The rates of induction of labour (IOL) have been increasing across the UK and prompting attention within maternity care. The National Maternity and Perinatal Audit (NMPA) produced a State of the Nation report on maternity services in September and have followed up with a snapshot audit specifically around IOL. Dr Amar Karia, Obstetrics clinical fellow, for the NMPA, explains the report and discusses the patterns between IOL and caesarean birth, with particular concern related to disparity across services.
Introduction and Background
Following the recent publication of our State of the Nation report in September 2025 (based on 2023 NHS birthing data from Great Britain), the National Maternity and Perinatal Audit (NMPA) team are proud to announce their next publication, focusing on Induction of Labour (IOL).
The Induction of Labour Snapshot audit expands on the results from previous reports, for the first time focusing on one of the most common interventions in maternity care. Successive NMPA State of the Nation reports have identified a pattern of increasing induction of labour rates, but also widespread variation in IOL rates, despite adjusting for factors such as maternal age, parity and previous caesarean birth.
The below table summarises the trends in IOL rates and variation in IOL rates between maternity care providers identified in our reports:
*Data from 2020-2022 has not yet been published due to considerable data delays from NHS England, and prioritisation of the most contemporaneous data. These data will be published on the NMPA website in due course.
** Does not include Scottish Data, due to a temporary exclusion of Scottish data from all national audits commissioned by the Healthcare Quality Improvement Partnership (HQIP) for this period.
***Data of sufficient quality and completeness from England for 2017-2018 period was only available for 51 English trusts due to dataset changes, meaning that total rate may not be fully representative of GB practice in 2017/18.
Funnel plots for IOL on the NMPA website (click here for a video on how to interpret funnel plots) show that across each of these reporting periods, there is a pattern of overdispersion, with the vast majority of trusts and boards having IOL rates more extreme than the limits of expected variation. This means that despite statistical models controlling for factors such as gestational age, parity, and previous caesarean birth, the complexities underlying these patterns are unexplained. Furthermore, these findings raised many other questions, including what happens to women and birthing people following IOL.
Induction of labour is typically offered when the perceived maternal or fetal risk of adverse outcomes outweighs the potential benefits of ongoing pregnancy, the risks of the induction process and potential risk of fetal development associated with an early birth. As advised by national guidance, general discussions about mode of birth preferences should be started early in pregnancy, and specific discussions around an offer of IOL should be patient centred, allowing the woman or birthing person time to speak with their partner, consult written materials and consider alternatives.
The decision whether to accept an induction of labour may be a challenging one for women and birthing people, and is often associated with giving up control from a carefully considered birthing plan to a more interventional and medicalised birth. There is strong evidence from meta analyses of both observational studies and interventional studies, which conclude that IOL does not increase the likelihood of a caesarean or instrumental birth. However, IOL is commonly longer than a spontaneous labour and will generally require more vaginal examinations than a spontaneous labour. The NHS recognises that an induced labour may be more painful than a spontaneous labour, and women and birthing people undergoing induction may be more likely to request an epidural than a spontaneous labourer. It’s no surprise therefore, that academic literature points towards IOL being frequently associated with a negative birth experience, although there are many components of the induction process and birthing care which may contribute to this.
Coupled with media scrutiny in 2021 arising from a draft consultation for NICE’s updated IOL guidelines, these factors inspired the commissioning of the IOL snapshot audit. Using the same 2023 GB data used in our recently published State of the Nation report, more than 170 000 records of induced labour were identified. This report focuses on all singleton pregnancies undergoing IOL from 24+0 weeks gestation which are expected to result in a live birth.
Analysis summary
The report has three main sections. Firstly, the report compares the characteristics (maternal age, country of the birth, gestation age, ethnicity, deprivation measured using quintiles of the Index of Multiple Deprivation (IMD), and past obstetric history) of women and birthing people undergoing IOL compared to the overall maternity population. Secondly, the report looks at the mode of birth following IOL (split into caesarean birth versus vaginal birth) and baby outcomes following IOL measured by 5 minute Apgar scores (a low score of less than 7, versus an Apgar score of 7 or more). A statistical model is used to examine trust and board variation in both outcomes, after performing case-mix adjustment (for maternal age, gestational age, parity and previous caesarean birth). Finally, the report looks at association between maternal characteristics and the aforementioned outcomes following IOL, using another statistical model to examine the association between individual characteristics and the outcome, when controlling for all other factors.
Overview of key findings and recommendations
Some of the key findings from the report are as follows. We identified that 30% of women and birthing people experienced a caesarean birth following IOL, with this rate varying widely between maternity care providers (IQR 26%–33% – the interquartile range (IQR) shows the middle 50% of results, showing the spread of data values around the median), with 40% of trusts and boards having a rate of caesarean birth following IOL more extreme than the limits of expected variation.
Whilst any pregnancy may conclude with a caesarean birth (for a myriad of reasons), the fact that on average 30% of IOL will result in a caesarean birth is an eye-catching finding. It is unclear how many women and birthing people are made aware of the chance of caesarean birth when counselled for induction of labour. This was recognised by our lived experience group, who helped with the interpretation of the data and co-development of a lay summary. As identified by our lived experience group, if women and birthing people possessed this information during the consent process, it is possible mode of birth plans may change for some. Counselling around IOL should also cover the reason for induction, recommended method(s) of induction, the expected duration and the possibility of other situations including uterine hyperstimulation and unsuccessful induction of labour as well as alternatives. This led to our recommendation for the use of local and national data during counselling of women and birthing people.
However, it is important to note that 30% is the average for Great Britain. Based on individual characteristics, the chance of caesarean birth following IOL may be higher for some, and lower for others. Whilst there was no meaningful difference in rates due to country of the birth, an IOL performed at 34–36 weeks gestation (just over 2% of IOL performed in Great Britain are performed before 37 weeks gestation), or after 40 weeks gestation were more likely to be associated with a caesarean birth than an IOL at 37–39 weeks of gestation. Differences in mode of birth following IOL were weakly related to deprivation (measured by IMD), with those from the most deprived quintiles (Q4 and Q5) being slightly more likely to experience caesarean birth following IOL than women and birthing people from the least deprived quintile (Q1). The factors with the clearest association with mode of birth were: age (women and birthing people with an age of 40 years or more were twice as likely to experience a caesarean birth following IOL than those aged less than 20), parity (women and birthing people having had a previous vaginal birth had a clear association with a vaginal birth following induction), and ethnicity, with women from Black, Asian, Mixed and Other ethnic groups being more likely to experience a caesarean birth following IOL than a white woman.
Regarding Apgar scores following IOL, 1.59% (IQR: 1.08%–1.97%) of babies born after IOL in Great Britain had an Apgar score of less than 7 at 5 minutes. These rates differed between the countries of Great Britain, with rates of low Apgar scores at 5 minutes higher in Scotland (rate: 2.53%, IQR: 2.09%–3.35%) than England (rate:1.48%, IQR: 0.99%–1.69%) or Wales (rate:1.90%, IQR: 1.10–2.24%). A similar pattern has been identified in the recently published State of the Nation Report, suggesting possible differences in Scottish data coding, or Apgar classification practices. Otherwise, the rates of low Apgar scores following IOL are within expected limits of variation for most trusts and boards.
While there was no relationship between maternal age and low Apgar scores, and a weak relationship between quintiles of higher deprivation (Q4 and Q5) and higher rates of low Apgar scores compared to the least deprived (Q1), there were considerable differences in the pattern between low Apgar rates after IOL and different ethnic groups.
The challenges of applying the Apgar scoring system to babies from minority ethnic backgrounds has been examined by the NHS race and health observatory. Following induction, babies born to women and birthing people from Black ethnic groups were more likely, and babies born Asian, Mixed or Other ethnic groups were less likely, to be assigned an Apgar score of less than 7 at 5 minutes than babies born to white women and birthing people. We therefore recommended that maternity services commissioners conduct structured reviews to understand why outcomes following IOL differ between care providers and different population groups, with an aim of trying to reduce variation in outcomes.
IOL data gaps in centralised maternity datasets
During this audit, we identified that there is lots of important information which would be collected in hospital records that does not pass downstream to centralised maternity datasets. Specific to induction, centralised data does not tell us important things such as the reason for induction, gestational age at induction, the method(s) of induction, and the duration of induction. For some elements, data is captured but the there is a lot of missing data, which really hampers interpretation. For example, our analysis suggests that 6% of inductions are unsuccessful. However, with trust and board rates varying between 0-20%, it is difficult to say how accurate this estimate is. The poor data quality is further impacted by the absence of a clear and agreed definition for what counts as an unsuccessful IOL. Further recommendations made in the report are targeted to maternity data controllers and software developers to improve the data flow into centralised maternity datasets for these key items, and to the Royal College of Obstetricians and Gynaecologists, to bring together stakeholders to develop a definition for unsuccessful induction of labour.
Further output from this report includes trust and board level results for their rates of caesarean birth and low Apgar score following induction of labour. Our report also contains a list of unanswered questions from national datasets related to induction of labour care, along with a list of proposed research priorities.
Conclusion
Induction of labour remains one of the most frequently performed procedures in maternity care in Great Britain. The majority (70%) of women and birthing people undergoing IOL will experience a vaginal birth. Despite this, outcomes and the experience for women and birthing people undergoing the same process in NHS maternity services may be very different. It is hoped that this information will allow care providers to reflect on their local data in comparison to national averages and use these data to support local quality improvement projects, and improved counselling. Midwives can use our report and lay summary to ensure that women and birthing people in their care have access to the information that will best support informed decision making around induction of labour. We hope this report draws attention to differences in IOL care across Great Britain and is a step towards improving care and reducing disparities in outcomes between different groups.
If you have any questions about the report, please contact nmpa@rcog.org.uk.
Amar Karia
Obstetrics clinical fellow, NMPA
November 2025