Independent report

Rapid review into data on mental health inpatient settings: final report and recommendations

Updated 21 March 2024

Applies to England

Ministerial foreword

Every patient deserves to be treated in an environment where they receive high quality care and are treated with dignity and respect, and their families and carers deserve to be reassured that their loved ones are safe. That is why I commissioned this rapid review to improve the way data and information is used in relation to patient safety in mental health inpatient care settings and pathways. I am grateful to the chair, Dr Geraldine Strathdee, for her passion and energy in driving it forward.

Firstly, I wish to pay tribute to everyone who contributed to the review’s findings; the sheer number of people who have engaged so enthusiastically with the review has proven that this is immensely important to everyone in the mental health sector. I wish to pay particular thanks to the experts by experience, both patients and carers, who engaged with the review, and contributed thoughtful and passionate insights. Your contributions have been absolutely vital to the success of the review, and have highlighted the need for patients, carers and staff to have a greater voice in flagging patient safety concerns.

The review’s findings demonstrate that we have some way to go to universally measure what matters in relation to patient safety. The recommendations have identified ways in which the system can improve how data and evidence is used more effectively, which has the potential to bring us closer to understanding the risks to delivering safe and high quality care and being able to intervene before things go tragically wrong.

This is an important opportunity to ensure patient safety is at the forefront of mental health care. I truly believe that the recommendations from this review can improve the way we use data to identify failures in care and risks to patient safety. As a result, I hope that, once implemented, these recommendations will help contribute towards saving lives and improving patient experience and therapeutic outcomes.

Maria Caulfield, Parliamentary Under Secretary of State for Mental Health

Chair’s introduction

I am deeply honoured to have been asked by ministers to chair this rapid review about the current data and information on mental health inpatient pathways and make recommendations about what changes are necessary in the way such data is collected, processed, reported and used for improvement.

As a clinician it has been my lifelong belief that ‘information is power’. Communicated in the right way, it helps individuals to feel more in control of their lives, staff to understand their services and organisations, and at every level it enables people to make better decisions to improve lives and care.

The aim of improving mental health information and information systems is ultimately about keeping patients safe and providing consistently high quality, evidence-based therapeutic treatments that enable patients to achieve the outcomes they need to have the quality of life they want back in their communities.

I am deeply grateful to the more than 300 people, representative of every part of the sector, from experts by experience through to frontline healthcare staff and including local and senior national leaders, who came forward to share openly their challenges, their concerns and their expertise. This work was only possible due to the committed and dynamic team from the Department of Health and Social Care mental health policy unit.

Many of our key findings were expected. People are deeply concerned about tragic instances of poor quality care and abuse. They are concerned about the unwarranted variations in the cultures and standards of care between teams, day and night services, and across organisations. They were concerned how best to support both patients and staff to feel safe so that every service could be one in which ‘yes’ would be the answer to the question ‘would I be content if a member of my family had to be admitted for treatment here?’ Almost everyone wanted open learning for benchmarking and to support rapid improvement and prevention of abuse.

A number of our findings were, however, very surprising. We had expected to hear that the data burden was high. But we had not expected the sheer scale of the challenge. Some clinicians reported spending half or more of their time entering data, taking time away from face-to-face therapeutic treatment and care of patients and undertaking the quality improvement work they and boards had committed to. One trust leader reported that business analysts spent well over half of their time flowing data into the ‘black hole’ where data was often not fed back or reported in a timescale so far from ‘real time’ as to lack utility.

Despite the data burden, patients and families and clinical teams told us that the system is not measuring what matters. They do not consider we are measuring what will truly have an impact on patient safety and outcomes. Too much data collection is about activity and processes and too little about patient experience, what therapeutic treatments are provided, and the ‘real time’ patient and clinician reported progress and outcomes. Despite the reduction in burden that could be achieved by use of digital tools, there was not a universal level of knowledge about, or provision of, these enablers.

Board members told us that culture is key. For a person to feel safe, the experience and culture of care needs to be kind, compassionate and hopeful, so they can make the progress that supports them to achieve a good quality of life back in their community. Staff also need to feel safe and supported to speak up early when a service does not meet the standards they want to deliver. Only a live and frequent presence by board and senior staff on wards could provide the assurance needed.

The review recommendations are based on the expert inputs from our stakeholders. They proposed practical, feasible recommendations. As many told us, they are proud of the wealth in England of our strong lived experience voice, our quality standards, therapeutic guidelines and examples of good and innovative practice. But they are deeply frustrated with the inability to ensure implementation of these for every patient without variation, and the lack of open accessible mechanisms to share and learn from what good looks like.

Implementation of the recommendations requires leaders and teamwork at every level from clinical teams to provider boards to integrated care systems (ICSs) coming together sharing values, vision, training and action.

During our rapid review, we have encountered outstanding leaders in every stakeholder group and sector. We wish them good speed in the implementation of these recommendations.

Dr Geraldine Strathdee, independent chair of the rapid review

Executive summary

The rapid review was commissioned by ministers in response to concerns that the right data and information to provide early alerts to identify risks to patient safety in mental health inpatient settings and prevent safety incidents was not available and was undermining efforts to improve care and keep patients safe.

We consulted with over 300 experts in mental health inpatient pathways, including experts by experience (including carers and families), healthcare assistants, nurses, psychiatrists, managers, clinical directors, chief executive officers (CEOs), chief clinical informatics officers, non-executive directors, integrated care systems (ICSs), safeguarding adults boards, regional and national leaders, academics, data experts, regulators, third sector organisations and others, and reviewed roughly 50 submissions, reports and other evidence that have been sent to the review team.

Our principal findings, based on our engagement and research, can be found below.

Purpose and objectives of the review

The purpose of the review was to consider the way that data and evidence relating to mental health inpatient settings and pathways was collected, processed and used to identify risks early, and mitigate them to protect to the safety of patients. This includes quantitative information and qualitative evidence from patients, carers and staff. Our objectives were to:

• review the data that is collected on mental health inpatient services by national bodies, regional teams, local systems, providers of NHS-funded care and others with a role in collecting information related to patient safety, and to understand how data streams are used and acted upon
• understand how the experiences and views of patients, families, staff and advocates relevant to mental health inpatient services are collected, analysed, collated and used
• understand whether data and intelligence are collected and used in such a way as to identify risk factors for inpatient safety and aid our understanding of patient and carer experience, whether people are receiving high quality care, whether people are cared for in a safe and therapeutic environment, and how data and intelligence are used by providers and local commissioners to reduce risk and drive a proactive culture of improvement
• identify ways in which the collection and use of data can better identify settings where patient safety might be at risk and to make sure that decision makers at all levels have the information they need to monitor and improve patient safety effectively - this should take into account the importance of minimising the burden of data collection, particularly for frontline staff

Read the full terms of reference for the review

While we were conscious of other investigations that related to failures in care in mental health inpatient settings, the review did not seek to investigate any particular trust or provider, nor to cover ground that was covered in other reports.

Summarised findings

We have divided the findings in our report into 5 key themes. These are:

1. Measuring what matters
2. Patient, carer and staff voice
3. Freeing up time to care
4. Getting the most out of what we have
5. Data on its own is not enough

Our findings are briefly summarised below.

Measuring what matters

We heard that, for mental health inpatient pathways, safe care is therapeutic care. This means that patients must be cared for in the most appropriate setting by staff trained in the full range of therapeutic interventions needed to meet the person’s mental health, physical, social, psychological and other needs, and must be provided with a range of daily therapeutic activities, and have their admission, leave and discharge managed effectively.

To reduce the risk of patient safety incidents, mental health staff and leaders need to focus not just on where things go wrong, but on what is needed for things to go right, such as providing safe, therapeutic, compassionate environments and functioning pathways.

We heard that culture is key. For a person to feel safe, the experience and culture of care needs to be kind, compassionate and hopeful, so they can make progress that supports them to achieve a good quality of life back in their community. We heard that staff also need to feel safe and supported to speak up early when a service does not meet the standards they want to deliver.

During the review, in collaboration with a range of experts, we developed a ‘safety issues framework’ (see Appendix 1) that sets out the key safety issues relevant to mental health inpatients and the factors that create safety risks or foster protective, therapeutic environments.

We conducted an initial scoping assessment of the extent to which national and local bodies are able to monitor the issues set out in the safety issues framework using the current data sets and information sources. This identified a gap between what is now available and what is needed. A large amount of data on activity and process measures is collected, as well as some on acuity and performance measures, but more systematic metrics on environment and workforce are needed and there are significant gaps in therapeutic care, outcomes and culture.

More generally, we found that more needs to be done to understand whether patients are receiving therapeutic care at all times as a vital component of care quality and a pre-requisite for safe environments and cultures.

Patient, carer and staff voice

There was a consensus that hearing from patients, carers and staff was essential, both about their experience of services and getting their input into the design of those services.

We found that there were barriers to getting real time, honest feedback from the people most closely connected to the wards through to managers and leaders at board level and beyond.

We found that sometimes the routes to give feedback were not clear or were difficult or time-consuming to use. When they did provide feedback, patients, carers and staff reported that sometimes they felt it was not listened to or acted on. Patients, carers and staff all reported that they sometimes felt unable or afraid to raise concerns or voice their feelings about safety and care on the wards for fear of the repercussions to themselves and others.

Many of the leaders we spoke to described to us the different initiatives they had in place to gather the views of patients, carers and staff, but were honest about the challenge they faced in getting useful responses that they could use in a timely fashion.

We found examples of good practice. Several of the provider leaders we spoke to told us about having experts by experience on their boards and in their leadership meetings, or about having patient councils that senior leaders could attend to hear patient voice first hand.

Others told us about their programmes of unannounced visits to wards day and night to gather informal or ‘soft’ intelligence from conversations with patients, carers and staff. The providers we spoke to highlighted their ‘Freedom to Speak Up Guardians’ for staff feedback. We also heard about technical innovations, such as the use of apps to allow patients, carers and staff the opportunity to give quick, real time feedback.

We heard from experts at all levels that visitors to inpatient settings, such as carers or advocates, play an important role in providing feedback and escalating concerns.

Experienced chairs and CEOs of highly rated services told us that it was vital to provide proactive, psychologically safe opportunities for staff to raise concerns early about standards they considered were trending towards being untherapeutic and unsafe. We were told, “By the time people blow the whistle, it’s too late; a lot of harm has been caused.”

We heard that staff and board members should ask themselves regularly, ‘Is this service of a standard that I consider would be a place where I or a member of my family would receive kind, compassionate, therapeutic care?’

Freeing up time to care

When we first established the review, one of our assumptions was that the data burden on staff was too high and that we would need to make recommendations to reduce it. However, we were not prepared for the sheer scale of the issue.

We heard that it was common for frontline nursing and clinical staff to spend as much as half their shifts in the office entering data. We were told by trust leaders that roughly half of their analysts’ time was used to flow data to national and local data sets instead of providing support for quality improvement to frontline staff and trust leaders. Senior clinical leaders cited data burden as one reason for staff retention challenges.

Frontline staff described a situation where they had to enter data multiple times into multiple systems. The systems they use are often outdated and do not communicate well with each other, and the requirements for data have grown over time as more and more demands are made to provide assurance to senior colleagues locally, regionally and nationally.

Staff told us that they were frustrated about the amount of time it took to enter data, a process that several described as ‘feeding the beast’, and we found that there was a real risk that the data burden on staff increased, rather than reduced, risks to safety as it reduced the amount of time staff could spend providing therapeutic care to patients.

We also heard about multiple data requests from different parts of the system, all with slightly different asks from each other, all of which needed their own separate returns and which often requested the same data with different definitions and formats. We heard this could be a particular challenge for independent sector providers that provide services across the country and NHS trusts covering large geographies. We heard that returns are required from multiple commissioners such as ICSs and provider collaboratives, often with different data returns and varied definitions. This was described by one senior leader at a trust as a ‘wild west’ situation.

Getting the most out of what we have

Another message that we heard throughout the review process from all levels was that there is no shortage of data on mental health - one senior national leader told us that mental health services were ‘swimming in data’. However, we found that local, regional and national bodies often do not make the most of the data they receive.

We found several local examples in trusts and independent sector providers of high quality dashboards that gathered and triangulated data about the key categories of risks to inform leaders. In these organisations and those that had less digital resource or had more recently moved to introduce electronic care records, we saw outstanding examples of action-orientated daily reviews in ward safety huddles where multidisciplinary ward teams systematically addressed the holistic needs of patients and carers. We were especially impressed where the skill mix in those teams had been built to address the range of needs of patients including those with autism and communication difficulties.

We heard that data is often not available in real time, and often has a time lag of weeks or months before it reaches frontline teams and local leaders. We found that, for data to be effective in providing early alerts to reduce risks to patient safety, it had to be available in as close to real time as possible in an accessible format.

We found that staff at all levels did not receive value from the data they enter - too little comes back to frontline staff and provider leadership that can provide them with insights about their patients and the performance of their services. At all levels, staff could recall almost no feedback from reports from national or regional data sets that provided them with an accessible report in real time that was useful to help them do their jobs.

Frontline clinical staff told us that they would value more targeted, directed, digestible information being sent to them in easy-to-use formats so they did not have to spend their time on data analysis, in particular information about patient experience and information that would help them compare themselves against other similar wards.

We heard that providers often do not have enough analytical resource to service the needs of all their teams, in part because of the amount of time analysts spend feeding information to external local and national data sets. This, and the lack of agreed standard operating processes for data entry, coding and reporting means that clinical staff often do the data entry and processing, which is not the best use of their time or expertise and often not part of their skillset.

Senior clinicians and professional training leads told us that they had had conflicting advice from local and national commissioners about the way to code diagnoses, patient needs and therapeutic interventions, so the data collected was often not in a consistent format with a consistent set of definitions.

We heard from non-executive directors and Quality Committee chairs that, while there was appreciation for the drive from NHS England to use statistical process control analyses, there was a need to move not only to have this analysis of trends, but to progress onto an understanding of the root causes and themes, benchmarking, and how that can inform the selection and implementation of quality improvement methods.

While many providers have developed impressive dashboards to monitor and improve quality and safety, more could be done at all levels to combine a range of data to identify the places where risks are most likely to arise. This ‘triangulation’ is important because, for example, it is difficult to draw strong conclusions about trends in the number of patient safety incidents reported by a service without understanding its composition of patients in terms of acuity, levels of detentions and provision of clinical interventions, and the skill mix of the available workforce. We were heartened to see these factors being brought together in innovative digital dashboards that also included diagnostic formulations, therapeutic interventions, patient reported outcome measures (PROMs) and clinician reported outcome measures (CROMs), all extracted into a single dashboard and reducing the burden of repeated data entry.

We heard that there was an appetite among providers to benchmark their services against other similar services, but we heard that, despite the valued presentations and outreach offered by organisations such as NHS Benchmarking Network, sufficient data often is not shared between providers to benchmark with sufficient rigour.

We also heard from staff at all levels, from ICS mental health leads, provider board members, senior managerial and professional leaders and others about their desire to receive shared training in the use and interpretation of ‘data for improvement’. They also requested a review of ‘mandatory training’, both in terms of the overall curriculum and the learning content so that it incorporated local data as well as the more educational content.

Data on its own is not enough

While nearly everyone we spoke to welcomed the focus on improving the use of data and evidence in mental health inpatient settings, we were also told frequently, especially by experts by experience, including carers, that data on its own is not enough - it has to lead to action. As one expert put it, “The question is ‘who cares?’ Who is actually going to do something about it?”

We were told again and again that the best way to identify where things are really going wrong is for leaders at all levels to visit wards and see for themselves. The data can help leaders to prioritise, identify the wards that may need the most attention, and establish the right questions to ask, but it is not a substitute for spending time walking the floor. We heard that the most effective leaders make regular, unannounced visits to wards and do so at all hours, including during night shifts.

We also heard about the importance of ‘soft’ intelligence - as one leader put it, “whispers on the wind” - and the need for leaders to act on information from all sources, rather than relying on one dashboard or data set. National safety leaders told us that there is clear evidence about the wards where ‘closed cultures’ are most likely based on known risk factors, including wards for people with learning disabilities and autistic people as well as children and young people inpatient services.

Recommendations

Recommendation 1

NHS England should establish a programme of work, co-produced with experts by experience and key national, regional and local leaders, including Care Quality Commission (CQC), ICSs, provider collaboratives, independent safeguarding bodies, professional bodies, provider representatives and third sector organisations, among others, to agree how to make sure that providers, commissioners and national bodies are ‘measuring what matters’ for mental health inpatient services, and can access the information they need to provide safe, therapeutic care. This programme should:

• consider what metrics need to be collected, shared and used at different levels to drive improvements in care quality and safety in mental health inpatient settings by the end of 2023. This work should build on the themes identified in the safety issues framework and pay due regard to inequalities. The output of the ‘measuring what matters’ work should then inform ongoing improvements to quality and safety oversight and support arrangements

• consider what enablers are needed to reduce burdens, improve data sharing and timeliness of reporting - based on co-produced principles to support a reduced data burden at all levels

Recommendation 2

Every provider and commissioner of NHS-funded care should have access to digital platforms that allow the collection of core patient information and associated data infrastructure to allow timely reporting of information to different decision makers. These systems need to:

• be compliant with the Digital Technology Assessment Criteria (DTAC)
• meet the requirements of the Digital Capability Framework (DCF) for mental health electronic patient records (EPRs)
• ensure usability, with effective workflows and interfaces to reduce administrative burden

The digital platforms and supporting data infrastructure must allow submissions into relevant national data sets, directly or through other interoperable platforms, and facilitate data flows between systems of different local provider organisations to support joined-up understanding of care pathways. These systems should allow the data collected to be made available to different decision makers, including CQC, at the appropriate level of aggregation and without requiring duplicative submissions, and allow benchmarking across trusts and independent sector providers.

NHS England’s Transformation Directorate should scope out options for how this ambition could be delivered, including cost implications and a value for money assessment to help providers meet this aim specifically for mental health, including specific ways in which mental health electronic patient record improvement and data sharing can be prioritised and interdependencies with other systems and programmes of work. These options should be presented to DHSC by the end of December 2023. DHSC and the NHS should continue to implement the commitments set out in Data Saves Lives, the data strategy for health and care aimed at tackling the cultural, technological and legislative barriers to better sharing of data across the health and care system.

Recommendation 3

ICSs and provider collaboratives should bring together trusts and independent sector providers, along with other relevant stakeholders such as independent safeguarding bodies, across all healthcare sectors to facilitate the cross-sector sharing of good practice in data collection, reporting and use. This forum should showcase examples of how data and information could be gathered and used to improve patient safety and quality of care and reduce the data burden on staff, including the ways that digital solutions can enable these improvements. It should also facilitate the rolling out of examples of good practice and digital innovation between all data commissioners and both NHS and independent sector providers, including the use of administrative staff and coding specialists to gather and process data, the use of consistent codes to record clinical activity, the design of optimal service pathways, and the use of analytical resource to process data and draw out trends and insights to inform quality improvement.

Recommendation 4

More work is needed to map the full range of data on deaths, including what is collected by which organisation and what can be done to improve it. DHSC, in partnership with NHS England and CQC and supported by key experts from across governmental and non-governmental organisations, should convene all the relevant organisations who collect and analyse mortality data to determine what further action is needed to improve the timeliness, quality and availability of that data. This follow up should be completed no later than autumn 2023.

Recommendation 5

Provider boards have a vital role to play in ensuring safety and quality of care in mental health inpatient pathways. We recommend the following actions to improve boards’ capacity to identify, prevent and respond to risks to patient safety:

• every provider board should urgently review its membership and skillset and ensure that the board has an expert by experience and carer representative
• every provider board should ensure that its membership has the skills to understand and interpret data about mental health inpatient pathways and ensure that a responsive quality improvement methodology is embedded across their organisations. They should expect those skills to be at least to a level that matches that of financial literacy on the board. They should review and update their recruitment and annual review processes in line with the recommendations of the Kark review of the fit and proper persons test. This should ensure that people with the necessary competencies, including data literacy skills, are appointed to the board and these skills and competencies are updated. Boards should consider annual mandatory training for their members on data literacy, in partnership with their local ICS and other system partners. Every board should provide Mental Health Act training so that at least half their non-executive directors are trained as associate hospital managers under the Mental Health Act and participate in hearings to best understand the clinical care provided, the challenges, and the views of patients, families and clinical teams for the patients
• CQC should assess and report on whether the membership of the boards of providers of mental health inpatient services includes experts by experience (including carer) representatives and whether boards are maintaining an appropriately high level of data literacy and quality improvement expertise on mental health inpatient pathways among their membership as part of their assessments
• every provider board should urgently review its approach to board reports and board assessment frameworks to ensure that they highlight the key risks in all of their mental health inpatient wards, as set out in the safety issues framework, and that they support the board to take action to mitigate risks and improve care, including both quantitative data and qualitative ‘soft intelligence’ such as feedback from patients, staff and carers. Provider boards should also set out in writing how they will make sure the voice of carers and family members is heard both at board level and with clinical staff and make sure this information is publicly available
• NHS England should review and update the guidance on board assessment frameworks

Recommendation 6

Trust and provider leaders, including board members, should prioritise spending time on wards regularly, including regular unannounced and ‘out-of-hours’ visits, to be available to and gather informal intelligence from staff and patients. Priority should be given to those units where there is a known higher risk of ‘closed cultures’ including services for people with learning disabilities and autistic people, children and young people, those with cognitive impairment and neurodevelopmental conditions and patients held under powers of detention.

Recommendation 7

All providers of NHS-funded care should review the information they provide about their inpatient services to patients and carers annually and make sure that comprehensive information about staffing, ward environment, therapeutic activity and other relevant information about life on the wards is available. CQC should assess the quality, availability and accessibility of this information as part of their assessment of services.

Recommendation 8

ICSs and provider collaboratives should map out the pathway for all their mental health service lines to establish which parties need access to relevant data at all points on the pathway and take steps to ensure that data is available to those who need it. To facilitate this, ICSs and provider collaboratives should make sure that their members have access to data literacy training relevant to mental health, including in relation to quality improvement and safety. They should also bring together the mental health population leads from across their footprint to map out the mental health needs of their local populations and the potential for primary, secondary and tertiary prevention as well as equitable access to safe therapeutic services.

Recommendation 9

ICSs will develop system-wide infrastructure strategies by December 2023 and the mental health estate needs to be fully incorporated and represented in these strategies and in subsequent local action plans. This recommendation is for local ICSs to review the mental health estate to inform these and future strategies, recognising there are evidence-based therapeutic design features that can contribute to reducing risk and improving safety. The review should include:

• identification of critical and significant safety issues and major derogations from National Health building notes, in particular where ligatures or unsafe observation areas are present. Including, where appropriate, updating the Estates Returns Information Collection (ERIC) to ensure that returns are thorough and underpinned by up-to date site surveys and identify safety risks relating to mixed sex accommodation
• a parallel identification of current capital plans which will reduce or remove these estates risks - for example, the eradication of mental health dormitory provision by March 2025, and the plans for mental health safety work announced at recent fiscal events
• a collaborative approach across the ICS and within individual providers to interrogate estates data to inform capital plans and investment priorities
• identification of additional beneficial work that is vital to the inpatient estate’s capacity to provide modern therapeutic interventions including self-management sensory rooms, rooms suitable for therapies and group psychological interventions, rehabilitation, occupational therapies rehabilitation treatments, and faith rooms
• provision for safe family rooms for visiting with children and other dependents, and a room in which a family member can stay overnight, especially when a young person is admitted for a first episode of illness

Recommendation 10

Ward visitors, whether unpaid carers, family members, friends or advocates, play an important role in providing feedback regarding the care provided and escalating any concerns. Providers should review their processes for allowing ward visitors access to mental health inpatient wards with a view to increasing the amount of time families, carers, friends and advocates can spend on wards. DHSC should consider what more can be done to strengthen the expectation for all health and care providers in England to allow visiting.

Recommendation 11

All providers of NHS-funded care should meet the relevant core carer standards set by the National Institute for Health and Care Excellence (NICE) and Triangle of Care, England. Regulators, including CQC and professional regulators, should consider how to monitor the implementation of these carer standards, especially where there is greater risk of unsafe closed cultures developing. ICSs should consider how to routinely seek carer feedback. Inpatient staff training programmes should identify how they can benefit from carer trainers. For patients detained under the Mental Health Act, families and carers should be part of all detention reviews.

Recommendation 12

Professional bodies, such as the Royal Colleges, should come together across healthcare sectors to form an alliance for compassionate professional care. This multi-professional alliance should:

• work together and learn from each other to identify ways to drive improvement in the quality of compassionate care and safety across all sectors, including mental health services, and how they can support staff to provide it
• along with their specialist data units, where they exist, contribute to the work set out in recommendation 1

Recommendation 13

Except where specified, these recommendations should be implemented by all parties within 12 months of the publication of this report. Government ministers, through the Department of Health and Social Care (DHSC), should review progress against these recommendations after 12 months.

Our vision for a better future

Throughout the review, we have heard from a diverse range of experts about what needs to be done to improve the way data and evidence is used to ensure safe, high quality care in mental health inpatient pathways. Using what we have heard, we have devised a set of principles that, taken together, set out a vision for the future where the potential of data and evidence is fully exploited so that the healthcare system is able to ensure the highest standards of care in all mental health inpatient wards and pathways.

We do not pretend that the recommendations in this report alone will achieve that vision. However, we believe that, when implemented, our recommendations will both mean a significant improvement to the way that data and evidence is used in the short term and lay the groundwork for the much more widespread use of the sort of sophisticated embedded digital solutions, such as natural language processing and artificial intelligence, that can safeguard patients and enhance the quality of care in the future.

Improving mental health information and information systems is ultimately about keeping patients safe and providing consistent, high quality, evidence-based treatments. Our vision will mean that key decision makers at all levels have the information they need to make informed and timely decisions and to improve safety and care.

Principles for the collection, analysis, sharing and use of data and information about mental health inpatient pathways

1. Data should be entered once and used multiple times.

2. All data entered should be analysed by informatics experts who can create high quality products to support decision making and promote safety and quality of care.

3. Anyone who enters data into a system should benefit from entering that data, for example, by having key information fed back to them in an accessible format in real time.

4. Patients, carers and staff should be able to provide feedback on care quickly, easily and anonymously, and should expect that feedback to be reviewed and to be used to improve services.

5. Those responsible for patient safety and quality of care in inpatient settings should agree ‘what matters’ for measuring safe and therapeutic care, and focus all data requests and infrastructure around collecting, using and sharing those measures. Inspired by UNESCO (The United Nations Educational, Scientific and Cultural Organisation) principles for the use of data in education^{[footnote 1]}, these data collections should:

• get the right data: collect the data that is relevant to safe, therapeutic care
• get the data right: collect data with precise definitions and appropriate measurements
• get the data right away: collect data that is timely and current
• get the data the right way: collect data through a rigorous process that can guarantee data quality and ensure consistency
• get the right data management: collect reliable data that is guaranteed by good quality control conducted by relevant stakeholders

6. There should be a clear and explicit rationale for all data requests, including why they cannot be met through other collections.

7. All decision makers should have the skills they need to interpret data on safety and quality of care and should be able to use those skills to draw insights from the data they are given and to act on those insights.

8. The data collected on patient safety and quality of care in mental health pathways should have consistent, nationally agreed definitions.

9. Patient data should follow the patient between services, so that patients only have to provide their information once and to support joined-up care between different services.

10. The insight and experience of patients, carers and staff is vital to understanding the risks and quality of care in services and should be integral to decision making at every level.

11. As far as possible, mental health services and ICSs should be joined up with other health and non-health sectors, such as police and local authorities, to facilitate the sharing of data and information to promote patient safety and to better meet the needs of local populations.

What a better system would look like

Decision makers at all levels would have access to robust, consistent, real time data so that:

• they could identify the wards and services most at risk and intervene to raise standards and protect patients, including those settings where closed cultures are most likely to develop, before those risks materialise
• they could learn from incidents in their own localities and nationwide more quickly and comprehensively
• they could have a better and more up-to-date understanding about what needs to be done to improve their services
• more thorough and comprehensive benchmarking would be possible between providers and integrated care boards (ICBs) so that everyone could understand how they compare to their peers and identify trends and areas for improvement
• ICSs could have clear, reliable indicators for decision making to understand and develop commissioning and collaboration plans to meet local population need
• innovations, interventions, and policies could be more easily assessed based on patient outcomes - this could also allow for greater equity of access and address unwarranted variation between services
• there could be improved accountability for leaders and decision makers in mental health inpatient settings
• frontline staff would spend substantially less time entering data and have much more time to spend providing therapeutic care, improving outcomes for patients

Key facts

This section aims to provide additional context to the report, including key facts about the inpatient population and the inpatient mental health sector. This is based on publicly accessible information on inpatients. Information on other relevant topics has been excluded from this section if an inpatient breakdown is not publicly available. Some of the facts included may be affected by data quality issues. Data provided is the latest available at the time of writing - some of the reasons for the time lags are discussed in other sections of the report.

NHS England estimates that during 2021 to 2022, 97,200 people were admitted as an inpatient with NHS-funded secondary mental health, learning disabilities and autism services in England (see Mental Health Bulletin, 2021 to 2022 Annual report).

An estimated 84% of inpatient admissions were into NHS settings and 16% into non-NHS settings^{[footnote 2]}. NHS settings accounted for 67% of in year occupied bed days, while non-NHS settings accounted for 33%.

69% of people admitted were recorded as coming from a white ethnic background, 9% were recorded as black or black British, 6% were recorded as Asian or Asian British, 3% were recorded as mixed, and 3% were recorded as ‘other ethnic groups’ - ‘not known’, ‘not stated’, and ‘unknown’ accounted for the remaining proportions.

30% of those admitted were recorded as from the most deprived quintile^{[footnote 3]}, 23% were recorded as from the second most deprived quintile, 17% were recorded as from the third most deprived quintile, 13% were recorded as from the fourth most deprived quintile, and 10% were recorded as from the least deprived quintile. The deprivation quintile for the remaining 6% was unknown^{[footnote 4]}.

When people are admitted as an inpatient with NHS-funded secondary mental health, learning disabilities and autism services in England, they are admitted into different types of inpatient units. NHS England estimates that, of the inpatients admitted during 2021 to 2022, 56% were admitted into adult acute beds, 15% were admitted into adult specialist beds, 14% were admitted to older adult acute beds, 2% were admitted into children and young people acute beds, while 1% were admitted into children and young people specialist beds. The remaining percentage were recorded as ‘invalid’^{[footnote 5]}.

There are multiple routes through which a person can be referred for admission into an inpatient service. NHS England estimates that during 2021 to 2022, the recorded categories with the highest number of admissions included ‘core community mental health’ services (68,100), ‘other mental health services’ (62,500)^{[footnote 6]} and ‘crisis and acute mental health activity in community settings’ services (61,100). The services with the greatest proportion of people in contact admitted as inpatients were ‘forensic services’ (41%), ‘early intervention team for psychosis’ (24%) and ‘crisis and acute mental health activity in community settings’ (21%)^{[footnote 7]}.

Some patients are formally detained in hospital under the Mental Health Act. NHS England snapshot data reported that approximately 16,200 people were detained in hospital as of 31 March 2022, of which nearly a third (29.6%) were detained under Part III of the act, meaning that they were involved in criminal proceedings or serving a sentence. NHS England estimates that 79% of detainees were detained with an NHS provider and 21% were detained with an independent provider (see NHS England Mental Health Act statistics).

NHS England estimates that, over 2021 to 2022, detention rates for black and black British people were over 4 times higher than that of white people under the Mental Health Act. NHS England also estimates that detention rates in the most deprived areas (153.3 detentions per 100,000 population) were more than 3.5 times higher than the rate of detention in the least deprived areas (42.1 per 100,000) - (see NHS England Mental Health Act statistics ).

CQC registers, inspects, and regulates NHS and independent sector healthcare providers. The range of organisations that provide inpatient mental health and learning disability services that come under CQC’s regulatory scope include NHS trusts (including mental health trusts, acute trusts and community trusts), independent sector providers, and community interest companies (CICs).

For specialist mental health services, CQC gives ratings for each ‘core’ service. These core services include 6 inpatient mental health and learning disability inpatient service types^{[footnote 8]}. CQC gives ratings for 5 key questions for each of the core services they inspect, one of which relates to ‘are services safe?’. Based on CQC ratings data as of 1 March 2023, the inpatient mental health inpatient ‘core’ services were rated as follows on safety (see CQC care directory with ratings):

• 77% of the NHS trusts and 59% of the independent sector locations with ‘acute wards for adults of working age and psychiatric intensive care units’ had a ‘requires improvement’ or ‘inadequate’ safe rating
• 26% of the NHS trusts and 46% of the independent sector locations with child and adolescent mental health wards had a ‘requires improvement’ or ‘inadequate’ safe rating
• 37% of the NHS trusts and 62% of the independent sector locations with forensic inpatient or secure wards had a ‘requires improvement’ or ‘inadequate’ safe rating
• 33% of the NHS trusts and 42% of the independent sector locations with long stay or rehabilitation mental health wards for working age adults had a ‘requires improvement’ or ‘inadequate’ safe rating
• 47% of the NHS trusts and 35% of the independent sector locations with wards for older people with mental health problems had a ‘requires improvement’ or ‘inadequate’ safe rating.
• 31% of the NHS trusts and 47% of the independent sector locations with learning disabilities or autism had a ‘requires improvement’ or ‘inadequate’ safe rating

As reported in the National Confidential Inquiry into Suicide and Safety in Mental Health (NCISH), in 2020 there were an estimated 70 suicides by mental health inpatients and an estimated 186 in the 3 months after discharge from mental health inpatient care in the UK (excluding Northern Ireland) (see National Confidential Inquiry into Suicide and Safety in Mental Health Annual Report 2023).

Based on CQC death notifications data, CQC estimates that in 2021 to 2022, out of the 270 deaths of patients detained under the Mental Health Act, 165 died of natural causes, while 50 died of unnatural causes, and 55 deaths were undetermined (see CQC Monitoring the Mental Health Act in 2021 to 2022).

CQC carried out a review of reports on patient safety incidents that staff had submitted through the NHS National Reporting and Learning System (NRLS) between April and June 2017. Out of nearly 60,000 total reports from NHS trust mental health wards, 1,120 were sexual incidents. More than a third of these (457) could be categorised as sexual assault or sexual harassment of patients or staff (see CQC (2018) Sexual safety on mental health wards).

Across the full range of health and social care services delivered or commissioned by the NHS or local authorities in England, people who present with behaviour that challenges are at higher risk of being subjected to restrictive interventions. Many restrictive interventions place people who use services, and to a lesser degree, staff and those who provide support, at risk of physical and or emotional harm (see DHSC (2014) Positive and Proactive Care: reducing the need for restrictive interventions).

The types of restrictive interventions that NHS England records include physical restraint (for example, prone, standing, kneeling), chemical restraint (for example, rapid tranquilisation, oral), mechanical restraint, seclusion and segregation. NHS England estimates that during 2021 to 2022 the restrictive interventions that the greatest number of inpatients were subject to were ‘physical restraint - standing’ (6,600 inpatients), ‘seclusion’ (4,500 inpatients), and ‘physical restraint - seated’ (4,200 inpatients), while 3,500 patients were subject to ‘physical restraint - prone’ (see Mental Health Bulletin, 2021 to 2022 Annual report).

There are several routes through which a person can register a complaint about the quality of a service, such as the NHS written complaints system, complaining directly to CQC, and through the Parliamentary and Health Service Ombudsman. CQC reported that in 2021 to 2022 they received 2,430 new contacts regarding the use of powers or how duties had been carried out under the Mental Health Act, which comprised a combination of complaints and matters dealt with as requests for advice. In addition, there were 6,500 contacts relating to open cases over the same time period (see CQC Monitoring the Mental Health Act in 2021 to 2022).

Methodology

For this rapid review, we aimed to gather as many views from as many people as possible. Over an 8 week period, we engaged with over 300 people with a variety of expertise relating to mental health inpatient settings, including experts by experience (including carers), frontline staff (including clinicians), data experts, board members, CEOs, chief clinical information officers, non-executive directors, commissioners, regulators and policy makers. We also drew on a range of existing evidence to support our evidence gathering and develop our conclusions.

The review team was composed of a secretariat provided by DHSC supporting the chair, Dr Geraldine Strathdee. The secretariat consisted of policy professionals and analysts, and was augmented by an independent mental health data and analysis expert identified by the chair.

We used a variety of approaches to make sure we could make the most of the expertise available to us and get the views of as many people as possible. These are set out below.

Experts by experience

The views of experts by experience, that is, people with lived experience of inpatient mental health services and their carers, were central to the development of this report, particularly for helping us to understand what good care looks like and the elements that promote patients’ safety on wards.

We commissioned Rethink Mental Illness to conduct a series of workshops and one to one interviews with experts by experience, including children and young people, people with a learning disability, autistic people and people from ethnic minority backgrounds, based on questions provided by the review team. Members of the review team also joined the workshops to hear from experts by experience directly. Rethink provided the review team with a summary of the discussions, which we have used to inform this report.

Deep dives

We conducted several ‘deep dives’ with individual trusts and independent sector (IS) providers to get the perspective of staff from ‘board to floor’ and to gain a better sense of how data and information flows between different parts of a range of trusts and how it is used to improve inpatient care. We selected the trusts for our deep dives on the basis of:

• size of trust
• CQC inspection rating
• geographical location

This ensured we spoke to a range of trusts with different perspectives on the use of data and intelligence. We also selected the independent sector providers on the basis of the range of services they offered. We selected 4 trusts and 2 independent sector providers for our deep dives.

During our deep dives, we conducted a range of semi-structured interviews with individuals and groups, including trust leaders and non-executive directors, clinical directors, chief clinical information officers, regional commissioners, ward managers and frontline clinical staff, to help us understand how the data value chain operates within trusts and between trusts and national and regional bodies - how the data and evidence is collected, processed, used and fed back to staff, the trust board and leadership team, and other parts of the healthcare system.

Focus groups, group meetings and one to one engagement

We spoke to over 150 experts in a range of fields in focus groups, group meetings and one to one engagements. They include leaders from national statutory healthcare and regulatory bodies, mental health trusts, independent sector providers, ICBs, provider collaboratives, professional bodies, third sector experts and other key experts from around the country. These engagements took the form of semi-structured interviews, which allowed the team to gather the insights we needed for the review while also leaving open the opportunity for our interlocutors to provide us with new or unexpected insights. We also received around 50 written submissions to the review from the people and organisations we spoke to and elsewhere.

Working group and advisory group

We established 2 groups to support and guide the review. Our working group, which met regularly during the review period, consisted of national and regional system experts, who were vital in helping to inform the direction of the review, provide expert insight about where we should be focusing our attention, and test our thinking and findings as they emerged.

Our advisory group consisted of our third sector partners and experts by experience, and was co-chaired by Mark Winstanley, CEO of Rethink Mental Illness. This group met twice, once at the beginning of the review to help us set the direction and priorities for the review and once towards the end of the review period to help us test and refine our recommendations.

Safety issues framework

As a building block for the rapid review into data on mental health inpatient settings, we developed a ‘safety issues framework’. This was both to check that there was a shared understanding of what we mean by safety in inpatient mental health and learning disability and autism services, and to serve as a basis for mapping the available data and indicators in relation to poor safety outcomes and the factors that contribute to these (See Appendix 1). The framework was developed through drawing on:

• a literature review on safety in inpatient mental health and learning disability settings, including factors that lead to abuse
• national policy and guidance including:
  • the NHS Five Year Forward View for Mental Health
  • NHS Patient Safety Strategy
  • Mental Health Act code of practice
  • national guidance on reducing the use of restrictive interventions
  • duty of candour regulations
  • relevant NICE guidelines and quality statements
  • Making Safeguarding Personal (Local Government Association)
• findings from published research and national clinical audits including:
  • the National Confidential Inquiry into Patient Outcomes and Deaths
  • the National Confidential Inquiry into Suicide and Safety in Mental Health
  • Getting it Right First Time (GIRFT) acute, rehabilitation and children and young people reports
  • published reports such as CQC’s report on sexual safety on mental health wards
• published data including from the National Reporting and Learning System on reported patient safety incidents
• the approaches to assessing safety within the system, most notably CQC’s single assessment framework, their guidance on identifying and responding to closed cultures and their published brief guides which set out professional guidance, legal requirements or recognised best practice guidance about particular topics in order to assist inspection teams
• the feedback from our evidence gathering and the advice from our expert working group which informed the structure of the framework to align with frameworks for conceptualising risks to patient safety

Data mapping, indicator mapping and workshops

In the early stages of the review, we held a workshop with data experts from a range of organisations to gain an understanding of the key data sources currently available at a national level. We also looked at these key data and data sources to understand the data flows at a national level.

The findings from the workshop and assessment of data flows were also used to develop the safety issues framework. Using the framework, we mapped the key data and data sources, as well as the indicators the data had been used to develop, to the different themes identified in the framework. This helped us identify gaps and potential areas for development (see Appendix 2).

During the review we also held 2 consensus workshops to explore the key issues we wanted to address:

• data on deaths among people receiving treatment in mental health inpatient pathways. This workshop examined the current approaches for measuring and managing mortality in mental health services with a wide range of experts. We have summarised the available national data on deaths in inpatients settings in Appendix 2

• the indicators that experts across the health system think are the most important for measuring patient safety in inpatient pathways. This workshop focused on ‘measuring what matters’, getting a range of perspectives on what matters for safe, therapeutic care and what we needed to change to get to a place where we measured those things effectively

Findings

In this chapter, we summarise what we learned through the course of the review. We have organised our findings into 5 key themes:

1. Measuring what matters
2. Patient, carer and staff voice
3. Freeing up time to care
4. Getting the most out of what we have
5. Data on its own is not enough

Each of these themes draws together a range of issues that were described to us over the course of the review. It should be noted that these themes are not mutually exclusive, but interdependent, and many of our findings cut across more than one theme.

We have also included a section that focuses on data on deaths in mental health inpatient pathways. This reflects the particular importance of this issue - every death in an inpatient mental health setting is a tragedy - but also the challenges that we identified over the course of the review around the collection and use of data on deaths that merit specific consideration.

In our findings, we refer to specific actors and organisations involved in the provision of healthcare. More information about these organisations can be found in in Appendix 3.

Measuring what matters

When we set out on this rapid review, one of our objectives was to understand whether data and intelligence are collected in such a way as to identify risk factors for inpatient safety and aid our understanding of whether people receive high quality care in a safe and therapeutic environment, and to identify ways that we can use data better so that decision makers have the information they need. To begin this process, we had to establish what supports the provision of safe, high quality care, and then set out to understand whether that was being measured and if so, how. Later in this report we lay out our safety issues framework (see Appendix 1), which illustrates what could be measured to support safe and therapeutic care, and our data and indicator mapping work (see Appendix 2), where we have established what is currently measured. However, in this section, we focus on what we were told about the current issues and what more needs to be done.

We found that commissioners, trusts, providers, regional bodies and national organisations are not always measuring what matters for patient safety in mental health inpatient pathways. We found that too often the focus is on indicators that measure activities, processes and when things have gone wrong, such as instances of self-harm, rather than the factors that could contribute to those events, including factors that could lead to the development of closed cultures.

We heard that safe care is therapeutic care. Patients should have access to a range of daily therapeutic activities, provided by the right blend of staff, and have their admission, leave and discharge managed effectively. One carer told us of their great sadness about how their loved one ‘seemed to lose their personality and their hope and any ambition for their future’ when they were held on wards where there were limited therapeutic activities. Many people told us that there should be a greater focus on measuring outcomes for patients and on patient experience, such as patient reported outcome measures (PROMs), patient reported experience measures (PREMs) and clinician reported outcome measures (CROMs).

We heard from clinical leaders and heads of professional bodies that data that measures the quantity and type of therapeutic intervention is often not available. They welcomed examples of where commissioning of specific pathways included the therapeutic interventions to be delivered and the outcomes to be measured. Likewise, where innovation is flourishing, for example, in the development of digital first pathways, these are often inclusive of PROMs and CROMs. We heard from the national audit and GIRFT programmes about major variations in accessible information, appropriate analyses, the delivery of clinical therapeutic practices and even the use of the powers of compulsory detentions under the Mental Health Act.

We found some providers were doing impressive work using, for example, natural language processing tools and patient experience feedback apps to help them gather and analyse qualitative data to gain a better understanding of the factors that contribute to safety risks.

We heard that patient safety data should be made available in real time for those that need to make accountable decisions for the purposes of prevention, early alerts, responsive safety actions and quality improvement.

We found that there are some key gaps in the availability of routine data sources to produce indicators to monitor certain safety issues. These include:

• the voice of people in hospital and their carers about the safety and quality of care and their involvement in care
• data on the culture on wards and staff attitudes towards people in hospital
• data on key aspects of safe therapeutic care, such as PROMs and CROMs

We found that there are data sets with the potential for relatively good information and potential for development, including the Mental Health Services Data Set (MHSDS), the patient level data set that is nationally mandated for providers of specialist mental health care, as well as a range of data sets relating to complaints and whistleblowing. However, the people we spoke to expressed concerns about how often some data collections are run and described problems with quality, coverage, accessibility and timeliness of publication. This included national clinical audits related to inpatient care which have provided invaluable data about access to and take up of clinical interventions, assessment of physical health needs, and outcomes for inpatients. There is a major programme of work to improve the coverage and quality of MHSDS data along with plans to reduce time taken to publish statistics from 12 to 6 weeks.

We found that information about the physical health needs of patients is not always collected and that physical health services are not always well integrated into mental health inpatient pathways. We also heard that, even where data about the physical health is collected, it is not always given the same priority by national and regional bodies, which can influence the attention being paid to physical health needs. This means that patients’ physical health needs sometimes go unmet. We were told that this issue is compounded by staff in mental health inpatient pathways being unable to access patient records from primary and acute services so staff are often unaware of historical issues, especially where the patient lacks capacity to share necessary information about their health.

We found that data and information about patients’ protected characteristics, including ethnicity data, are often not collected or triangulated with other information about patients, meaning that opportunities to address inequalities or to learn valuable information about the people accessing the mental health inpatient pathway are often missed.

Patient, carer and staff voice

From the outset, we wanted to place patient, carer and staff voice at the centre of this review. One of our objectives was to understand how the experiences and views of patients, families, staff and advocates relevant to mental health inpatient services are collected, analysed, collated and used. We heard repeatedly that the voices of patients, carers and staff were essential to understanding where risks to patient safety could develop and how to address those risks, especially in those settings where poor or closed cultures were at risk of developing. However, patients, carers and staff were also uniquely able to tell us what matters for patient safety and quality of care, and their input was invaluable to the development of the safety issues framework.

We learned that patients and carers do not always feel that their voices are heard on wards. In many cases, we were told that it was not clear how patients and carers could provide feedback on their care. Where mechanisms were clearly available, experts by experience - both patients and carers - told us that they felt action often was not taken as a result of their feedback, and that sometimes concerns or feedback were ‘blocked’ from being sent to senior managers by middle management, especially in settings with a poor or ‘closed’ culture. We were told that, when carers or patients raised concerns with CQC, they often felt that action was not taken, or was not fed back clearly. Several carers also told us that they or their loved ones were afraid to raise concerns because of fears about the possible impact for the patient on the ward - they ‘didn’t want to make things worse.’

Experts by experience noted that advocates could be helpful when it came to making patients’ voices heard, but that the quality and availability of advocates can be variable.

While we heard that there can be challenges to gathering data and evidence from patients, including some settings being reliant on paper-based collection of data, several trusts are trialling digital solutions to gather feedback from patients in real time and use it to inform senior leaders and boards about patient experience. We found these are best used as a supplement to, rather than a replacement for, other means of gathering feedback from patients and that ultimately leaders need to prioritise spending time walking their wards and talking to patients. This is critical in identifying those locations where closed cultures are more likely to develop.

We found that comprehensive, robust and timely information about the patient experience on wards and in inpatient pathways was not always available to key decision makers.

Carers reported knowing that there were National Institute for Health and Care Excellence (NICE) and Carers Trust Triangle of Care standards, but that these standards were not always implemented. They wanted to know how they could support their loved one in their care and care planning, play their part in advocacy and support the development of services. One carer told us, “We see the challenges services are facing, the high use of bank and agency staff on wards, and we want to be seen as partners in our loved one’s care, not the enemy.”

We found that carers in particular value information about what life is like on the ward, including staffing, daily routines and therapeutic activity. One carer told us, “When you first enter this whole world, you can feel very lost and helpless as a carer.”

We were told that frontline staff do not always feel that their voice is heard, or that they are empowered or able to speak up when they see things going wrong. While providers have Freedom to Speak Up Guardians, we found that sometimes staff felt intimidated by the process or worried about the impact of speaking up, especially where the Guardian was a senior manager. Staff told us that when they had raised concerns or issues they did not always feel that action was taken, or, where action was taken, that the outcome of the action was not well communicated. However, this was far from universal, with some staff telling us that they did feel confident raising concerns, including positive feedback for the Freedom to Speak up Guardians where they had been implemented well. We found that staff value clear, anonymised feedback routes. They also value hearing about action taken based on feedback.

We found that ward culture can also have a significant impact on staff’s confidence or ability to raise issues - for example, we heard in a closed or toxic culture staff may be discouraged from reporting issues, whereas in more open, learning cultures staff may be more likely to be encouraged to raise their concerns in the spirit of learning and improving.

Freeing up time to care

At the outset of the review, ministers emphasised that one of the fundamental issues we should be considering through this process was how to reduce the burden of data entry on staff, and frontline staff in particular. We learned a great deal about the sheer scale of the data burden on staff - for example, we were told that frontline staff often spend as much as half their shift entering data. This has several consequences, including on staff turnover and burnout. We also, perhaps most importantly, found an impact on the care of patients, as the amount of time staff can spend providing therapeutic care to patients is reduced, which in turn could itself contribute to risks to patient safety. This is why we have named this section ‘Freeing up time to care’.

We found that frontline clinical staff spend too much time entering data. Staff often have to enter the same data multiple times into multiple different systems and often those systems are outdated, difficult to use and do not communicate with each other. In many cases, we found that the data entered was ‘process focused’ or ‘task focused’ to provide reassurance to senior managers and to regulators that certain routine actions had been taken. We also found that, in some places where providers had introduced additional data collection requirements to service local dashboards, there was a risk that this could add to the data burden on staff, reducing further the time they had available to spend with patients.

We heard that commissioners, such as ICBs and provider collaboratives, often ask for different information from each other and from what is collected in local or national data sets. This leads to a lack of consistency and results in a significant extra burden on both frontline clinical staff and the resources of providers. We found that this could be a particular challenge for independent sector providers as they often cover multiple areas and can fall within the commissioning footprint of several ICBs and provider collaboratives. This was described by one senior leader as a ‘wild west’ situation, where duplicative requests for data are made by different parts of the system with little consistency or coordination.

We found that independent sector providers, on average, are less likely to submit comprehensive data to national and regional data sets than NHS providers, despite it often being a contractual requirement and necessary to support benchmarking. Independent sector providers told us that the data burden is very high, and that they do not get value from the data they provide.

While it was clear that data and information provided by frontline staff was critical to ensuring safety, there was widespread agreement that this should be streamlined as far as possible, focused on the most important things, and tools should be made available to make entering data as easy as possible. We heard some examples of trusts that had been able to make some progress on this issue, including one trust that had managed, through negotiation with their local ICBs, provider collaborative and NHS England, to reduce the amount of data they submitted from 6,500 performance indicators to 3,000.

Getting the most out of what we have

At the start of the review, we were tasked with understanding what data is collected in mental health inpatient settings, how it is used and how the use of data could be improved to identify those settings where patient safety might be at risk. Through the course of the review, we found that there is already a substantial amount of data collected on mental health in general and on inpatient settings in particular - as one expert we spoke to put it, ‘mental health services are swimming in data’. We have identified potential areas for development in existing national data sets in Appendix 2. In this section, we set out what we learned about the way data is processed and used, and the ways that could be improved.

We heard that there are often significant time lags in processing, uploading and sharing data. While data is often entered in real time by frontline staff, the work involved in processing, analysing and sharing information, exacerbated by the multiplicity of data sets and lack of interoperability between different systems, means that data presented to boards, commissioners and national data sets were often weeks or months out of date.

We found that frontline clinical staff often get very little value for the data they enter. Several frontline staff that we spoke to referred to feeling like they were ‘feeding the beast’ and that they did not receive useful outputs from the data they entered. Where information was available to staff, it was often not presented in a useful format and was confusing and time-consuming to navigate. This made it difficult for them to spot important trends or issues as they arise, or to understand the perspectives of the patients in their care. We heard that frontline staff need and want clear and up-to-date information about the patients on their wards and about how well their ward is doing in comparison with others. They told us that they need this information given to them in a clear and easily decipherable format, and do not want to spend time having to extract these insights themselves from dashboards. They also told us that they are particularly interested in feedback from patients about their experience on their ward to help them learn and improve.

We found that there is often not sufficient analytical resource in trusts to service the needs of all staff, from frontline clinical staff to leadership and board members. This meant that we found that the needs of senior leaders and national data sets were sometimes prioritised over the needs of frontline staff and that information was not always provided to key decision makers in a useful and actionable format. We were told by one trust leader that some analysts spend as much as half of their time flowing data to national data sets. This sometimes means that non-analysts, such as clinical staff, do the data entry and processing, which is not the best use of their time or expertise and often not part of their skillset.

We heard that there was a lack of consistency across England in the way that key measures about patients were coded and recorded on systems, including information about their needs, diagnoses and therapeutic interventions. We were told by several clinical leaders that the varying use of ICD 10, ICD 11 and SNOMED coding meant that the same information could be recorded in different providers in different ways, and that they would value clarity on which they should use. We found that this variation led to a greater burden on staff at all levels and that it risked the collection of high quality data that could be used to generate improvements.

We found that key information relating to the quality of care and patient safety in mental health inpatient pathways was not routinely shared. This means that benchmarking with peer trusts is often not possible, and opportunities to learn from others and improve services are often missed. We heard that providers are sometimes reluctant to share information for fear of inviting negative attention from NHS England, CQC or the press. We have also heard that the absence of a robust national picture of service provision that builds up from ward level makes it difficult to develop reliable comparative analysis, particularly in relation to different types of inpatient services.

We found that key information about patients does not always follow the patient through their therapeutic journey. This means that risk factors relating to the patient sometimes are not immediately identified or communicated to frontline staff on admission to the ward or to community services on discharge from the ward. It also means that patients and carers often have to give the same information multiple times to multiple members of staff, which increases the burden on staff and can be distressing or frustrating for patients and carers. We found that while some key systems, such as the electronic patient record, were designed to support the sharing of data, in practice they were time-consuming and cumbersome.

Key information about patients is also often not passed between agencies, including between trusts, when a patient moves between different areas. This means that sometimes the full picture about the needs of a patient is not available, or there are delays in getting the right information. We heard that this was a particular challenge for independent sector providers, as they often cover multiple areas and admit patients from different parts of the country. We heard that it is important to make sure that key information about a patient’s care needs is shared between all health and social care partners, including commissioners and providers. Trusts, independent sector providers, ICBs and provider collaboratives should also explore how they can share more data with other local services, such as local authorities and education providers, to support the patient on their therapeutic journey.

We found that data and information about patients’ care and safety risks are often not triangulated effectively at ward level. This means that opportunities to identify trends and emerging risks are sometimes missed. We heard that not all trusts had the technology or resources to develop integrated dashboards and found that there may be benefits to trusts working together and sharing their expertise to ensure consistent, high quality assurance across all mental health inpatient pathways. We also found that, where dashboards had been developed, they had been developed to meet the needs of individual trusts or providers and were not easily scalable to regional or national level. Nevertheless, several trusts and independent sector providers have developed excellent dashboards that provide rich information about the quality and safety of care in their services. We heard these work best when the data in the dashboard is triangulated at ward level, as this allows insights to be drawn about care at a level as close to the patients as possible.

Through our data and indicator mapping, we have identified difficulties in bringing together a comprehensive picture of aspects of inpatient safety across the NHS and independent sectors at the national level, particularly where there are different reporting requirements, routes for reporting or where requirements to participate in data collections differ. Due to existing protocols, published data that does not distinguish between missing data and low numbers or which rounds figures limits the ability to undertake accurate analysis.

We heard that data transparency is essential, and that providers needed to make more data more widely available. This will allow more benchmarking of different services and will help to increase cooperation across different services to drive improvement. We found that NHS England and CQC have a role to play in fostering a supportive culture of mutual transparency where data can play a role in driving up standards across the country. This will also support CQC in the implementation of its new single assessment framework.

Colleagues representing the Independent Safeguarding Adults Boards Chairs Network told us that there is wide variation on the quality of safeguarding concerns being reported from mental health inpatient services. We heard that the availability of information was often dependent on the quality of the professional relationships within local systems and willingness of organisations to be open and transparent.

Data on its own is not enough

While we were struck by the level of enthusiasm for and engagement with the review’s focus on data and evidence, the people we spoke to at all levels, especially experts by experience, were clear with us that data on its own is not going to improve the quality and safety of care that patients receive in mental health inpatient settings. Leaders at all levels, and especially in trust and provider boards, need to take understand the risk factors in their own services and to take action to address them.

We found that key decision makers at every level often felt like they did not have the skills or capacity to make the best use of the data and information available to them and were therefore not able to gain the insights they needed. This sometimes led to requests for additional data from trusts, providers and staff, some of which are duplicative and which added to the burden on staff. We heard that, even where data is triangulated and presented clearly, a lack of skills and training means that staff at all levels are often unable to identify when a change or a trend is significant.

We found that the data collected by trusts, providers and regional and national bodies is only valuable if it is used well and leads to action. Leaders at all levels need to make sure they have measures and training in place to use data and information to identify where there might be an issue and then should be swift, rigorous and meticulous in investigating and resolving it. This includes in particular making regular, unannounced visits to wards, including out of hours and during night shifts.

We heard that data has a critical role to play in supporting system-wide quality improvement (QI) at commissioner and provider level. Commissioners and providers should work together to establish QI programmes that drive up the quality of safe, therapeutic care in mental health inpatient pathways ensuring that staff have time to undertake this work. One advanced provider told us they were trialling a standard where no more than 20% of staff time was spent in data collection to free up time to care and for Quality Improvement.

Data on deaths in mental health inpatient settings

Over the course of the review, we spoke to a wide range of people about the way that data about deaths among people who use mental health inpatient services is collected, analysed and used, including holding a consensus workshop that brought people together to dive into the issues and agree how the collection, analysis and use of the data can be improved. A breakdown of the range of organisations that collect and use data on service user deaths in mental health inpatient services, as well as the data that is either shared with providers or other system partners, or available in the public domain, can be found in Appendix 2. It is important to note that we did not just consider deaths by suicide and related deaths, but all deaths among people in mental health inpatient pathways, including those as a result of physical health needs.

We found that there are several organisations that collect and report on deaths of people with mental health problems and on people with a learning disability but that these collections are fragmented, which presents significant challenges in providing an overview of how many people die while in contact with inpatient services and the cause of their deaths. There is no published national overview of the deaths of people in inpatient mental health settings nor of the total number of deaths of people in contact with mental health services at provider level. However, there are opportunities through the linking of mental health, health and mortality data sets to provide this overview including the causes of people’s deaths due to both physical health conditions and due to suicide.

We heard that providers are not always made aware when someone has died, particularly if they have gone absent from the ward or died while on leave, after transfer to an acute hospital or following discharge home or elsewhere. Providers have no ready access to information about the cause of a person’s death to enable learning and have to follow up each person’s death to find out the cause. We also heard that the definitions applied to produce suicide statistics and delays in death registration mean that figures are thought to be an underestimate and there is likely to be attrition at each stage of the process, particularly if deaths are not referred to a coroner in line with national guidance or if there are variations between coroners about reaching a conclusion in relation to suicide.

Not all deaths are required to be reported nationally and the reporting requirements differ for NHS and independent providers - for example, we heard that not all deaths will be reported by NHS providers to national incident reporting systems as there may not be an associated patient safety incident. We heard that this leads to challenges in getting a complete picture of reported deaths where there are multiple reporting routes. We also heard that there are legislative barriers which prevent some organisations accessing available data for monitoring or safeguarding purposes.

The Office for National Statistics (ONS) produces a range of mortality statistics, including the main causes of deaths using the International Classification of Diseases but there are variations in the way deaths are categorised by other organisations. We were told that the use of natural and unnatural in relation to deaths can be unhelpful in that they relate to the way someone has died rather than the cause of death. We also found that there may be delays in determining cause of death, particularly if a person’s death requires referral to a coroner, with implications for the timely production of statistics and learning.

National statistics are often reported annually and there can be a time lag in relation to the data they reflect, which means that it can be very difficult to get access to comprehensive data on deaths in a timely fashion. There are a number of initiatives that have tried to address this through the use of real time surveillance including the Suicide Prevention Programme, which has used this approach to track the number of probable suicide deaths locally and regionally, and the National Child Mortality Database, which collects data on all deaths by suicide among children and young people. However, careful thought needs to be given to the most appropriate methods for the analysis of deaths in mental health inpatient settings given they are comparatively rare events.

Case studies

During the course of the review, we identified several case studies of innovative practice that help to illustrate how specific providers have attempted to address some of the issues we have identified, both in England, the UK and internationally. These case studies are not intended as endorsements of particular products or approaches, but to demonstrate some of the opportunities provided by new technologies and innovations.

For ease of reference, we have arranged these according to the key themes we set out in the executive summary, above.

Measuring what matters

The examples in this section set out how clinicians, data experts and policy makers tried to ensure that they and their institutions are ‘measuring what matters’ and taking account of therapeutic care as a key driver of safety in mental health inpatient pathways.

We were shown a range of safety, quality improvement and early alert dashboards by the trusts and independent sector providers we spoke to, and we are grateful to everyone who showed us the great work they are doing. One such example is from St Andrew’s Healthcare’s integrated safety and quality improvement platform. Over the last 5 years, their clinical, information and technical leads have been developing a dashboard that brings together into a coordinated and triangulating platform data from all their internal data collections. This includes data on safe staffing, level of acuity, rates of observations and safety incidents, patient self-reported assessment and progress reports for both physical and mental health, delivery of therapeutic care interventions, PREMs, PROMs and CROMs. The dashboard can be used to report at the level of patient, ward, unit, and organisation, improve patient safety and care, enhance decision making, improve data analysis and support quality improvement. Embedded digital tools such as tablets and apps are used where possible so that staff can gather information in a low burden way, and in a structured format that can, in real time, provide information on safety trends. The safety framework ward dashboard has been used to support reductions in the use of seclusion and restraints. They are also developing a ward ‘heatmap’, built into the ward dashboard, that can flag wards of concern from a safety perspective.

Community mental health teams in Mersey Care NHS Foundation Trust are using a management and supervision tool (MaST) that uses predictive analytics to examine data and information from multiple sources to detect risk of crisis and inform decision making based on service user needs and the likely resource required to provide safe, high quality and effective care. MaST allows staff to prioritise their interventions and move to a preventative rather than treatment focused model of care. Adapted versions of MaST have been launched in early intervention teams and older adults services as well as Mersey Care’s social prescribing ‘Life Rooms’, engaging with a subset of service users early to help identify any unmet needs to reduce the likelihood of people’s health deteriorating, potentially avoiding the need for mental health crisis services. Data on PROMs has also been incorporated. Realised benefits include a reduction in hospital admissions, improved compliance with 72 hour follow up, reduction in caseload sizes, and improved discharge management. MaST’s adoption is estimated to have resulted in a cost saving of £1.7 million in the 6 month period after MaST was introduced, according to an independent evaluation (PDF, 1MB) of the service. The team are currently evaluating whether it can be rolled out in mental health inpatient services.

Patient, carer and staff voice

In this section, we have described some examples of how healthcare providers have tried to improve the way they gather and use insights and feedback from patients, staff and carers, including how to gather these insights in as close to real time as possible.

In Australia, the New South Wales (NSW) Ministry of Health runs a ‘Your Experience of Service’ (YES) survey. This is a consumer experience measure tool that aims to improve mental health services in community as well as hospital settings from the perspective of service users. It is offered on paper and electronically to NSW hospital and community service users. The collection of patient feedback has been ongoing since 2015. They receive around 20,000 to 25,000 returns a year, with good representation of age, gender, Aboriginal consumers and people receiving involuntary care. This data is then used to publish a ‘Your Experience of Service’ annual report. Questions are set up to be easy to interpret and be answered using simple rating methods. In terms of data collection, the processes are standardised across community and hospital settings. All data is all funnelled into one central database. This means that the data can be reused effectively internally for improvements and to benchmark against others. There is also a requirement that services provide data monthly to an analytics branch, so there is a continuous stream of data input which can potentially be accessed in ‘real time’ for early alerts in deterioration of patient experience.

This example comes from Berkshire Healthcare NHS Foundation Trust and iWantGreatCare. Berkshire Healthcare’s vision is ‘to be recognised as the leading community and mental health service provider by our patients, staff and partners’. Outstanding patient experience is at the heart of this strategy. Berkshire foundation trust developed a standardised patient experience measurement framework to be used across all their 160 services in children’s, community and inpatient mental health. They undertook a patient and staff consultation to develop their live patient experience tool, focusing on what patients want to talk about, when, how, and how often. They heard that patients want to be given a choice of how they provide feedback, which is vital to ensure feedback mechanisms are accessible for all. Patients also wanted to see that their feedback is being used to make service improvements and want this regularly communicated. This would, in turn, drive engagement, increase participation and enrich feedback. Based on what they were told by patients and carers, they partnered with an independent healthcare review organisation, iWantGreatCare, to develop a set of 7 questions that help them to gather consistent, reliable and comparable information about patient experience. Details of the project, including the consultation and the list of 7 questions, are available on Berkshire foundation trust’s website.

Freeing up time to care

In this section, we identify some examples of the steps that providers have taken to try to reduce the burden on staff and free up time to care by streamlining processes or introducing new technologies.

Through their programme ‘Shaping our Future’, East London NHS Foundation Trust (ELFT) embarked on work to redesign the way they delivered services to meet the post-COVID ‘new normal’ and improve standards of care, co-designed with experts by experience and frontline staff. Through this work, they identified that a lot of frontline staff and business intelligence capacity was spent on data entry and reporting of a total of 6,500 contractual key performance indicators (KPIs). ELFT and its integrated care system commissioners and partners came together to review the 6,500 KPIs. They agreed a set of guiding principles for the basis on which data should be gathered and established task and finish groups that included commissioning, operational, quality and performance leads to review the KPIs and agree changes in line with the principles. Within the first year of the work, the number of KPIs that ELFT reported against was reduced from 6,500 to 3,000. ELFT found that, in addition to other benefits, they were able to improve the way they provided data to their commissioners, with greater use of statistical process control. Staff and patient time was freed up to continue to engage with quality improvement in both hospital and community settings.

Getting the most out of what we have

In this section, we describe some examples of national and international initiatives that aim to maximise the value from the information that providers and others already collect.

Transformation Partners in Health and Care (formerly Healthy London Partnership), in conjunction with stakeholders and the NHS Benchmarking Network, designed and commissioned the London mental health dashboard in 2016 - one of the first publicly available dashboards. It was initially created as a strategic planning tool, a trusted single reference point and source of data for London’s mental health system. Bringing together 276 metrics from 27 sources of data, it serves as a benchmarking tool allowing comparisons between different systems, organisations and providers across London to support service and quality improvements.

‘Deep dive’ reports are produced to summarise the trends and current status of health and wellbeing of the population in the London region. The dashboard also enables bespoke reporting at provider level. The London mental health dashboard has been commended for its data transparency, use of multiple data sources in one accessible point and for its inclusion of wider determinants of health to give a holistic overview of the health system. It has been beneficial to identify trust and regional level issues and where improvement can be driven collaboratively. We have included this case study as an example of how trusts can come together with each other and with other organisations to share information. We were told that this has helped the participants understand their populations better and identify areas where services can be improved.

People with psychosis form the highest proportion of those admitted to inpatient wards. The Psychosis Population Health Management Platform was developed by South London and Maudsley NHS Foundation Trust in collaboration with clinicians, academics, informaticians, computer scientists, biostatisticians and researchers. The programme demonstrates how inpatient and community data can be used by individual clinicians, teams and organisations. It demonstrates the potential for ICBs to use data in a way that enhances clinicians’ ability to improve clinical outcomes in the local population through targeted prevention and by using an individualised risk calculator to support early intervention^{[footnote 9]}.

At this stage the platform is based on mental health trust inpatient and outpatient information, but in the future it can be used to identify what treatments will best help patients discharged home and to primary care, and support clinicians to deliver both mental health and physical health treatments. The use of advanced analytics support such as artificial intelligence is being considered. We heard that, by using a robust data set from an advanced model which combines structured data elements with Natural Language Processing output, and innovative and interactive visualisations based on open standards through the CogStack platform, the Psychosis Population Health Management Platform has transformed the way South London and Maudsley NHS Foundation Trust can scrutinise local clinical data and participate in integrated care.

To find out more, watch the YouTube video, ‘Psychosis Population Health Management Platform’ from Dr Rob Harland at the Kings Health Partners annual conference, 2021.

Watch the Psychosis Population Health Management Platform video on YouTube.

Measuring what matters: what should be measured in relation to patient safety in mental health inpatient pathways

Throughout the review process, we have been seeking to understand what matters when it comes to preventing and reducing risks to patients in inpatient mental health pathways. What we have heard consistently from all levels is that currently data collection focuses too much on measuring adverse events and process measures and not enough on preventative factors and clinical outcomes for patients. This is not to say that measuring adverse events is not important and that information should not be collected - in fact, that data has contributed to some important changes, such as the work to remove ligature risks from inpatient settings.

To this end, we have identified the key issues that should be measured when seeking to identify and address patient safety risks in mental health inpatient pathways. This is the safety issues framework, below. This set of issues is not exhaustive and further work is required to identify what exact metrics should be measured, by whom, and when. However, it is intended to guide future work at all levels so that everyone involved in the provision of care in mental health inpatient pathways understands what they need to consider when seeking to address risks to patients.

The guiding principle for these indicators is that safe care is therapeutic care. While ‘safe’ and ‘therapeutic’ may suggest differences of focus or activity, we have found that in reality the 2 concepts are connected and should not be thought of separately - you cannot have one without the other.

We explored the question of ‘what safe care looks like’ in collaboration with our experts by experience panel and our carers panel, who provided vital insight into what matters to them for patient safety and what should be measured.

Our panel told us that safety is made up of both physical elements, including having positive ward environments, safe levels of staffing and robust safeguarding procedures to protect people, and emotional, relational and psychological elements, such as feeling supported, having well-run wards with positive environments, and positive relationships with staff.

Several participants said that the state they were in during their mental health crisis led to stigma from staff when they were well, which affected their care and safety.

All participants agreed that trust is crucial for care and for safety and that trust is about continuity of care and security. One participant said that safety is also about accountability from staff and institutions.

The qualifications of staff were also mentioned - people tended to feel more safe when staff were more qualified (for example, those who were qualified in trauma informed care), or when more senior qualified staff were leading and supporting less qualified staff. One participant said, “It’s not just about numbers, it’s about the quality of staff.” The issue of overrepresentation of agency staff on wards was mentioned as a major factor. One panel member said, “If you see agency staff, you think this is going to be a bad shift.” Participants valued having a regular relationship with staff to support therapeutic engagement.

Many participants said that a structure in place for the day helped with their sense of safety and good care.

Many participants said that having one to one sessions and a range of therapeutic activities was a key part of feeling safe and heard in inpatient wards. One participant told us:

One to ones are necessary and there is a lack of them. There is no psychologist in the ward I am on, so we have no therapeutic intervention, that is crucial to your recovery. How am I going to recover if I don’t have this support in place?

Physical restraint was raised as a major issue, with one participant saying, “restraining is a big issue. Unsafe restraints happen often and I personally have been hurt in restraint.” We heard this could also be an issue for staff, with some staff reporting feeling traumatised by restraint events.

Participants also told us that understanding patients’ physical health needs and comorbidities is vital for keeping them safe. More than one told us about patients being put on medication that caused them to put on weight and affected their physical health.

Our panel told us that the key risk factors that influence safety include:

• the frequency and quality of observations
• understanding patients’ physical as well as mental health needs
• learning from self-harm and injury
• the competencies and blend of skills in the staff multidisciplinary team
• hearing from and learning from staff and patient views at all levels
• the ward culture

The safety issues framework

As a building block for the review, we have identified a range of safety issues relevant to inpatient mental health care. The range of safety issues has been based on a brief review of recent literature on safety in inpatient mental health and learning disability settings, including factors that lead to abuse, with reference to relevant national policy and guidance and informed by the approach to assessing safety within the system and feedback through our evidence gathering.

Our consultation with stakeholders indicated that we should take a broad view of safety in these settings to include evidence-based protective factors and approaches that promote safety as well as risk incident factors. This incorporates factors that might result in harm to people in hospital, their visitors and staff as well as considering the harm that may result from a lack of therapeutic safety, poor culture and unsafe governance. Although the main focus is on safety in the inpatient setting, the review is considering the pathway of care from admission to discharge and the risks to people after discharge from hospital given this is a time of high risk to safety.

There are a number of frameworks for conceptualising risks to patient safety and for evaluating incidents^{[footnote 10]}. With reference to these, we have summarised the risks to inpatient safety differentiating between: 

• poor safety outcomes where there may be the greatest risk of harm or which could result in injury or death
• individual, organisation and practice factors which can contribute to poor safety outcomes, including the harm that can be caused by the use of unsafe practices where people are admitted to hospital without effective therapeutic care

The safety issues framework is set out below, and Appendix 1 contains further detail of the topics which relate to each of the key safety issues. We have used this framework as a basis for mapping available data, indicators and analysis, to identify potential gaps, and to inform our recommendations.

Poor safety outcomes

Poor safety outcomes relevant to mental health inpatients include:

• abuse
• assault and conflict
• self-harm
• sexual safety
• patient accident
• other incidents that may result in harm
• negative impact on physical health
• negative impact on mental health outcomes
• death

Factors which may contribute to poor safety outcomes

Factors that contribute towards poor safety outcomes among mental health inpatients include:

• profile of people using the service (context)
• service characteristics (context)
• lack of service user and carer involvement
• lack of therapeutic clinical care and treatment
• inappropriate use of coercive or restrictive interventions
• unsafe ward environment and lack of safe, therapeutic estates
• lack of safe physical health assessment and treatments
• lack of safe, high quality and effective staffing
• lack of positive and therapeutic culture
• unsafe medicines management pathways and practices
• unsafe inpatient care pathway
• poor learning culture
• lack of leadership for safety and governance
• high data burden but lack of real time information for rapid improvement

Appendix 1: the safety issues framework - mapping of topics to key safety issues for mental health inpatient settings

A: poor safety outcomes

Abuse

Includes:

• harassment and intimidation
• discrimination
• verbal abuse
• threats of harm
• financial abuse
• bribery
• neglect

Assault and conflict

Includes:

• violence, aggression, conflict and physical assault leading to harm or injury to staff or patients
• behaviour that challenges staff

Self-harm

Includes:

• self-harm
• suicidal behaviour
• self-neglect
• self-starvation

Sexual safety

Includes:

• sexual harassment
• sexual exploitation
• sexual assault

Patient accident

Includes:

• falls
• injuries

Other incidents which may result in harm

Includes harm caused by:

• restraint
• illicit drug taking and/or alcohol use

Negative impact on physical health

Includes:

• worsening of health conditions or new health conditions which are not identified, monitored or treated
• unsafe medicines management
• illness due to outbreaks of infection (for example, COVID-19)

Negative impact on mental health outcomes

Includes emotional and psychological distress.

Death

Includes death due to:

• suicide
• self-harm
• preventable death due to physical health problems
• drugs and alcohol
• homicide
• self-neglect
• accident
• treatment, procedures or errors in care

B: individual and practice factors which can contribute to poor safety outcomes

Profile of people using the service (context)

Including:

• protected characteristics
• inherent risk factors related to people being highly dependent on staff for their basic needs, being less able to speak up for themselves without support, being unable to leave the service of their own accord (for example, subject to the Mental Health Act or the Mental Capacity Act) or remaining in the service for months or years
• diverse people with diverse needs on wards

Service characteristics (context)

Including:

• isolated service with limited access to community services and facilities and less opportunities for friends and family to visit
• decline in safety and overall performance and lack of adherence to national guidance and standards

Lack of service user and carer involvement

Including:

• lack of individualised and personalised care, including culturally appropriate care
  plans which are not co-produced
• not involving people to manage risks
• not promoting independence, choice and control
• lack of access to independent advocacy
• lack of active involvement of carers, friends and families
• lack of involving service user and carers as patient safety partners

Lack of therapeutic clinical care and treatment

Including:

• lack of delivery of evidence-based care including holistic assessment of needs, care planning, diagnosis, consent to care and treatment, provision of appropriate therapies
• ineffective risk assessment and risk management
• poor observation practices
• ineffective relational security
• ineffective procedural security (for example, appropriate application of search procedures)
• ineffective application of Mental Health Act and Mental Capacity Act
• clinician reported outcome measures (CROMs), patient reported outcomes measures (PROMs) and patient reported experience measures (PREMs) not in use
• inappropriate use of electroconvulsive therapy (ECT)

Inappropriate use of coercive or restrictive interventions

Including:

• excessive, increasing use or repeated uses of restrictive interventions including restraint (physical, chemical, mechanical), seclusion, segregation; lack of following guidance if interventions are used including debriefing after incidents
• not taking action to prevent the escalation of situations that lead to the use of restrictive interventions. Not using restrictive interventions as a last resort
• being on a locked ward with high bed occupancy
• application of blanket restrictions

Unsafe ward environment and lack of safe, therapeutic estates

Including:

• unsuitable, untherapeutic estate with design of features that can increase risk (for example, hidden corners) and lack of access to facilities that can promote safety (for example, availability of sensory rooms)
• unclean and poorly maintained estate with ineffective infection prevention and control
• lack of provision of equipment, facilities and technology to promote safe care
• dormitory accommodation not eliminated
• lack of choice of single sex accommodation
• ligature risks not mitigated or eliminated
• services in an isolated location which limits access to community services and facilities and limits opportunities for friends and family to visit

Lack of safe physical health assessment and treatments

Including:

• lack of physical health assessment
• not monitoring adverse effects of medication
• not investigating causes of poor physical health
• not responding to deteriorating health and managing long-term conditions

Lack of safe, high quality and effective staffing

Including:

• poor therapeutic culture and ethos of staff team
• staffing levels, capacity and skill mix (including balance of registered and unregistered staff) unable to meet service user need
• lack of continuity of staffing, high use of agency staff and high sickness, vacancies and turnover
• lack of provision of regular support, training and supervision
• ineffective recruitment practices
• lack of promotion of staff well-being
• low staff morale and satisfaction
• staff feeling afraid to go onto the wards
• whistleblowing alerts

Lack of positive and therapeutic culture

Including:

• people not being treated with kindness, compassion, dignity and respect
• service users unsatisfied with care
• carers unsatisfied with care
• negative staff attitudes (for example, absence of caring values) and behaviour (for example, punishing regime) towards service users
• lack of provision of and engagement in activities, including on evenings and weekends
• lack of access to leave, or leave being badly managed, leading to unauthorised leave
• lack of access to one to one time with staff
• not embracing new ideas, external visitors and good community connections
• increasing number and ineffective follow up of safeguarding incidents
• lack of interprofessional collaboration

Unsafe medicines management pathways and practices

Including: 

• ineffective administration and management of medicines
• dispensing medicines out of line with national guidelines
• ineffective monitoring

Unsafe inpatient care pathway

Including:

• admission to hospital that may have been avoidable or short lengths of stay where admission may have been inappropriate
• lack of timely access to an appropriate bed in the least restrictive setting and as close to home as possible
• inappropriate out of area placements
• inappropriate placement of young people on adult wards
• ineffective management of admission to hospital and transfers between services
• unsafe and delayed discharge, discharge planning not starting when someone is admitted to hospital, excessive lengths of stay
• ineffective communication, information sharing and services not working together to promote safe discharge
• ineffective follow up by community services after discharge

Poor learning culture

Including:

• inconsistent reporting and lack of learning from patient safety incidents and events
• people not informed and supported when things go wrong with care and treatment, lack of honesty and lack of an apology when things go wrong (duty of candour)
• lack of learning from complaints, concerns and compliments
• inconsistent approaches, delayed investigations and insufficient time to carry out thorough investigation of incidents and complaints

Lack of leadership for safety and governance

Including:

• lack of access to or lack of support from Freedom to Speak Up Guardians
• leaders failing to monitor and address issues raised by staff, people using the service, relatives and visitors to the service
• leaders failing to learn and improve from the feedback from those who speak up
• poor support for whistleblowers
• poor management and negative relationships between staff and senior colleagues
• absence of manager or leader
• lack of openness and transparency
• lack of internal oversight and poor governance for safety in inpatient mental health settings

High data burden but lack of real time information for rapid improvement

Including:

• inaccurate data that does not provide a comprehensive view about safety; delays in data submission
• lack of availability of digital technology to support live monitoring
• lack of training and support for staff to enable data literacy
• ineffective use of data to monitor services
• lack of information produced to inform improvement, governance and oversight

Appendix 2: data and indicator mapping

Mapping data sets and indicators to the safety issues framework

We have used the safety issues framework developed for the review as a basis of mapping available national data and indicators relevant to inpatient care and safety. This has helped us to identify potential gaps to inform our recommendations. The tables below present the mapping of national data sets and indicators to the safety issues framework. These have been informed by the workshops we held with a range of external stakeholder data experts.

Our mapping is not fully comprehensive but offers an initial picture of the coverage of data and indicators in relation to the framework, and where there may be gaps. The focus of our mapping is on data that is collected locally by providers and used at the national level to produce provider level analysis, although the findings may be insightful for other levels, such as ICBs and provider collaboratives. Where possible, we have differentiated between the availability of data and indicators for the NHS and independent healthcare (IH) sectors and have noted key issues and areas for potential development.

Note that we use the following acronyms in the tables below:

• Care Quality Commission (CQC)
• children and young people (CYP)
• electroconvulsive therapy treatment (ECT)
• Estates Returns Information Collection (ERIC)
• Freedom to Speak Up (FTSU)
• General Medical Council (GMC)
• Getting It Right First Time (GIRFT)
• Learn from patient safety events (LFPSE)
• Mental Capacity Act (MCA)
• Mental Health Act (MHA)
• Mental Health Services Data Set (MHSDS)
• National Clinical Audit of Anxiety and Depression (NCAAD)
• National Reporting and Learning System (NRLS)
• NHS England (NHSE)
• Office for National Statistics (ONS)
• patient reported experience measures (PREM)
• Patient Led Assessments of the Care Environment (PLACE)
• Royal College of Psychiatrists College Centre for Quality Improvement (Rpsych CCIQ)
• UK Health Security Agency (UKHSA)

Poor safety outcomes

Safety outcome	National data sources and collections	Indicators developed	Areas for development at national level
Abuse	Independent providers - CQC notification NHS providers - NRLS and LFPSE (NHSE)	Changes in pattern of reporting abuse or allegations of abuse - for independent providers only	No comprehensive overview across NHS and independent providers in relation to abuse or allegations of abuse
Assault and conflict	MHSDS (NHSE) Specialised commissioning (NHSE)	CQC (MHSDS) indicators for NHS and independent provider inpatient services on rates of assaults on patient and multiple assaults Specialised commissioning metrics on rates of assaults for patients and staff (for subset of specialised commissioning services)	Data and indicators available for assaults on patients in MHSDS but there is potential for developing more granular analysis on the types of services that these assaults take place in. However, this is dependent on coverage, and MHSDS assault data is not publicly available Limited data and indicators available for assaults on inpatient staff No known source of data for behaviour that challenges staff
Self-harm	MHSDS Specialised commissioning	CQC (MHSDS) indicators for NHS and independent provider inpatient services on rates of self-harm and multiple instances of self-harm Specialised commissioning metrics on the proportion of patients who have self-harmed and have required immediate intervention or treatment (for subset of specialised commissioning services)	Data available and indicators developed although data not publicly available Potential for testing more granular (service specific) analysis for MHSDS
Sexual safety	Specialised commissioning	Specialised commissioning metrics on the rate or proportion of sexual safety incidents (for subset of specialised commissioning services)	No comprehensive overview of incidents related to sexual safety across NHS and independent providers Difficulties with identifying relevant or specific incidents within national incident reporting systems
Patient accident	May be reported to NRLS and LFPSE	No specific indicators identified	Potential for time series monitoring of reported incidents
Other incidents that may result in harm	CQC notifications Specialised commissioning May be reported to NRLS and LFPSE	CQC (notifications) indicators on the changes in pattern of reporting serious injuries and police incidents Specialised commissioning metrics on the proportion of harmful incidents reported, ‘never events’ or serious incidents (for subset of specialised commissioning services)	Potential for further use of MHSDS in relation to injuries resulting from restraint Subject to testing, incidents reported to LFPSE that equate to CQC notifications may be possible to identify for NHS, with the potential for expansion of indicators that CQC apply to independent providers
Negative impact on physical health	No known data available	No known indicators available	Potential for testing use of linked data sets MHSDS and hospital episode statistics (NHSE) to develop indicators May be relevant content from outcome measures if data quality improves
Negative impact on mental health outcomes	No known data available	No known indicators available	LFPSE has strengthened the harm gradings in relation to the psychological impact from incidents. May be relevant content from outcome measures if data quality improves
Infection	Situation reports including for COVID-19 (NHSE) Seasonal flu and COVID-19 vaccination (UKHSA)	No known indicators available relating to hospital acquired infection Proxy measure at trust level, for example. proportion of staff vaccinated for flu	Data published for acute trusts on hospital acquired infections May be potential for further development of analysis relating to infection
Death	MHSDS (linked to ONS deaths data) CQC notifications Deaths where there has been a patient safety incident reported to NRLS and LFPSE Specialised commissioning	CQC indicator on suicides reported for detained patients (NHS and independent providers) CQC indicator on the changes in patterns of reporting detained patient deaths (NHS and independent providers) and other inpatient deaths (independent providers) GIRFT indicators on discharge due to death Specialised commissioning metrics on the proportion of patients that die for subset of specialised commissioning services	No comprehensive overview of the numbers of people who died while in contact with mental health inpatient care, although CQC indicator was developed (currently suspended) and there is potential for further development using linked data sets MHSDS recording of discharge due to death thought to be of poor quality

Factors which may contribute to poor safety outcomes

Contributory factor	National data sources and collections	Indicators developed	Issues and potential for development
Profile of people using the service (context)	MHSDS NHS benchmarking for CYP	NHS benchmarking CYP indicator on admissions by ethnicity and gender Specialised commissioning metric for men admitted to adult eating disorders services Range of measures (GIRFT, NHS benchmarking, specialised commissioning) on admission under the MHA	Patient level data sets offer potential for analysis by protected characteristics No indicators available in relation to other inherent risk factors due to either lack of data or poor data quality No clear view at ward level about diverse people with diverse needs on wards, although MHSDS cluster information may be a potential option
Service characteristics (context)	ONS urban and rural classification CQC ratings Specialised commissioning	CQC indicators on geographic isolation of independent provider inpatient services, sector risk score and time since last inspection Specialised commissioning metric for adherence to forensic mental health services standards	CQC ratings available at inpatient services level which could be used to give an overall view of performance
Lack of service user and carer involvement	PLACE (NHSE) Specialised commissioning NCAAD (Healthcare Quality Improvement Partnership)	CQC indicator based on PLACE score for privacy, dignity and well-being Specialised commissioning indicator for adult secure services on provider engagement with carer support and involvement toolkit NCAAD measures on co-production of care plan, admission information, information about medication, carer involvement	No routine source of national data on experience of service users and carers Some relevant indicators developed by national clinical audits but not collected routinely
Lack of therapeutic clinical care and treatment	MHSDS Specialised commissioning NCAAD	GIRFT data quality measure on recording of diagnosis Range of specialised commissioning indicators relating to Care Programme Approach, Care Education and Treatment Reviews or standardised assessments being carried out for specific services NCAAD measures in relation to referral and receipt of recommended therapies and comorbid diagnosis Various GIRFT, specialised commissioning and NCAAD indicators relating to use or change of score in outcome measures Specialised commissioning indicators for some services on completion of PREM or participation in patient experience exercise	Relevant measures about access to and take up of relevant evidence-based treatments developed for other services. There may be potential for using MHSDS to generate some of these if data quality improvements are made Absence of data and measures that can indicate quality of some practices (for example. observation, relational and procedural security, effective application of MHA and MCA) GIRFT and NHS benchmarking findings indicate variable use of routine outcome measures and use of paired data is rare GIRFT findings indicate there is a wide variation in recording and availability of patient experience data and the majority is based on small samples No known indicators available on ECT although a voluntary peer review network is available for ECT clinics (Electroconvulsive Therapy Accreditation Service - ECTAS)
Inappropriate use of coercive or restrictive interventions	MHSDS NHS benchmarking Specialised commissioning Bed occupancy data (KH03) (NHSE) MHA visit data (CQC)	Range of measures developed by NHSE, CQC, specialised commissioning and NHS benchmarking on use of restraint (including types of restraint), seclusion or segregation Measures developed by NHS benchmarking, specialised commissioning and CQC on bed occupancy CQC indicator for independent services on inpatients on locked wards	Wide range of measures available on use of restrictive interventions, although they use different methodical approaches. The granularity of their analysis is also different Data available on debriefing after restrictive intervention incident but no known indicators developed No known indicators based on use of blanket restrictions
Unsafe ward environment and lack of safe, therapeutic estates	PLACE CQC notification ERIC (NHSE)	CQC indicators based on PLACE score for disability, dementia, and cleanliness of the environment CQC indicator for independent providers on changes in pattern of reporting events that stop services (may not be specific to ward environment)	PLACE does not cover all aspects relevant to safety of the environment, although may be potential for further use of granular information - limited data available for independent providers in PLACE National data collected via ERIC, although the data does not reflect the suitability of the estate No known indicators available in relation to dormitory accommodation, choice of single sex accommodation and ligature risks
Lack of safe physical health assessment and treatments	Specialised commissioning NCAAD	Specialised commissioning indicators for some services on access to GP, primary care and dental services, physical health assessments, physical health plan or body mass index (BMI) change NCAAD measure on offer of intervention where evidence of a physical health problem	Potential for building on indicator development from national clinical audits if data was more routinely available (for example, via MHSDS) to monitor physical health assessments in inpatient mental health services
Lack of safe, high quality and effective staffing	NHS staff survey (NHSE) Electronic staff records (NHS Business Services Authority) NHS benchmarking Specialised commissioning	Range of CQC trust level indicators based on NHS staff survey and electronic staff record but not specific to inpatient care Range of indicators developed by NHS benchmarking and specialised commissioning on staffing levels, sickness absence, spend on bank and agency staff Specialised commissioning (CYP) indicators on care hours from registered nurses and healthcare support workers and ratio of substantive to bank and agency staff Specialised commissioning indicators for some services on supervision and training relevant to specific services CQC indicator on changes in pattern of whistleblowing (independent providers)	No nationally available data on culture on wards Published national data sets on staffing and staff views for NHS cannot be disaggregated to give view of inpatient care and do not include most independent providers Although electronic staff record has provision to collect data on training, data quality is thought to be poor No known data or indicators on the effectiveness of recruitment practices
Lack of positive and therapeutic culture	CQC ‘Give Feedback on Care’ Specialised commissioning NHS staff survey	CQC indicator on bad experiences of care and for changes in pattern of reporting safeguarding for independent providers Specialised commissioning metrics for some specialised commissioning services on completion of patient or carer satisfaction surveys, on proportion of patients being offered meaningful activity, rates of escorted leave episodes and for one service on percentage of investigated safeguarding reports CQC trust level indicator on teamwork (from NHS staff survey) but not specific to inpatient care CQC indicators for NHS and independent providers on rates of unauthorised absences and for independent providers, changes in pattern of reporting unauthorised absence	No mechanism for national collation of feedback from service users or carers or to give a comprehensive view of satisfaction with care or on staff attitudes No known current indicators on access to one to one time with staff, or on openness to embracing external visitors and involvement from the community Available data relating to staff feedback about teamwork not specific to inpatient care
Unsafe medicines management pathways and practices	Specialised commissioning NCAAD Prescribing Observatory for Mental Health Rpsych CCIQ	Specialised commissioning metrics for CYP services on physical monitoring of those prescribed antipsychotics NCAAD indicators on reviews including response to medication and side effects, and on prescribed medications within British National Formulary limits	National clinical audits have developed relevant methods for assessing medicines management but unclear whether there is a source of relevant and robust data outside of the national audit process Relevant indicators likely to have been developed by the Prescribing Observatory for Mental Health but information is only available for members
Unsafe inpatient care pathway	NHS benchmarking GIRFT (NHSE) Specialised commissioning MHSDS Out of area placements (NHSE) CQC notifications Delayed transfers of care (NHSE) NCAAD	Range of NHS benchmarking and GIRFT indicators related to community provision which may have implications for avoidable admissions and profile of people admitted Specialised commissioning indicator for one specialised commissioning service on involvement of Crisis Resolution Home Treatment in admission and for others on waiting times for transfer Range of indicators on bed availability, length of stay, distance to treatment, out of area placements, waiting times, delayed discharges CQC indicator on changes in reporting placement of young people on adult wards Specialised commissioning measures for some specialised commissioning services about the provision of comprehensive information to receiving clinician and GP NCAAD has used measures to assess standards about sending discharge documents to GP within a 24-hour timeframe and on 24 hours’ notice of discharge given to service users and carers NHS benchmarking and specialised commissioning measures (for some specialised commissioning services) on readmission and NCAAD measures on community follow up	Wide range of indicators covering various aspects of the acute inpatient care pathway, although consensus may be needed to identify the best measures. Further work may be needed to ensure the availability of data to monitor the pathway for other types of inpatient services (for example, mental health rehabilitation services) or the application of relevant indicators to all inpatient service types Data collected and published via MHSDS on placement of young people on adult wards, but no known indicator based on this Some measures relating to communication for discharge and on readmissions but not available for all services Published MHSDS data includes metrics relating to 72 hour follow up of patients discharged from inpatient mental health units, so there is potential to use this data as a basis for an indicator
Poor learning culture	NHS staff survey NRLS Central alerting system (Medicines and Healthcare products Regulatory Agency, MHRA) CQC data on reported complaints and MHA complaints NHS written complaints (NHSE) Investigation of complaints by Parliamentary and Health Service Ombudsman	CQC trust level indicators on staff feeling secure about raising concerns about unsafe clinical practice (NHS staff survey) CQC trust level indicators on reporting practices including proportion of patient safety incidents resulting in harm, risk of under-reporting patient safety incidents, comparative reporting rates and median time taken to report (NRLS) CQC risk-based indicators on trust responses to national patient safety alerts Specialised commissioning indicators for some specialised commissioning services on rate of complaints and source of complaints CQC indicator for independent providers on changes in pattern of reporting complaints to CQC	CQC indicators based on NRLS do not give a specific view on reporting practices within inpatient services and give no overview of reporting practices in independent providers The transition from NRLS to LFPSE will have implications for the production of indicators as trusts transition at different times There are a range of data sources relating to complaints, but these are not used as the basis of indicators. There may be difficulties in getting a specific view relating to inpatient care in relation to some sources No known indicators available on the investigation of incidents and complaints
Lack of leadership for safety and governance	FTSU data (National Guardian’s Office) NHS staff survey GMC enhanced monitoring and national training survey Whistleblowing alerts to CQC	CQC trust level indicators based on NHS staff survey scores, GMC data on quality and safety of training and trainees’ satisfaction, and on whistleblowing alerts open for at least 10 weeks CQC indicator for independent providers on presence and absence of registered manager	FTSU data available but no known indicators based on this or on the quality of support for whistleblowers Trust level indicators are not specific to inpatient services and there is no parallel data for independent providers No known data sources or indicators available relating to the governance and oversight of safety in inpatient mental health services
High data burden but lack of real time information for rapid improvement	MHSDS	NHSE publish a range of metrics or information about submission to MHSDS and MHSDS data quality CQC and GIRFT have developed specific indicators based on this	No current data sources or known indicators relating to the availability of digital technology, although NHSE have plans for new digital maturity ratings No current data sources or known indicators relating to training and support to enable data literacy, the use of data to monitor services, or on what information is insightful

Data on deaths in mental health inpatient settings

As part of our data mapping work, we scoped out the range of organisations that collect and use data on service user deaths in mental health inpatient services. The below sets out some of the key organisations, as well as the data that is either shared with providers or other system partners, or available in the public domain.

Key organisations that collect and use data on deaths	What is available in the public domain or shared
Providers	NHS trusts publish information as part of Learning from deaths in the NHS requirements
ONS collects data about registered deaths among the general population	ONS publishes a wide range of national statistics on mortality including on suicide, excess and avoidable deaths. Additional reports are published on further analysis of data (for example, Prevention of Future Death Reports for Suicide submitted to coroners)
National Confidential Inquiry into Suicide and Safety in Mental Health (NCISH), which is based on ONS data on suicides and deaths of undetermined intent. NCISH collects information from clinicians where people have been in contact with mental health services. During the pandemic NCISH set up a system to monitor real time data	NCISH is the official source of statistics on suicides among people with mental health needs and publishes national figures on the numbers of deaths of people due to suicide, including those in contact with services. NCISH identifies national trends and themes, and makes general national recommendations for clinical practice and policy to improve safety (see their 2023 annual report). NCISH share their scorecard with providers and highlight whether the provider is an outlier
Learning from lives and deaths (LeDeR) (NHSE) - collects information about the deaths of people with a learning disability and autistic people to improve care, reduce health inequalities, and prevent early deaths	LeDeR summarises the lives and deaths of people with a learning disability and autistic people who died in England in annual reports (see their 2021 annual report)
NHSE collects data via the NRLS and LFPSE about deaths where there has been a patient safety incident. NHSE also collects data via the Strategic Executive Information System (StEIS) about serious incidents, including those resulting in death	NHSE publishes statistics on patient safety incidents reported to NRLS (and in future LFPSE), including deaths. NHSE also publishes the following: National patient safety incident reports Organisation patient safety incidents workbooks An explorer tool ‘Never events’ figures are published from StEIS
CQC collects data on the deaths of patients detained under the MHA (NHS and independent providers) and about the deaths of other service users (direct notification for deaths in independent provider settings). CQC accesses data on deaths in the NHS through NRLS and LFPSE	CQC publishes information on the deaths of detained patients (for example, annual report to Parliament Monitoring the Mental Health Act COVID-19 Insight reports Data shared with Ministerial board on Deaths in Custody Summary information shared with mental health NHS trusts and Community Interest Companies (CICs) through CQC Insight reports
The Ministry of Justice (MOJ) collects data from manual statistical returns submitted by all coroners to produce statistics about deaths referred to coroners. Coroners collect evidence to determine the cause and manner of death.	The Courts and Tribunals Judiciary publishes Reports to Prevent Future Deaths and the Chief Coroner’s annual report contains statistics available about deaths referred to coroners
The Independent Office for Police Conduct (IOPC) collects data on all incidents involving a death or serious injury while in contact with the police	IOPC publishes annual data showing how many people have died after contact with the police - figures include any deaths of people detained by the police under the MHA
NHSE links MHSDS and ONS mortality data for some analysis (‘deaths asset’)	NHSE publishes analysis on Excess under 75 mortality rates in adults with a serious mental illness which measures the extent to which adults with a serious mental illness (SMI) die younger than adults without a serious mental illness
Office for Health Improvement and Disparities (OHID) uses ONS data and NHSE’s ‘deaths asset’ to produce analysis on the premature mortality and excess under 75 mortality rates in adults with severe mental illness	OHID Fingertips tools are publicly available (see links to the relevant tools for severe mental illness and suicide prevention). The SMI premature and excess mortality indicators are available at local authority level and suicide rates are available at local authority and ICB level

Appendix 3: glossary of terms

Acute care

Acute care is for patients who need specialist support and treatment which cannot be provided in the community during an acute stage of their illness. This can be after an accident, during the most unstable phase of an illness or following surgery.

Acute trust

Acute NHS trusts provide services such as accident and emergency departments, inpatient and outpatient medicine and surgery and in some cases very specialist medical care. They provide secondary care, ranging from relatively small district hospitals to large city teaching hospitals.

Care Quality Commission (CQC)

CQC is the independent regulator of health and adult social care in England. CQC makes sure health and social care services provide people with safe, effective, compassionate, high quality care and encourages care services to improve. CQC registers, monitors, inspects and regulates services and takes action to protect people who use services. If there are concerns with the quality of care from a provider, such as a safety concern, CQC takes action to encourage improvement and hold the provider to account to protect people who use regulated services from harm and the risk of harm, and to ensure they receive health and social care services of an appropriate standard. CQC also has a duty under the Mental Health Act 1983 to monitor how services exercise their powers and discharge their duties when patients are detained in hospital or are subject to community treatment orders or guardianship. CQC’s Mental Health Act reviewers visit and interview people currently detained in hospital under the act, and requires actions from providers when CQC becomes aware of areas of concern or areas that could improve. CQC also has specific duties under the Mental Health Act, such as to provide a second opinion appointed doctor (SOAD) service, review Mental Health Act complaints, and make proposals for changes to the code of practice.

Chief clinical informatics officers (CCIOs)

Chief clinical informatics officers (CCIOs) within a trust take responsibility for driving continuous clinical process improvement focused on patient outcomes and efficiency, ensuring clinical adoption and engagement in the use of technology, and developing clinical information that supports and enhances organisational reform.

Community interest company (CIC)

A community interest company (CIC) is a special type of limited company which exists to benefit the community rather than private shareholders.

Community trust

Community trusts provide community health services to people in the community.

Crisis team

Crisis teams support people who might otherwise need to go to hospital. Some common examples of crisis intervention include suicide prevention telephone hotlines, hospital-based crisis intervention, and community-based mental health services mobilised during a disaster. 

Department of Health and Social Care (DHSC)

DHSC is a ministerial government department supported by a number of agencies and partner organisations. It leads the nation’s health and social care to help people live more independent, healthier lives for longer. DHSC is responsible for supporting and advising its ministers, setting direction to protect and improve global and domestic health, and making sure it and its partner organisations deliver on agreed plans and commitments.

Detention

Detention is when a person is made to stay in hospital and may be treated against their wishes.

Early intervention team in psychosis

Early intervention in psychosis (EIP) teams are multidisciplinary teams set up to seek, identify and reduce treatment delays at the onset of psychosis and promote recovery by reducing the probability of relapse following a first episode of psychosis.

Experts by experience

Experts by experience are people who have recent personal experience of using or caring for someone who uses mental health inpatient services.

Forensic mental health service

Forensic mental health service provides treatment, rehabilitation and aftercare for people who are mentally unwell and who are in the criminal justice system.

ICD 10

ICD 10 is the 10th revision of the International Statistical Classification of Diseases and Related Health Problems which was used as a diagnostic tool.

ICD 11

ICD 11 is the 11th revision of the International Statistical Classification of Diseases.

Integrated care system (ICS)

ICSs are partnerships of organisations that come together to plan and deliver joined up health and care services, and to improve the lives of people who live and work in their area. They are made up of integrated care partnerships (ICPs) and integrated care boards (ICBs).

An ICP is a statutory committee jointly formed between the NHS ICB and all upper tier local authorities that fall within the ICS area. The ICP will bring together a broad alliance of partners concerned with improving the care, health and wellbeing of the population, with membership determined locally. The ICP is responsible for producing an integrated care strategy on how to meet the health and wellbeing needs of the population in the ICS area.

An ICB is a statutory NHS organisation responsible for developing a plan for meeting the health needs of the population, managing the NHS budget and arranging for the provision of health services in their area

Independent sector healthcare provider (ISHP)

An independent sector healthcare provider (ISHP) is a private sector healthcare company that is contracted by the NHS in the provision of healthcare or in the support of the provision of healthcare.

Inpatient mental health services

Mental health inpatient services are for people who can no longer be supported at home and need to be admitted to hospital due to severe mental health problems. These can be functional mental illnesses (such as depression or schizophrenia) or organic mental illnesses (such as dementia).

Mental Health Act

The Mental Health Act 1983 is the main piece of legislation that covers the assessment, treatment and rights of people with a mental health disorder.

Mental health trust

Mental health trusts provide health and care services for people with mental health problems.

National Confidential Inquiry into Suicide and Safety in Mental Health (NCISH)

The National Confidential Inquiry into Suicide and Safety in Mental Health (NCISH) is an internationally unique project collecting in-depth information on all suicides in the UK since 1996.

National Institute for Health and Care Excellence (NICE)

NICE is a public organisation which produces evidence-based guidance and advice for health, public health and social care practitioners.

NHS England (NHSE)

NHS England provides national leadership for the NHS and supports NHS organisations to work in partnership to provide high quality care and better outcomes for patients. This includes supporting all healthcare providers to minimise patient safety incidents and drive improvements in safety and quality. NHS England works with the healthcare system to develop and implement patient safety policy such as the NHS Patient Safety Strategy. The strategy includes the delivery of national programmes and initiatives such as Mental Health Safety Improvement Programme. NHS England also directly commissions a number of specific services that sit outside the provider collaborative model, including high secure hospitals, secure acquired brain injury units and women’s enhanced medium secure units. NHS England has merged with NHS Digital, which means that NHS England has assumed responsibility for running the vital national IT systems which support health and social care, as well as the collection, analysis, publication and dissemination of data generated by health and social care services, to improve outcomes.

NHS trust

An NHS trust is a local body that is set up for the purpose of providing health services in a given geographic area. Trusts can act as providers of healthcare services or as commissioners and sub-contract to other providers.

Occupied bed days

The total number of occupied beds for each day in a period.

Parliamentary and Health Service Ombudsman

The ombudsman makes final decisions on complaints that have not been resolved by the NHS in England and UK government departments and other public organisations. 

Psychiatric intensive care

Psychiatric intensive care is for service users compulsorily detained, usually in secure conditions, who are in an acutely disturbed phase of a serious mental disorder. There is an associated loss of capacity for self-control, with a corresponding increase in risk, which does not enable their safe, therapeutic management and treatment in a general open acute ward.

Provider collaborative

A provider collaborative is an NHS-led group of providers of specialised mental health, learning disability and autism services who have agreed to work together to improve the care pathway for their local population. They include providers of children and young people mental health inpatient services, adult low and medium secure services and adult eating disorder services. They take responsibility for the budget and pathway for their given population. The provider collaborative is led by an NHS ‘lead provider’, which is accountable to NHS England (the commissioner) for reporting, delivering agreed high quality outcomes and performance across all of the specialised services within their provider collaborative.

Quintile

When the values for an indicator are divided into 5 equal groups, each grouping is a known as a quintile. Each quintile represents 1/5 or 20% of the range of values for the indicator. The first quintile represents the lowest 1/5 of values from 0% to 20% of the range.

Restrictive intervention

Deliberate acts on the part of other persons that restrict an individual’s movement, liberty and or freedom to act independently in order to:

• take immediate control of a dangerous situation where there is a real possibility of harm to the persons or others if no action is undertaken
• end or reduce significantly the danger to the persons or others
• contain or limit the person’s freedom for no longer than is necessary

Secondary care

Secondary care includes:

• planned or elective care - usually in a hospital
• urgent and emergency care, including 999 and 111 services, ambulance services, hospital emergency departments, and out-of-hours GP services
• mental health care

SNOMED

SNOMED is a structured clinical vocabulary for use in an electronic health record. SNOMED gives clinical IT systems a single shared language to facilitate the exchange of information between systems.

Specialised mental health services

Specialised mental health services are primarily for people with a serious mental illness or disorder. They provide a tertiary level of care and are commonly highly trained psychologists and psychiatrists who also have decided to specialise in a particular field.

Statistical process control (SPC)

Statistical process control (sometimes known as statistical quality control (SQC)) is the application of statistical methods to monitor and control the quality of a production process. This helps to ensure that the process operates efficiently, producing more specification-conforming products.

Tertiary care

Tertiary care is highly specialist treatment, such as: neurosurgery, transplants, plastic surgery, secure forensic mental health services.

Ward

A group of hospital beds with associated treatment facilities managed as a single unit for the purposes of staffing and treatment responsibilities.

---

1. Source: UNESCO Asia and Pacific Regional Bureau for Education (2011) Systematic Monitoring of Education for All - Training Modules for Asia-Pacific. Bangkok: UNESCO Bangkok: Module A2, pages 8 to 9. ↩

2. Patients can be admitted more than once in any year, and whether they are admitted to an NHS setting or non-NHS setting may differ each time. ↩

3. Quintiles are based on the English indices of deprivation. The English indices of deprivation measure relative levels of deprivation in 32,844 small areas or neighbourhoods, called Lower-layer Super Output Areas, in England. ↩

4. Although a greater proportion of people in inpatient settings are from the more deprived quintiles, the proportion of people in contact with services that were admitted as inpatients was similar across the 5 quintiles (3% across the 4 most deprived quintiles and 2% in the least deprived quintile). ↩

5. NHS England (2022). See Mental Health Bulletin, 2021 to 22 Annual report. Patients can be admitted more than once in any given year, and the bed type admitted to may differ each time. ↩

6. This captures a number of services, including general psychiatric service, substance misuse team, rough sleeping services. ↩

7. NHS England (2022). See Mental Health Bulletin, 2021 to 22 Annual report. Patients can be admitted through multiple routes in a given year and the route of admission may differ each time. ↩

8. Each inpatient ‘core’ service may be provided from one or multiple locations. NHS providers are more likely to have multiple locations from which these services are operated while independent sector providers are more likely to have core services rated for services provided at one location. ↩

9. Implementation of a real time psychosis risk detection and alerting system based on electronic health records using CogStack - PMC (nih.gov) ↩

10. For example, The measurement and monitoring of safety from the Health Foundation, the Yorkshire Contributory Factors Framework, or the SEIPS quick reference guide and work system explorer. ↩