre-term birth rates in different locations.
VBAC rates - both the percentage of women planning a VBAC and for VBAC success - are declining despite there having been no change in the evidence base or the guidelines.
There is no sign that the OASI care bundle is reducing injuries, several years after the roll-out began. It will be interesting to see whether this changes in coming years.
Questions to ask your local Trust or Board
What differences in policy or practice might be driving their rate of inductions and the outcomes following induction if these differ from the national averages?
Are women being provided with local information about the rates of unplanned caesareans and low Apgar scores that is specific to their own characteristics?
What is your local VBAC rate? What proportion of women plan a VBAC and what proportion of these are successful? How are the benefits and risks of this being presented to women?
The AIMS Journal spearheads discussions about change and development in the maternity services..
AIMS Journal articles on the website go back to 1960, offering an important historical record of maternity issues over the past 60 years. Please check the date of the article because the situation that it discusses may have changed since it was published. We are also very aware that the language used in many articles may not be the language that AIMS would use today.
To contact the editors, please email: journal@aims.org.uk
We make the AIMS Journal freely available so that as many people as possible can benefit from the articles. If you found this article interesting please consider supporting us by becoming an AIMS member or making a donation. We are a small charity that accepts no commercial sponsorship, in order to preserve our reputation for providing impartial, evidence-based information.
AIMS supports all maternity service users to navigate the system as it exists, and campaigns for a system which truly meets the needs of all.