National State of Patient Safety 2024:
Prioritising improvement efforts in a system under stress

About this report

This report presents the national state of patient safety in England in 2024. It includes an updated analysis of the publicly available patient safety data, to show how performance in patient safety has changed since our first report was published in 2022. This report also includes an updated timeline of developments in patient safety policy and practice and a review of progress against our recommendations.

The special focus of this report is priorities for patient safety. We conducted a series of activities to understand patient safety priorities from the perspective of different stakeholders – patients and the public, health and care workers, care providers, and national bodies. The aim was to explore how safety improvement efforts can be better targeted, in the context of an NHS under significant pressure. Our work included:

• Review of publicly available information from government and national bodies on national patient safety priorities.
• Desk research to identify opportunities to use other data and intelligence to prioritise patient safety issues.
• Analysis of published Patient Safety Incident Response Plans (PSIRPs) to identify local NHS organisations’ patient safety priorities.
• A poll commissioned from YouGov asking health and care workers and the public to rank patient safety priorities, followed by a stakeholder workshop to discuss the findings.

The conclusions of this work, along with our recommendations, can be found in Part 4.

The UK as a whole performs about average on rates of deaths from treatable causes, such as sepsis and blood clots, when compared to OECD countries (using 2022 data for the UK, as England-level data is unavailable). If the UK matched the top decile of countries, this would equate to 13,495 fewer deaths per year. In our 2022 report, this figure was 12,675 deaths (see Figure 1, using 2019 data).

Safety outside of hospital

IGHI and PSW  are committed to identifying and analysing patient safety data that provide a more holistic view of patient safety. This includes the recognition that safety issues affect people in settings outside of hospitals, including in community and mental health settings, as well as in their own home. While some national data collections include data in these settings, they are often still in relation to people’s stay in hospital.

Data on the number of people who return to hospital as an emergency within 30 days of being discharged, is one such example. This indicator helps us to understand how well the NHS supports people to recover effectively at home following a stay in hospital. Analysis shows that, between 2013/14 and 2022/23, the proportion of people readmitted as an emergency within 30 days of their most recent discharge increased from 12.4% to 14.2%. Data on the number of people who died by suicide within three months of an in-patient discharge is another example. Analysis shows that, between 2011 and 2021, the number of suicides fell from 299 to 193.

1.2   Five themes from the data

Our analysis below is summarised into five broad themes, identifying areas where the data suggest safety improvement efforts could be focused. There are limitations in identifying patient safety priorities based on the data alone, therefore these themes sit alongside the priority areas identified following our other research activities (see Part 3).

1.2.1 Regional variations in safety

Adverse effects of medical treatment refer to death or disability caused by a procedure, treatment, or other exposure to the healthcare system. Overall, disability-adjusted life years (DALYs) resulting from this have increased in England since 2005. However, this impact is not evenly spread, for example with rates in the North East of England more than twice as high as in Greater London (see Figure 2).

Box 1: Disability-adjusted life years (DALYs)

WHO states that “one DALY represents the loss of the equivalent of one year of full health. DALYs for a disease or health condition are the sum of the years of life lost to due to premature mortality and the years lived with a disability due to prevalent cases of the disease or health condition in a population.” DALYs therefore give a fuller indication of the burden of unsafe care on individuals, populations and society as a whole, than looking at mortality alone. They are typically shown per 100,000 population.

The Summary Hospital-level Mortality Indicator (SHMI) compares the number of patients who die following their admission to hospital with the number that would be expected to die. SHMI takes account of patient characteristics, such as age and underlying health conditions, to enable comparison between acute trusts. Analysis of the proportion of trusts with a higher-than-expected number of deaths (up to January 2024) suggests worsening performance in London and some improvement in the Midlands and East of England (see Figure 3).

As of January 2024, all acute trusts in London were categorised as either having fewer deaths than expected, or an expected number of deaths – in January 2024, this figure fell to 94%. In the Midlands and East of England, this proportion changed from 92% in January 2020 to 97% in January 2024.

An important measure of ambulance service performance is the time it takes from receiving a 999 call to a vehicle arriving at the person’s location. Analysis of Category 1 and 2 calls (up to July 2024) – which include life threatening emergencies and urgent situations – shows marked variation in performance across the different regions of England. For Category 1 calls, the difference between the regions with the lowest and highest average response time is 2.1 minutes. For Category 2 calls, this difference is 11.5 minutes. Category 2 response times also show an overall increase throughout 2023 (see Figures 4 and 5).

 1.2.2 Safety concerns in maternity services

Analysis of the most recent MBRRACE-UK reviews – an audit programme that collects data on deaths of mothers and babies – and data from the Office for National Statistics provides cause for concern. It shows that rates of neonatal deaths and maternal deaths have risen for the first time in a decade, and are continuing to rise. Rates of stillbirth also increased for the first time within this period, before falling again (see Figures 6-8).

Stillbirth is defined as the delivery of a baby at or after 24 weeks of pregnancy who shows no sign of life. Between 2013-2020, the stillbirth rate fell by 18.4%, from 4.6 to 3.8 per 1,000 births. In 2021, this figure rose to 4.1 – an increase of 8% on the previous year, followed by a drop to 3.9 stillbirths per 1,000 births.

A neonatal death describes a baby born at or after 20 weeks who dies before they are 28 days old. Between 2013-2020, the neonatal death rate fell by 17%, from 1.7 to 1.3 per 1,000 live births. This figure rose to 1.5 in 2022 – an increase of 15.4%.

A maternal death is defined as a woman who dies during, or up to six weeks after, pregnancy from causes linked to their pregnancy. Between the 2011-2013 and 2017-2019 periods, maternal deaths remained relatively steady, following several years of continual decline. However, between the 2017-2019 and 2020-2022 periods, the maternal death rate (including women who died due to complications of COVID-19) increased from 8.8 to 13.4 deaths per 100,000 maternities – a statistically significant increase of 52.3%.

Disparities in outcomes for women remain since our last report. Women living in the most deprived areas have a maternal mortality rate more than twice as high as women living in the least deprived areas. There remains an almost three-fold difference in maternal mortality rates amongst women from Black ethnic backgrounds compared to White women, and a two-fold difference for women from Asian ethnic backgrounds.

1.2.3 Staff concerns about safety, particularly in ambulance services

Perceptions of safety in the workforce, evidenced in results from the annual NHS staff survey, continue to be a concern, despite recovery to pre-pandemic levels in some areas. For example, the proportion of staff who thought there were enough staff for them to do their job properly was 32.4% in 2023, similar to levels in 2019 (see Figure 9). However, this means that around two thirds of staff still believe they are unable to carry out their jobs fully due to workforce shortages. Only around half of patients (55.7%) responding to the NHS inpatient survey in 2023 said there were enough nurses on duty to care for them in hospital, compared to 58% in 2019 (see Figure 10).

Despite a small drop, 28.8% of staff from Black and Minority Ethnic groups experienced harassment, bullying or abuse from colleagues in the previous twelve months, a figure that remains higher than for White staff (Figure 11).

Perceptions of safety from staff within ambulance services show the greatest need for action. As of 2023, ambulance staff are the least confident that their organisation will act on staff or patient concerns (49.8% and 61.5% compared to the national averages of 56.8% and 70.5%, respectively, Figures 13 and 14). Ambulance staff also report markedly higher rates of burnout (39.1% compared to the national average of 30.4%, Figure 12).

Ambulance staff also report witnessing errors, near misses or incidents that could have hurt patients more than staff in other services (38.3% compared to the national average of 33.2%) (Table 2).

1.2.4 Risk of slipping in ‘previous safety wins’

In the decade leading up to 2018, the NHS achieved and sustained dramatic reductions in rates of healthcare acquired MRSA and Clostridium difficile. This was accompanied by improvements in outcomes for people with venous thromboembolism (VTE, or blood clots in the veins) and hip fractures. However, analysis of the most recent data shows that rates of Clostridium difficile have increased by 54% between 2018/19 and 2022/23 (Figure 15). Following a spike during the pandemic, deaths from VTE have fallen, but still remain above the pre-pandemic level (Figure 16).

1.2.5 Safety while people wait for their care

The NHS understandably had to prioritise services during, and in the immediate aftermath, of the COVID-19 pandemic. The resulting impact can be seen across many areas of care, but most notably in the ability of people to access the care they need in a timely way. For example, waiting lists for elective care were at 7.6 million people in June 2024, while 25% of people continue to wait more than four hours for a treatment decision in A&E (see Figures 17 and 18).

Data on the impact of extended waits on people‘s safety is not routinely collected. However, poor access to care is a clear proxy for unsafe care, and reports on the impact of extended waits for ambulances and specialist appointments, for example, are becoming increasingly common. As stated in our 2022 report, it remains a priority for the NHS to collect near real-time data on people’s safety during extended waits for care, alongside the work to address the waits themselves.

The Global State of Patient Safety

In 2023 we published a report on the Global State of Patient Safety. The report examined global trends using publicly available patient safety data. Overall, as in England, we found a mixed picture for global patient safety, with some notable gaps in data collection. For the UK, data was only available for 68 out of the 89 indicators we looked at. Based on a novel patient safety ranking of OECD countries, using four key patient safety indicators, the UK ranked 21^st out of 38. Norway came first in the ranking, followed by Sweden and South Korea, highlighting the opportunity for shared learning and improvement for England and the UK.

The Darzi Review stated that the NHS in England is an outlier in terms of clinical negligence payments made to people who suffered harm as a result of their care. In 2023/24, payments increased to £2.9 billion, or 1.7% of the entire NHS budget, which represented double the share of total health spending as New Zealand, ten times the level of Australia, and twenty times as much as Canada.

In 2022, the OECD estimated the direct costs of treating patients who have been harmed during their care to be around 13% of total health spending in developed countries. Excluding patient safety incidents that may not be preventable, this was estimated to be 8.7%. Applying this figure to the NHS in England, the direct costs of unsafe care would be £14.7 billion in 2023/24. This represents a significant use of resources, in addition to the human and societal cost of harm caused by unsafe care.

Direct costs include the tests, treatments and other activities associated with delivering the care needed to ameliorate the effects of harm. They do not include the indirect costs of unsafe care. Indirect costs are harder to quantify, but are important for understanding the wider impact of unsafe care, and to inform the prioritisation of patient safety issues. Such costs could include, for example:

• Breakdown of trust between patients and the healthcare system, where public satisfaction with the NHS in 2023 fell to the lowest recorded level since surveys began in 1983.
• Impact on staff morale, mental and physical wellbeing, rates of burnout and sickness (see Part 1), and loss to the service.
• Lost productivity from staff having to develop workarounds within poorly designed and unsafe systems to keep patients from harm, or from delivering some processes that fail to improve safety.
• Wider economic and social burden, such as lost productivity by affected patients and families, and the impact on people’s quality of life (see Box 1).
• The opportunity and business costs from reporting and managing incidents, implementing the Duty of Candour, managing staff through performance processes, and from implementing recommendations that are not evidence based.

These wide-ranging human and economic costs make a clear and urgent case for investing in effective safety improvement efforts. However, the array of safety issues facing the NHS makes the process of prioritising efforts challenging. We therefore carried out a range of research activities, listed below, to shed light on patient safety priorities from the perspective of different stakeholders – patients and the public, health and care workers, care providers, and national bodies:

• Review of publicly available information from government and national bodies on national patient safety priorities
• Desk research on how data is, and could be, used to inform national patient safety priority setting
• Analysis of published Patient Safety Incident Response Plans to identify individual care providers’ patient safety priorities
• A poll commissioned from YouGov asking health and care workers and the public to rank areas of care that the NHS should give extra attention to, followed by a stakeholder workshop to discuss the findings.

 3.1 National patient safety priorities

There are a range of different organisations, strategies, plans, initiatives and events that inform and set national patient safety priorities in England. These are set in both a proactive way, for example, through the national patient safety strategy, and a reactive way, for example, through public inquiries and safety events. These different mechanisms are described below.

3.1.1 Proactive priority setting

In 2019, NHS England published its first Patient Safety Strategy (which was later updated in 2021). The strategy set out NHS England’s ambitions to better understand, prioritise and address patient safety issues, and included 61 areas for action. These action areas ranged in scale and complexity, with some being sub-divided into multiple further actions, and many requiring significant work and careful coordination across the system – just one example being the implementation of NEWS2 to prevent harm caused by unrecognised or untreated patient deterioration. The strategy originally identified the following four national safety improvement priorities “because of their potential to enable the most significant impact on patient safety”:

• Preventing deterioration and sepsis
• Medicines safety
• Maternal and neonatal safety
• Adoption and spread of tested interventions – such as care bundles to improve safety in emergency abdominal surgery and discharging people with COPD.

In 2023/24, NHS England issued a list of eight priority areas for leaders and Patient Safety Specialists, focusing on the infrastructure to deliver safer care. These can broadly be divided under the three categories below:

• Implementation of new requirements – such as the new Patient Safety Incident Response Framework (PSIRF), the framework for involving patients in patient safety, the Patient Safety Syllabus and the Learn from Patient Safety Events service.
• More effective implementation of existing requirements – such as the Medical Examiner responsibilities, and adherence to National Patient Safety Alerts.
• Creating the conditions to support safer care – such as the development of a Just Culture and the delivery of the national patient safety priorities identified above.

Other national plans and guidance documents contain commitments and priorities on patient safety. For example, the annual Priorities and Operational Planning Guidance sets out what the NHS will deliver each year. In 2024/25, this included commitments to reduce ambulance response and A&E waiting times, and improve the overall quality and safety of services, particularly maternity and neonatal services.

The National Institute for Health and Care Research (NIHR) funds six Patient Safety Research Collaborations (PSRCs). PSRCs are partnerships between universities and NHS trusts that support patient safety research to address one or more of the following Strategic Patient Safety Challenges:

• Improving patient safety intelligence and understanding of patient safety challenges
• Improving organisational patient safety culture and practice
• Patient safety behaviours
• Effective patient safety practices
• The patient safety impacts of alternative service delivery models
• Ergonomics, design and human factors
• Clinical risk scores (validation, implementation and outcomes)

 3.1.2 Reactive priority setting

Safety events are a source of patient safety priorities. National Patient Safety Alerts are issued by NHS England following review of clinical incidents reported through the national reporting system and other sources. Decisions on whether to issue an alert are based on the remit of NHS England and the nature of the risk. Since November 2019, 19 alerts have been issued and remain in force, covering areas such as the administration of medicines, use of medical devices, and procedures for carrying out medical procedures.

The regulatory system itself provides another source of patient safety priority setting. Research has identified at least 126 regulatory bodies with a remit for patient safety (this figure excludes commissioners of care). The Darzi Review identified an increase of people working in regulatory type functions in England, from just over 2,000 in 2008 to more than 7,000 in 2024. This increase in the number of people working in regulatory roles equates to a growth from an average of five per provider to more than 35 over the same time period (acknowledging some organisations had merged during this time).

The Health Services Safety Investigations Body (HSSIB, formerly HSIB) is one such body with regulatory functions, conducting national investigations into patient safety concerns across England. Between 2017-2023, HSIB completed 84 national reports, making 236 national safety recommendations, 210 safety observations and 59 safety actions, to 57 organisations. A report by HSSIB in 2024 highlighted how the ‘noise’ created by the significant volume of recommendations across the sector meant providers were struggling to implement improvement work. Other bodies include the Care Quality Commission, the Parliamentary and Health Service Ombudsman, and more recently, the Patient Safety Commissioner.

Perhaps the most high-profile source of patient safety priorities are the outcomes of major reviews, investigations or inquiries, many of which are triggered by widescale failures in care. In 2024, the House of Commons Health and Social Care Committee commissioned an expert panel to evaluate the Government’s progress on meeting patient safety recommendations. Evidence submitted by the Government identified 508 recommendations pertaining to patient safety from 12 public inquiries and reviews between 2010-2022 – averaging one inquiry and 42 recommendations per year. Occasionally, a tragic case can trigger new guidance and action to prevent future harm, such as the introduction of Martha’s Rule following the death of 13 year old Martha Mills, giving families the right to a rapid review of a patient’s condition.

3.1.3 Making the priority-setting process transparent and evidence-based

There is a plethora of patient safety priorities in England, which are determined through multiple routes. Each priority requires urgent attention and action across the health and care system, which the government and healthcare organisations often struggle to keep pace with. It is not always clear how longer-term, proactive patient safety priorities are chosen. There would be value in publishing the evidence and decision-making process used to determine them. The latest thinking and evidence should be used to inform this process, in conjunction with patients, families, carers and clinicians.

3.2 Using data to inform national patient safety priority setting

Based on work already being done, and existing research that has the potential to be applied in practice, we highlight below three examples of how data and insights can be used to inform patient safety priority setting at a national level.

3.2.1 Using claims and litigation data to set national strategic priorities

Each year NHS Resolution resolves a small number of compensation claims that arise from avoidable intrapartum brain injuries. These incidents are devastating for the child, their family and those who care for them, involve lifelong care needs and can result in multi-million pound compensation payments from NHS funds. Preventing just one of these cases would save untold anguish to families and the healthcare staff involved, some of whom go on to leave the profession. It would also result in millions of pounds in savings to public funds.

NHS Resolution estimated an annual ‘cost of harm’ for the cost of claims resulting from incidents that took place during the financial year 2023/24 as £5.1 billion. Of this, 49% relates to maternity (compared to 52% in 2022/23). As a result, NHS Resolution continues to prioritise addressing the causes and management of incidences of brain injury sustained at birth. Improving maternity outcomes is one of NHS Resolution’s four strategic priorities for 2025. This includes the following initiatives:

• The Early Notification (EN) Scheme: launched in 2017, it proactively investigates specific brain injuries at birth for the purpose of determining whether clinical negligence played a role in causing harm. The EN scheme has established a Maternity Voices Advisory Group (MVAG) to provide external stakeholders, in particular families and their representatives, with a forum through which they can advise and support future service developments. An evaluation of the Scheme is underway with findings expected in 2025.
• The Maternity Incentive Scheme (MIS): launched in 2017, it creates a financial incentive for trusts to deliver against ten safety actions agreed by a wide range of stakeholders. The Maternity Incentive Scheme – NHS Resolution is fundamental to driving improvements in the safer delivery of care in maternity and neonatal services. This Scheme is also in the process of being formally evaluated.

3.2.2 Understanding the broader impact of unsafe care to design national safety improvement programmes

Research by Hauck and colleagues at Imperial College London calculated healthy life years (HLYs) lost in English hospitals due to six major types of adverse events: pressure ulcers, post-operative sepsis, central line infections, deaths in low mortality procedures, deep-vein thrombosis or pulmonary embolism, and hip fractures. HLYs refer to refer to the years of good life the patient would have had if the adverse event had not occurred. The researchers found that preventable pressure ulcers had the greatest impact on people’s lives, resulting in 26 HLYs – in comparison, central line infections accounted for less than 1 HLY.

The data used in the study covers the period from 2005/06 – 2009/10. During this time, a major national initiative called Matching Michigan was launched to drive down rates of central line infections. There were several good reasons to address this safety issue, and to choose the particular intervention to address it:

• It was a clearly defined problem with high public profile (healthcare-associated infection)
• It could be addressed within a bounded care environment (intensive care units)
• There was an apparently codified solution (a bundle of technical interventions such as appropriate hand hygiene)
• There was a successfully implemented intervention to replicate (the Keystone programme in Michigan).

However, the research by Hauck et al suggests that the overall burden from pressure ulcers at the time was far greater. We also know that action was already underway in many ICUs to tackle the problem of central line infections before Matching Michigan came along.

3.2.3 Learning from coroners’ death investigations to make system-wide safety improvements

Coroners across England and Wales investigate the circumstances of deaths that have an unknown, unnatural or violent cause. This includes a duty to issue Prevention of Future Death reports (PFDs) to individuals, organisations or government departments when they consider action should be taken to prevent further deaths. Research has shown that PFDs contain rich information and provide essential lessons for the NHS and other services, but that this is not consistently acted upon.

Coroners’ PFDs were difficult to use and analyse until the development of the Preventable Deaths Tracker (PDT), a national vigilance tool that harnesses information from coroners to minimise premature deaths. The PDT uses real-time analytics to understand who does and does not respond to the coroner, finding that only half included a published response from the individuals or organisations they were sent to – despite this being a legal requirement.

An analysis conducted by the BBC in 2023 found around one in five PFDs (109) highlighted long waits for NHS treatment, or a shortage of medical staff or other resources such as beds. Of these, 26 involved mental illness or suicide, and 31 involved ambulances and emergency services. Research into PFDs has also focused on specific clinical areas – one example being maternal deaths.

An analysis of over 4,000 PFDs published between 2013 and 2023 using the PDT identified 29 maternal deaths reported by coroners in England and Wales. The reports frequently highlighted gaps in national guidance, inconsistent local guidelines, and poor communication – but only 38% of reports received a response. The researchers also compared the data with MBRRACE-UK. Such an approach highlights the value in complementing case-level insights with larger-scale data analysis when understanding and prioritising patient safety issues.

3.3 Patient Safety Incident Response Plans

The Patient Safety Incident Response Framework (PSIRF) sets out how the NHS responds to patient safety incidents. PSIRF became operational in September 2023, replacing the Serious Incident Framework. PSIRF represents a step change in how the NHS establishes and maintains safe systems. It aims to achieve:

• Compassionate engagement and involvement of those affected by patient safety incidents
• Application of a range of system-based approached to learning from patient safety incidents
• Considered and proportionate responses to patient safety incidents
• Supportive oversight focused on strengthening response system functioning and improvement.

NHS trusts are required to produce a Patient Safety Incident Response Plan (PSIRP), describing how they will achieve these aims. Trusts are also required to include details of the patient safety issues they are giving particular focus to locally, by consulting stakeholders and reviewing data. PSIRPs therefore provide, in theory, useful insights into the types of locally important patient safety issues that NHS trusts are prioritising.

At our stakeholder workshop attendees were concerned that, while organisations have begun to implement PSIRF, they need ongoing support, resource and expertise to do it effectively. This includes the ability to identify and use all appropriate data and insights to set local priorities, and to turn learning into action for the benefit of patients and staff. An important condition for this was for national organisations to continue to emphasise that the primary aim of PSIRF was learning and improvement, not blame and performance management.

3.3.1 Examining Patient Safety Incident Response Plans (PSIRPs)

To enable our analysis, we collated every publicly available PSIRP and extracted information on their local patient safety priorities, and how they planned to respond to them and make improvements. PSIRPs were collated using existing research, from exploring individual NHS trust websites, and from general web searching. We then conducted a Large Language Model (LLM) analysis to better understand key themes emerging from the PSIRPs. For further information on our approach, please see our Technical Annex.

3.3.2 Availability of PSIRPs

Our analysis found notable variation in whether PSIRPs had been made publicly available, and whether the relevant information was included or clearly identifiable. As of 12 July 2024, 76% of trusts had a publicly available PSIRP (162 out of 214 trusts). This figure varied by region, Integrated Care System (ICS) and trust type:

• Across regions, public availability of PSIRPs ranged from 55% in the South East to 95% in the East of England and the South West.
• Across ICSs, public availability of PSIRPs ranged from 33% to 100%.
• Across trust types, public availability of PSIRPs ranged from 67% for mental health trusts to 90% of ambulance trusts.

Three percent of the publicly available PSIRPs could not be included in our analysis, as the relevant information was not included or clearly identifiable. Of the publicly available PSIRPs:

• The number of local patient safety priorities per PSIRP ranged from 1-30. The mean number was eight.
• 86% (140 out of 162) of PSIRPs included planned responses for each priority.
• 67% (109 out of 162) of PSIRPs included improvement routes for each priority.

3.3.3 Local patient safety priorities

Our analysis found that the most commonly identified local patient safety priorities were focused on the issues listed below (see Figure 19):

• Pressure ulcer prevention and tissue care management.
• Timely identification and management of patients at risk of deterioration.
• Falls prevention.
• Delayed, missed or incorrect diagnoses, including for cancer.
• Infection prevention and control.
• Medication safety.

In addition to the priorities identified above, our analysis found that trusts also commonly prioritised the improved management of patient safety incidents more generally, including the enhanced involvement of patients and families.

These priorities most commonly reflect safety issues in acute trusts as they make up over half of the 229 NHS Trusts. We will be conducting further analysis to understand the issues in specific settings of care, such as in mental health and community trusts. We are also developing a searchable database of PSIRP priorities that will enable people in the NHS to identify other organisations working on similar patient safety issues, to support the sharing of potential solutions.

Our PSIRP analysis approach

Our analysis of the PSIRPs utilised a topic modelling technique. Topic modelling is a machine learning approach that uncovers themes within a collection of documents. The main advantage of this method is the speed with which key insights can be drawn to support rapid learning and action. Further details on our method can be found in the Technical Annex. Manual thematic analysis can enable more detailed insights to be generated, alongside the use of expert judgement, but is time-consuming.

Our model clustered certain words based on their degree of similarity, from which we created the themes shown in Figure 19. These six themes were identified where we had a high degree of confidence in the similarity of the words, and in their meaningfulness. These themes only contain about 12% of the total number of local patient safety priorities extracted from the PSIRPs we analysed. Reasons for this include:

·  Some clusters of words were not related closely enough to recognised patient safety terminology.

·   Some clusters of words were relevant to multiple patient safety themes, therefore they were not assigned to a single theme.

·  Some priorities included many words not associated closely enough to patient safety, preventing accurate categorisation.

However, our overall analysis suggests that many of the uncategorised priorities were likely to be related to the six themes identified above, even though they were not included within our final selection.

We also analysed the planned responses and anticipated improvement route through which trusts would respond to each priority, but no distinct themes emerged from the output. This was due in part to non-specific terms frequently being used by trusts. For example, under anticipated improvement routes, frequently used terms included "inform ongoing improvements" and "create local recommendations".

Overall, there is a notable degree of alignment across NHS trusts in terms of the patient safety challenges they are seeking to address. Historically, the NHS has struggled to implement the recommendations from national inquiries into patient safety failures, or effectively manage the safety risks that arise within local organisations. There is, therefore, benefit to the system coming together to share, agree and implement evidence-based interventions to address them, using approaches based on the latest safety science (incorporating human factors and systems thinking)

PSW is conducting work in partnership with Thiscovery to further understand the approaches NHS organisations are taking to address some of these common challenges, and guide thinking to support the more effective adoption of proven interventions.

3.4 Polling the public and health and care workers

In June and July 2024, IGHI collaborated with YouGov to conduct a poll of the general public and health and social care workers in England, asking them to rank 10 patient safety priorities (see Box 3). The poll was conducted to gain a better understanding of what people consider to be the areas most in need of attention to ensure the safest possible care.

The polling question and list of priorities was initially derived from the relevant patient safety literature and the themes identified in Part 1 of this report. They were further revised with patient safety experts, health and care workers, and patient and public partners. The findings of the poll were then presented to a stakeholder group, including members of the public and healthcare workers, to better understand their implications. More information on the survey can be found in Box 2.

Across all three groups, waiting for urgent care was ranked as the number one patient safety priority. This includes people waiting for ambulances, for treatment in A&E, or for an urgent appointment with a cancer specialist (see Figure 20). Forty-seven per cent of the general public, 41% of NHS workers, and 45% of social care workers ranked it as their highest priority. Across all three groups, people using digital services (such as video consultations or patient apps), was consistently ranked the lowest priority.

This was also true when looking at the distribution of responses from the public by region in England. For all regions, waiting for urgent care was ranked as the top priority (ranging from 40% of respondents in the North East to 51% of respondents in the East of England, see Figure 21). Across all regions, the use of digital services was consistently ranked the lowest priority.

Insights from workshop attendees highlighted how workforce burnout, combined with anxiety from patients and the public about accessing care when they need it most, may have contributed to the prominence of urgent care in their responses. Attendees stressed, however, that urgent care is not a “discrete” issue. Problems in A&E or ambulance services often highlight issues with other NHS services, such as the ability to provide adequate mental health services before people experience a crisis, or adequate social care once somebody is medically fit for discharge from hospital. Therefore, tackling issues in urgent care is a complex problem requiring whole system solutions.

In our 2022 report, we concluded that the lack of timely access to care, evident in long waiting times for services, should be considered an urgent patient safety issue. The results of this poll underline this conclusion. They also suggest that current definitions of patient safety should be revised, to cover the avoidance of harm during the provision or absence of healthcare. Seeing patient safety in this way has implications for where improvement efforts are focused.

Among the general public, people from Black ethnic backgrounds ranked care for mothers and babies almost as high as waiting for urgent care, and notably higher than for any other ethnic group (see Figure 22).

Safety for mothers and babies from Black ethnic backgrounds is a significant patient safety issue. Data consistently report that women from Black ethnic backgrounds are three to four times more likely to die in pregnancy and childbirth than White women. Recent analysis also shows that women from Black ethnic backgrounds are up to six times more likely to experience some of the most serious birth complications during hospital delivery. The results of this poll show that these safety concerns are front and centre of the minds of those most affected by these issues.

For NHS workers, all groups of respondents except those working in ambulance services ranked “people waiting for urgent care” as the highest priority. Respondents working in ambulance services ranked people using mental health services as their number one patient safety priority. This includes access to, and care provided in, community mental health services and inpatient units, for both children and adults (see Figure 23).

Mental health emergencies are among the most challenging and time-consuming cases for ambulance staff to attend to, partly due to the need to refer people to other services. Recent analysis showed that national datasets do not capture thousands of such people as they are often not taken to hospital. Insights from our stakeholder workshop emphasised the pressure ambulance crews are under to care for people experiencing a mental health episode, and concerns about an insufficient number of crisis units to support their ongoing care. Social workers in the younger age groups, and health visitors, also prioritised people using mental health services, as well as mothers and babies, as a higher priority than other groups (see Figures 24 and 25).

Box 3: About the YouGov poll

The poll was conducted using an online survey. It was administered to members of the YouGov panel of more than 2.5 million people who have agreed to take part in surveys. Emails are sent to panellists selected at random from the base sample, providing them with a survey link.

The general public poll was conducted between 21 June – 3 July. The total sample size was 2060 adults. The figures have been weighted and are representative of all English adults (aged 18+).

The NHS workers poll was conducted between 19 – 26 June. The total sample size was 1079 adults. The figures have been weighted and are representative of NHS staff.

The social care workers poll was conducted between 28 June – 11 July. The total sample size was 511 adults working in adult social care. The sample is not representative of the adult social care workforce. Analysis is based on 418 social care workers who are based in England, for comparison purposes.

For further information on the survey, please see our Technical Annex.

Our analysis of the patient safety data highlights concerns about the safety of maternity services, the culture of safety in the workforce, and unwarranted variations in care and outcomes. It also highlights the importance of constant vigilance in areas of past progress, and better data to monitor people’s safety while they experience delays in their care.

Our poll of the public and health and care workers found that waits for urgent care was the area of greatest concern. Analysis of local patient safety plans identified six patient safety problems that many NHS organisations are commonly prioritising. However, our analysis also showed an opaque process for setting national patient safety priorities, and a system that is failing to keep pace with the significant volume of recommendations it receives.

The NHS has the potential to become a world-class learning organisation. To achieve this vision, change is urgently required. Patients in some parts of the NHS cannot currently benefit from the safety improvements being delivered in other parts. Also, high-quality data – at the local and national level – are not always collected or used in a “problem sensing” way that enables early detection of patient safety issues. However, identifying problems and making recommendations alone is not enough. The NHS currently faces an overload of priorities and actions, placing the system at risk of losing focus on the most pressing patient safety concerns.

Combined with the severe pressures being placed upon the NHS, it is imperative that the NHS must finally start to ‘act like a sector’ and focus its limited resources where it counts. We therefore set out two ambitions below for the health and care system, which we believe will support the long-term improvement of patient safety in England:

1. Local NHS organisations must be supported to adopt evidence-based interventions to tackle the most common safety problems causing significant harm to patients. Our analysis of trust patient safety plans identified six common problems that many organisations are tackling, such as pressure ulcers and patient falls. Adopting proven interventions to common problems like these would finally see the NHS truly acting like a National Health Service. We envisage a future where the first port of call for NHS organisations is a repository of such interventions, along with the support they need to implement them, rather than developing their own solutions from scratch.
   
2. National organisations must agree on a focused set of patient safety improvement priorities for the system to rally around. Our analysis found a crowded landscape of patient safety bodies, an opaque process for national priority setting, and evidence that the system cannot keep pace with the volume of recommendations it receives. We envisage a future where patients and healthcare workers are partners in the development of these priorities, and where national organisations rationalise their own activities to ensure the NHS is supported to deliver improvements against them.

Improving maternity and neonatal safety

Based on our analysis in this report, improving safety and reducing inequalities in maternity and neonatal services is a pressing patient safety priority. This is an area that has been subject to considerable attention, initiatives and policy commitments in recent years. Although the government will not meet its ambitions to halve rates of stillbirths, neonatal and maternal deaths and brain injuries by 2025, the data show (see Figures 6-8) that progress was made in the preceding years.

It is possible that a combination of national ambition and local action made a real difference in outcomes, but weak or absent reporting and evaluation of improvement programmes has undermined our ability to understand what worked. It is also apparent that progress has since stalled, and in many places worsened, by not building on what worked or responding quickly enough to warning signs in the data.

We believe that, by applying our recommendations to maternity and neonatal services, it is possible to establish a coherent set of national policies and actions, and ensure the support is there for frontline teams to learn from and adopt proven practices to make care safer for mothers and babies.

These recommendations are timely, as government looks to set the priorities for the NHS more broadly in its 10 Year Plan, due in spring 2025. They also come as Dr Penny Dash reviews the plethora of bodies responsible for patient safety. We hope that both the 10 Year Plan and the Dash Review will embrace and endorse these two recommendations, with clear patient safety priorities that have been co-produced with patients and staff, as well as the necessary improvement support for NHS organisations to progress them.

We will revisit the National State of Patient Safety in 2026, when we hope to see more progress than has been made over the last two years. We will also consider how we can find evidence on issues such as the impact of culture, workforce welfare and retention, effectiveness of regulation, leadership and reputation management that all influence patient safety.

Acknowledgements and suggested citation

This report was produced by Imperial College London's Institute of Global Health Innovation. It was made possible by the generosity of the charity Patient Safety Watch founded by the Rt Hon Jeremy Hunt MP, led by James Titcombe OBE. We extend special thanks to James for his expert input and advice throughout the production of this report.

The authors would like to thank the following people who have provided valuable input and advice in the production of this report: Georgia Butterworth, Mike Durkin, Bryony Dean Franklin, Rebecca Lawton, Darren Ashcroft, Suzette Woodward, Christoforos Pavlakis, Charity Gondwe, Georgia Richards, Megan Bidder, Naomi Assame, Annie Hunningher, Winsome Thomas and Gurdees Watson.

Thank you to the team within the Institute of Global Health Innovation who supported the development of the report: Victoria Murphy, Laura Horga and Anna Lawrence-Jones.

The authors received infrastructure support from the NIHR North West London Patient Safety Research Collaboration. Patient Safety Watch would like to thank Joe Kiani for his dedication and international patient safety leadership and the Masimo Foundation for their generous financial support.

Suggested citation: Illingworth J, Fernandez Crespo R, Hasegawa K, Leis M, Howitt P, Darzi A. The National State of Patient Safety 2024: Prioritising improvement efforts in a system under stress. Imperial College London (2024).