National review of maternity services in England 2022 to 2024

The National maternity inspection programme has identified widespread issues affecting the quality and safety of maternity services in England.

In the programme, we rated 47% of services as requires improvement or inadequate. Many of our concerns are not new – in our Getting safer faster: key areas for improvement in maternity services report, we highlighted that maternity services stood out from other services as not making safety improvements fast enough. Similarly, our Safety, equity and engagement in maternity services report identified that issues such as poor relationships between obstetric and midwifery teams, and failure to engage with and listen to local women, continue to affect the safety of some hospital maternity services.

Throughout the programme, the safety of women using maternity services has remained a key concern. This is reflected in our ratings, as no service was rated outstanding for being safe. In fact, for the safe key question, the majority of services were rated as requires improvement (47%), while 35% were rated as good and 18% were rated as inadequate. Where we had the most concerns, we used our enforcement powers to require trusts to make significant improvements to protect people from risk of harm.

We found a range of issues affected how safe services were. These ranged from compliance with training requirements, particularly in key areas such as measuring babies’ heart rates and safeguarding, to how well services identified and managed the risk of deterioration in both women and babies. We also found concerns in relation to infection prevention and control in some services, with poorly maintained estates adding to their inability to provide safe care to women (see the Estates section of this report).

Throughout the programme, our inspection teams received high levels of challenge from some leaders working across the sector, which led to concern that poor care within maternity is being normalised. But all services must recognise the long-term, significant impact that pregnancy and birth can have on women. Many women told us about how their mental health had suffered before, during and after birth. In the UK, 4% to 5% of women develop post-traumatic stress disorder (PTSD) every year after giving birth and data from MBRRACE-UK shows that although extended perinatal mortality rates decreased across the UK in 2022, they remain higher than in both 2019 and 2020.

Many of the issues we highlight in this report present serious risks to safety, such as unacceptable levels of variation in key areas such as triage However, in this section we look specifically at the way services reported, learned from and communicated with women following incidents.

Incident reporting

Although most services managed patient safety incidents well, more work is needed in this area to ensure that where women suffer serious harm in maternity services do not go unreported and are graded correctly. Issues and inconsistencies around incident reporting were identified as concerns in Dr Bill Kirkup’s report on maternity services in East Kent.

We are concerned that a lack of reporting – either because of a recognised complication that the trust does not believe meets the definition of a patient safety incident or that staff are overstretched – is leading to harm becoming normalised and opportunities for learning being missed.

For most of the inspection programme, services were reporting incidents to the National Reporting and Learning System (NRLS) – a central database for all patient safety incident reports. NRLS defines a patient safety incident as “any unintended or unexpected incident(s) that could have, or did, lead to harm for one or more person(s) receiving NHS funded healthcare.” Towards the end of the programme, NHS England introduced the new Learn from Patient Safety Events (LFPSE) service and guidance, which has replaced the NRLS. This provides clearer definitions and distinguishes between physical and psychological harm.

Recognised complications may be common for staff and may not always meet NHS England’s definition of a patient safety incident, which means they do not always need to be reported to NRLS or LFPSE. However, these complications can have a significant and long-lasting impact on women, and trusts have a statutory duty to notify CQC of such events. Trusts can do this through LFPSE and should monitor and respond to trends in these commonly occurring obstetric complications at a local level.

Many services did not have this oversight of commonly occurring obstetric complications. We found that services often had to access several different dashboards to get an overall picture of patient-related outcomes, which could at times be contradictory and unclear. In addition, we found inconsistencies in how trusts managed key metrics such as blood loss. Despite available guidance, not all services were measuring blood loss in all deliveries. This risks potential under-reporting and could mean national dashboard comparisons are less meaningful for oversight and improvement. We also found some services did not report all incidents of delays in care and controls of postpartum haemorrhage.

Not reporting incidents at a local level suggests a tendency for services to accept that maternity incidents are inevitable and that nothing in a woman’s care or treatment may have contributed to them. But this is not always the case. Previous successful initiatives have shown that incidents such as shoulder dystocia, where a baby’s shoulder becomes stuck, can be preventable. For example, in 2000, North Bristol NHS Trust introduced simulation training to reduce shoulder dystocia. Since training was introduced, the trust believes that no babies have suffered permanent injuries. We also know that the likelihood and impact of postpartum haemorrhages can be effectively reduced with good antenatal monitoring of haemoglobin levels. Our concerns are reinforced by the recent Birth Trauma Inquiry, Ending the Postcode Lottery for Perinatal Care, which described a “maternity system where poor care is all-too-frequently tolerated as normal”.

While we recognise that postpartum haemorrhages (PPH) are not entirely preventable, services need to use evidence-based practice and guidance to optimise outcomes for women and acknowledge the impact that it can have on them. In addition, we know that women from Black and Asian backgrounds have an increased risk of PPH. Perinatal care for women from ethnic minority backgrounds should focus on preventative measures to optimise outcomes. However, as highlighted in our section on inequalities, not all services we inspected were monitoring outcomes by ethnicity.

Pressures on staff

When a patient safety incident occurred, most services managed this well in line with national guidance. However, we were concerned to find instances of patient safety incidents going unreported to NRLS because of time constraints. We found a significant number of incidents were not reported as staff were overstretched. Until more action is taken to ensure that incidents are recorded properly, and in a timely way, opportunities for improvement can be missed. Services rated as good and outstanding have a culture where incident reporting is encouraged, and feedback loops and improvement actions are normalised.

Maternity services tend to generate a significant number of incidents compared with other areas within a trust, and our inspection programme found that the size and make-up of governance teams were not always sufficient. Services often did not involve risk and governance managers, meaning midwifery staff were required to review incidents themselves. We were concerned about the impact of this on the quality and speed of reviews and the knock-on effect on staffing levels if midwives do not have protected time to review incidents. We consider this in more detail in the section on staffing.

Grading of incidents

As well as problems with reporting incidents, we are also concerned about variation in the way incidents were graded. The final report of the Ockenden review highlighted the importance of correctly grading patient safety incidents, ensuring the level of harm recorded reflects the actual harm the patient suffered.

NRLS states :

Maternity, fetal and neonatal incidents such as intrauterine deaths should be reported to the NRLS, however a degree of harm of death should only be chosen if it is considered that a patient safety incident, such as an omission in care during the antenatal period, has led to or contributed to the death. The degree of harm can be amended and re-uploaded to the NRLS after further investigation.

The way trusts and clinicians interpreted NRLS guidance on reporting incidents varied. While this variation exists, there is room for confusion, loss of learning and potential harm. Grading incidents based on whether omissions in care contributed to them, as outlined by the NRLS, does not take into consideration the actual physical and psychological harms that women experienced.

We saw evidence of this from incidents that are defined as ‘major obstetric emergencies’ (including uterine inversion and major haemorrhages over 2 litres) regularly being graded as no harm or low harm. Incidents graded as lower harm might mean opportunities to investigate and learn are missed. It could also result in no follow-up care or monitoring being organised, which may harm mothers and their babies. For example, one service used the perinatal mortality review tool, which showed an incident was graded less severely than it should have been. The trust originally highlighted that care issues ‘may’ have made a difference to the outcome for the baby, but a further review showed these issues were ‘likely’ to have made a difference.

We know that traumatic birth experiences can have a significant lasting impact on women and their families. Through our review, people told us about their experiences and the impact on them:

This experience [was] not one I wish to ever have to go through again, this will be my first and LAST baby. When I think of my birth experience and the aftercare, I cry every time, it was purely awful.

I am now undergoing therapy for PTSD... I find it incredibly traumatic to explain what happened in detail.

I would love another baby at some point but am emotionally scarred and find the whole ordeal difficult to talk about so this is something I am very concerned about.

LFPSE defines a patient safety incident as “something unexpected or unintended that could have or did lead to harm for one or more patients”. Recording guidance states, “if in doubt, it is always better to record a patient safety incident using the available information and best judgement”.

The guidance advises that where an unintended or unexpected outcome has been observed, but there is any uncertainty about whether an unintended or unexpected incident has occurred, the event should always be recorded to LFPSE to support national learning.

The new Patient Safety Incident Response Framework (PSIRF), introduced during the programme, moves away from the grading of incidents and prioritises compassion and engagement with the people involved in patient safety incidents. It also has a focus on improvement. We will monitor how trusts implement and use PSIRF in future inspections and ensure that harm and trauma are still given the appropriate consideration.

Our inspections found that even in a very defined system such as the previous serious incident framework, there was variation and under-reporting. Under PSIRF, providers should agree their incident response plan with their integrated care board (ICB). We will assess how trusts have done this, looking specifically at plans for maternity and neonatal services.

Investigating and learning from incidents

We expect leaders and staff to have a good understanding of service improvement, using processes to ensure that incidents are learned from. Leaders should encourage reflection and collective problem-solving.

While we found some pockets of good practice, the overall picture of how services investigated incidents was mixed. We were encouraged to find some services with midwives who specialised in learning from incidents and action plans being developed as areas were identified for improvement.

At some services, managers reviewed incidents potentially related to inequalities (see the inequalities section for more information). For example, one trust interrogated data to identify the impact of ethnicity on outcomes. Following this, the service recommended increased scanning for Pakistani women after data revealed they have a higher risk of having babies that are Small for Gestational Age (SGA). At another service, following a baby abduction incident, an abduction policy was implemented and security staff were employed. The service also introduced 2-hourly security rounds and a sign in register.

Our improvement resource provides more information on how services learned from incidents well. However, this good practice was not consistent across services. An investigation by MSNI into one service noted that staff did not acknowledge the needs of people with a learning disability using the maternity service.

We also found that delays in the reviewing process meant learning from incidents was slow-paced and learning was not always shared effectively with staff. Concerningly, in a small number of cases, it was not clear whether the service had produced any ongoing action plans or monitoring. In other instances, action plans were not up to date, or did not fully reflect the findings of the reported incident.

These issues expand beyond maternity services. A study published in the Journal of Patient Safety found that too often hospitals develop action plans with weak or ineffective interventions, which can fail to address key issues and result in significant gaps in translating investigations into meaningful improvement. It found plans typically included individual-focused interventions, even when problems were systemic.

Although we saw pockets of outstanding practice in many areas, there is a need to support trusts to adopt solutions that are working well in other maternity services. The lack of a system-wide approach to sharing learning is preventing maternity services from driving improvement by implementing strategies and interventions that work well elsewhere.

There are opportunities in the Patient Safety Incident Response Framework (PSIRF) to improve the way maternity services identify and embed learning from incidents through directing investigation resources towards incidents that they can learn most from. At one PSIRF early adopter site, we found the trust had created a continuous improvement and learning team that comprised midwives, patient safety and quality improvement practitioners. This team reviewed all incidents reported as moderate or above in the previous serious incident framework and identified learning opportunities. We welcome the increased focus on quality improvement and compassionate involvement of those affected by patient safety incidents.

Some serious events in maternity services have national requirements for reporting, such as intrapartum stillbirths and maternal deaths, which are reportable to the Maternity and Newborn Safety Investigations programme (MNSI). However, additional metrics for serious maternal morbidity outcome would improve oversight. These could include maternal admissions to the intensive therapy unit, returns to theatre, and maternal collapse.

Transparency and accountability

While recognised complications such as postpartum haemorrhages, obstetric anal sphincter injury (OASI), or shoulder dystocia do not always constitute a patient safety incident and may be recognised by staff as complications, it is vitally important to acknowledge the trauma experienced by the woman at the centre of each incident. Women need to understand what has happened to them, their recovery, and any potential impact on future pregnancies, but we are concerned that this does not always happen. Although research has identified improvement in this area, it shows there is still work to be done to make sure families are involved in investigations. Like other national reports, we heard through our Give feedback on care service that women did not always get a timely debrief or explanations of events, and this had had a negative impact on them.

Under the Health and Social Care Act 2008 Regulation 20: duty of candour requires providers to act in an open and transparent way. It aims to protect people’s right to openness and transparency from their health or care provider and encourages families to talk about their experiences openly and without fear as they begin healing. This can also help build people’s understanding of risk in future pregnancies. But the duty of candour only applies in certain situations, and we are concerned that when incidents are out of scope of the duty of candour, women do not always receive the debrief they need to process what has happened to them.

As well as the statutory duty of candour for all health and care providers, there is also a wider professional duty to be open and honest following incidents where the statutory duty of candour does not apply. The Nursing and Midwifery Council and the General Medical Council issued joint guidance on the professional duty of candour. The guidance is not intended for circumstances where a patient’s condition gets worse due to the natural progression of their illness. It applies when something happens with a patient’s care, and they suffer harm or distress as a result. There are opportunities to develop the principles of being open and honest with women in all scenarios, including after recognised complications of pregnancy.

We noted that in some trusts, staff can view potential complications as being normal – particularly during the intrapartum phase (during labour). However, we know from speaking to women who have experienced trauma that some of these ‘normal’ complications can have a significant impact. For example, although a grade 3 perineal tear may not warrant a patient safety incident, nor would it necessarily require the duty of candour to be instigated, it is vital that women still have the opportunity to discuss what happened, why it happened, and what it means for their future.

Through our Give feedback on care service, women told us about the impact of their traumatic birth experiences:

I'm still traumatised, developed high level of anxiety and obsessive thoughts…

However after the traumatic time… even now 3 months on I am very upset about this… the first few weeks of my baby’s life were marred by flashback.

I have been left with trauma. Worst experience of my life.

We also found that potentially serious incidents such as massive obstetric haemorrhage were normalised by many services if they perceived that they had ‘managed’ everything in line with guidance (generally either the Royal College of Obstetricians and Gynaecologists’ Prevention and Management of Postpartum Haemorrhage guidelines or the All Wales Maternity & Neonatal Network Guidelines on Prevention and Management of Postpartum Haemorrhage). Despite a number of services thematically reviewing incidents, we found this did not always translate into learning and improvement, such as a reduction in rates of PPH. In addition, even though services may have ‘managed’ an episode of haemorrhage well, a review and explanation of events would still be vital to help women to process their experience.

In many cases, managers involved women and their families in the investigation of incidents, which is a key part of incident response under PSIRF. We also heard about the importance of compassionate staff who provide people with clear information in a supportive setting:

The team have been incredibly kind with our questions and making the next steps very clear, which makes them less daunting… They've really validated our experience and helped us to feel like what we are going through matters.

We found examples of good practice where services applied duty of candour and issued letters in the first language of the family affected by the incident , but this was not always the case. We found evidence of inequality in how some services reviewed incidents. For example, in one service there were potential delays to the duty of candour process because the women involved did not speak English. It is vital that women are given the opportunity to be involved in investigations concerning their care. Not having English as a first language should not exclude people from being part of this important process.

At another service, we found good practice such as appointing a family liaison midwife to provide continuity of support throughout the process and auditing compliance with the duty of candour. But we also saw in a significant number of services that, although staff apologised following incidents, they were not always open and transparent with women and their families. Moreover, staff did not always provide clear information on the reason why things happened. Similarly, we identified occasions where women and families who were affected by serious incidents had not been involved in the investigation process, or their involvement was delayed. Through our Give feedback on care service, we have heard from women who are still waiting for answers and want to ensure mistakes are not repeated:

I had a traumatic labour which resulted in a uterine inversion. I was rushed to theatre to be operated on, could not bond with my new baby and had to have a blood transfusion… We have since been into hospital for a meeting to discuss what happened but I still have no answers and I was meant to be contacted to have another meeting with a midwife and all these months later I am still waiting. I think the service I received was absolutely atrocious in what should have been a wonderful experience. The surgeon was amazing and so were some of the nurses on the ward. I hope something will be done about the care I received as I know I am not the only one and I wouldn't recommend to anyone.

Through our engagement with families who have suffered a bereavement, we heard concerns about the lack of a complaint route, as services like PALS do not look into complaints where a patient has died. Family members also explained that policies and procedures following a loss can be left to staff to interpret, echoing our concerns around variation in the quality of follow up and communication. The families suggested that people affected by maternity failings should be involved in delivering training to midwives to ensure all families receive clear information and appropriate care in the future.

A pilot of Maternity and Neonatal Independent Senior Advocates started recently in England. The role has been introduced to support women and families affected by problems in maternity care. Maternity and Neonatal Independent Senior Advocates will help ensure that the voices of women and families are listened to and acted on. They will play an important part in ensuring women understand what has happened to them.