HSSIB investigation report - Mental health inpatient settings: Creating conditions for learning from deaths in mental health inpatient services and when patients die within 30 days of discharge (30 January 2025)

Summary

This is one of a series of Health Services Safety Investigations Body (HSSIB) investigations on the theme of patient safety in mental health inpatient settings. The investigation examined how providers conduct timely and effective investigations into deaths of patients receiving care in inpatient units or within 30 days of discharge. This included a review of local, regional and national oversight frameworks, as well as data collection mechanisms.

The aim of the investigation was to understand how providers learn from deaths, and how they use that learning to improve. The investigation sought to understand the impact on individuals involved in the care of patients who died while in mental health inpatient care or shortly after discharge. It reflects the experiences of families, carers and staff.

The investigation recognises the complex nature of mental health inpatient care and discharge. The findings present opportunities to improve systems and practices in mental health services, with potential relevance to other healthcare settings in England.

Content

The investigation identified significant challenges in maintaining safety, conducting effective investigations, managing data on deaths and ensuring system-wide learning. These findings reiterate findings from other reports on inconsistencies in data reporting, lack of consistent terminology and difficulty in cross-provider comparisons. The investigation found gaps in discharge planning, crisis service accessibility and access to community therapy that were potentially contributing to poor patient outcomes, including deaths.

The investigation highlighted system-level issues in service commissioning, patient flow, integrated working and accountability, compounded by a lack of system-level learning and application. The investigation found there is a culture of blame in which individuals and organisations are afraid about safety investigation processes. The report emphasises the need for a systemic approach to safety investigations and learning with a focus on collaboration, transparency and oversight, with a shift from procedural practices to a culture rooted in empathy, person-centred care and active involvement of families.

Findings

Understanding how providers ensure timely and effective investigations

• Investigations into patient safety incidents in mental health do not always take a system-wide perspective, limiting the ability to capture the full complexity of care.
• The current national framework for incident response faces implementation barriers in mental health settings, due to the differences in care requirements compared to acute physical health settings.
• Training for the implementation of the Patient Safety Incident Response Framework includes developing knowledge of systems thinking and system-based approaches to learning from patient safety incidents. However, some organisations described their training focused on acute physical health contexts and does not sufficiently account for the mental health care context.
• There is no national system to track and ensure the implementation of investigation recommendations, resulting in limited strategic oversight of patient safety investigations and a lack of structured learning for improvement.
• Many families feel marginalised and excluded from the investigation process, experiencing investigation processes as a ‘tick box’ exercise and without a culture of transparency, learning, and accountability.
• Processes for learning from deaths are feared by families, staff and organisations because of a reported focus on blame which does not align with the stated goals of an effective safety culture that is orientated around learning to support systemic change.
• Families often feel excluded from care processes, with their concerns about safety planning and risk mitigation often overlooked, which complicates their ability to help keep their family member safe.
• Legal processes within organisations may unintentionally shut down opportunities for learning, fostering a culture of defensiveness rather than reflection.
• Staff lack the time, permission and safe spaces to support open, reflective conversations about patient safety incidents, which are essential for learning and improvement.
• Patient safety incident investigations, and other associated investigation processes if a death occurs, often do not consider the emotional distress experienced by all affected. This results in compounded harm.

The investigation also became aware of areas of mental health inpatient care where investigations had not effectively addressed ongoing concerns about inpatient mental health care:

• Gaps were identified in discharge planning, crisis service accessibility and community therapy provision, and staff skilled in mental health, resulting in people being left in unsafe situations where they may self-harm.
• There is significant variability in therapeutic engagement and a lack of personalised care which has left some patients feeling hopeless and disconnected.
• The term ‘therapeutic engagement’ may be interpreted differently across mental health services. This has resulted in some approaches becoming clinically focused rather than person focused.
• Providers told the investigation that incidents of people using items of clothing to ligature resulting in catastrophic self-harm was increasing. However, the investigation did not identify specific guidance on how to reduce and respond to non-anchored ligature risks, or on managing access to known ligature risk items.
• Staff face ongoing challenges in balancing ‘least restrictive approach’ policies and the therapeutic benefit of decisions about care, with the need to ensure patient safety, often creating tension in care delivery.

Examining national, regional, and local oversight and accountability frameworks for deaths in mental health inpatient services

• Many previous national recommendations to improve the care of patients with mental health needs have not been taken forward to date, leaving no clear plan for implementing the recommendations.
• Some integrated care boards do not have full oversight of patient safety risks across all the services they oversee. Instead of having a clear, formal structure for accountability in patient safety investigations, they often rely on informal relationships or collaborations between providers.
• This lack of complete oversight can make it difficult to ensure that patient safety investigations are thorough and standardised across different organisations, leading to potential inconsistencies in addressing safety issues.
• Some integrated care boards and regional teams struggle with gathering and analysing data on patient safety due to resource and reporting limitations. This means they do not always have a complete picture of the risks at a system-wide level.
• Data gaps limit the ability of integrated care boards to identify co-morbidities (when patients have multiple health conditions) and understand health inequalities, especially among people with serious mental illness. This creates challenges for integrated care boards to address broader health trends and inequities within their regions effectively.
• The involvement of patient safety partners and people with lived experience in safety meetings is variable, limiting their ability to contribute.
• Some non-executive directors with responsibility for safety and quality struggle to scrutinise and interpret complex data sets on patient safety and deaths due to the volume and lack of triangulation of information presented. This limits meaningful oversight and learning.

Examining the mechanisms that capture data on deaths (and near misses) across the mental health provider landscape, including up to 30 days after discharge

• There is inconsistency in data reporting. Mental health providers report deaths and near misses in varied ways, using different definitions and methods. This inconsistency makes it difficult to compare data across providers and understand overall trends in patient safety.
• There is not a standardised national system requiring providers to report deaths in the same way. This means that each provider’s reports may look different, which reduces the reliability of data for understanding patient safety across the board.
• There is not a single, comprehensive database that includes all deaths and near misses within mental health services, including those occurring within 30 days after a patient’s discharge. This makes it hard to see the full picture of patient safety outcomes and identify patterns or risks.
• There is not a centralised organisation or process effectively overseeing and co-ordinating data on deaths. This lack of oversight limits the ability to identify systemic issues, reduce duplicated efforts, and drive consistent improvements across mental health services.
• There is currently limited co-ordinated effort among organisations that produce data relating to deaths, and individual providers may be collecting and analysing similar data on their own. This leads to duplicated work, wasting time and resources that could be better used if there was improved collaboration. It also results in considerable variation in the data being presented.

Recommendations and observations

HSSIB makes the following safety recommendations:

• Safety recommendation R/2025/052: HSSIB recommends that the Department of Health and Social Care works with NHS England and other relevant stakeholders, to clarify national expectations for meaningful and restorative learning from patient safety events and deaths in mental health services. This is to ensure effective learning is supported through processes that provide high-quality and transparent investigations within a culture of compassion.
• Safety recommendation R/2025/053: HSSIB recommends that NHS England works with other stakeholders to define the term ‘therapeutic relationship’. This is to support building trust and compassionate relationships between staff and patients from admission to inpatient settings through to discharge, to improve patient outcomes.
• Safety recommendation R/2025/054: HSSIB recommends that NHS England, working with other relevant national bodies, develops guidance on how to reduce and respond to non-anchored ligature risks. This will help staff to support people who attempt to hurt themselves with non-anchored ligatures and improve patient safety whilst maintaining a therapeutic environment.
• Safety recommendation R/2025/055: HSSIB recommends that the Department of Health and Social Care creates a national oversight mechanism that supports co-ordination, prioritisation and oversight of safety recommendations to implementation across the system. This is to ensure that recommendations from public inquiries, independent patient safety investigations and other patient safety investigation reports, as well as prevention of future death reports from inquests, are analysed and monitored and reviewed until their implementation using a continuous quality improvement approach to learning.
• Safety recommendation R/2025/056: HSSIB recommends that the Department of Health and Social Care working with NHS England, and other relevant stakeholders, develop a comprehensive, unified data set with agreed definitions for recording and reporting deaths in mental health services to include deaths that occur within a specific time period after discharge. This will support any revisions required to the current NHS England Learning from Deaths Framework. The creation of a comprehensive, unified data set would enhance system-wide visibility, co-ordination and collaboration, reduce duplication of effort, and maximise the impact of improvement work through strategic oversight.

HSSIB makes the following safety observations:

• Safety observation O/2025/057: Integrated care boards and organisations that provide mental health care can improve patient safety by working together to support the facilitation of cross-organisational investigations and learning. This should be achieved in a way that enables people involved in an investigation to come together to share perspectives and build relationships to enable learning. This may provide opportunities for effective and meaningful organisational learning and facilitate reparation and trust-building for everyone involved.
• Safety observation O/2025/058: Organisations that provide mental health care can improve patient safety by adopting a comprehensive person-centred care approach that prioritises the individual needs, preferences and rights of each patient. This approach should ensure consistent access to meaningful therapeutic activities, actively involve families in care planning and decision making, and create supportive environments tailored to the sensory and emotional needs of neurodivergent individuals.
• Safety observation O/2025/059: NHS boards can improve patient safety by supporting their non-executive directors (NEDs) with responsibility for quality and safety to attend NED-specific training on quality of care and patient safety. This may include modules on compassionate leadership, the importance of psychological safety, safety science in investigations and techniques for supportive challenge. By fostering these skills, NEDs can better understand the complexities of healthcare delivery, engage meaningfully with staff, and ensure that patient safety and quality care remain at the forefront of their governance role.
• Safety observation O/2025/060: Integrated care boards and organisations that provide mental health care can improve safety by involving people with lived experience and family carers in coaching for executive leaders. This could include creating learning networks within provider collaboratives. By embedding these roles, executive teams and non-executive directors would receive direct insights from those with personal experience of mental health services, helping them to co-produce learning from deaths and drive improvements in care.

Related HSSIB reports

• Learning from inpatient mental health deaths and near misses: assessment of suicide risk and safety planning (September 2024)
• Mental health inpatient settings: creating conditions for the delivery of safe and therapeutic care to adults (October 2024)
• Harm caused by mental health out of area placements (November 2024)
• Mental health inpatient settings: Supporting safe care during transition from inpatient children and young people’s mental health services to adult mental health services (December 2024)