Mental health crisis care: care of patients in emergency departments

People’s experience in the emergency departments

• Prolonged waits, crowding, overstimulation, lack of privacy, limited observation and unmet basic needs could increase distress, deterioration and risk to people in crisis, other people using EDs, visitors and staff.
• Safety risks were not experienced equally. Protected characteristics, communication and sensory needs, neurodevelopmental needs, trauma, previous experiences and wider circumstances could affect how people experienced the ED environment, how their distress and needs were understood, and how able they were to participate in care and decision making.
• Children and young people could experience prolonged waits in EDs or paediatric settings while assessments continued, when responsibility for care was unresolved, or while suitable onward provision was sought.
• EDs were often not designed or resourced to provide prolonged therapeutic mental health crisis care. Dedicated mental health spaces varied in availability, design and suitability, and there were no nationally defined service standards or built environment standards. Some mental health assessment rooms did not meet Royal College of Psychiatrists’ Psychiatric Liaison Accreditation Network standards.
• People with complex social circumstances, housing instability or co-occurring substance use could fall between services, leaving EDs managing needs beyond their intended role.
• Variation in the capacity, accessibility, integration and responsiveness of mental health crisis services meant that they could not always provide a timely or suitable alternative to ED attendance for patients in mental health crisis.

Workforce – knowledge, skills and resources

• Workforce skills, confidence and capacity were not always matched to the needs of people coming to an ED in mental health crisis.
• Staff described gaps in specialist skills, legal knowledge, trauma-informed care, therapeutic engagement, de-escalation and clinical supervision.
• Workforce pressures also affected staff wellbeing, contributing to burnout, compassion fatigue, moral distress and potential moral injury.
• Enhanced therapeutic observation and care was variable, with some services relying on agency or security staff for one-to-one supervision.
• Liaison psychiatry provision was variable, with differences in staffing resources, commissioning and delivery of Core 24 standards limiting timely mental health input in EDs.
• Limited availability of approved mental health professionals (AMHPs) and Section 12 doctors, particularly out of hours, delayed Mental Health Act assessments and exposed gaps in national oversight of statutory workforce capacity and resilience.

Oversight and decision making

• Families and carers were not consistently involved in a person’s assessment, safety planning or discharge decisions in EDs, and information shared by family members was not always used to support safety.
• Many people already known to mental health services did not appear to have co-produced crisis or relapse prevention plans.
• Arrangements in the ED for urgent mental health admission for patients in mental health crisis, under Section 140 of the Mental Health Act, were not always clear or effective.
• Wider healthcare pressures increased pressure on EDs, including limited mental health inpatient beds, delayed Mental Health Act assessments due to limited access to AMHPs and Section 12 approved doctors, variable liaison psychiatry provision and limited alternatives to ED attendance.
• Governance, accountability and escalation of safety concerns was variable, with evidence of staff feeling tension and fear. Escalation meetings were often experienced as shifting responsibility and risk between services.
• Delayed admission to mental health inpatient care, prolonged stays in the ED, the exclusion of families from information and decisions about their family member, and a lack of support for the person once discharged from the ED were not consistently recognised, captured or used to monitor safety incidents or to inform safer care.

Information sharing and systems

• Poor interoperability and information sharing between IT systems meant that ED and liaison psychiatry staff could not consistently access or share risk history, crisis plans, safeguarding information, recent assessments and family-held information to support safe patient care.

Interagency working

• Delays or gaps in appropriate health provision can transfer risks between organisations responsible for providing care under Section 136 pathways.
• Interfaces between services including police, ambulance, drug and alcohol, social care, voluntary sector and other services were inconsistent, increasing operational risk and leaving EDs managing needs beyond their intended role.
• Section 136 pathways were not always supported by clear ownership across police, health and local authority services, timely access to health-based alternatives, shared risk management, effective transfer of care or sustained multi-agency learning. This is reported to have become more challenging following Right Care, Right Person implementation.

HSSIB makes the following safety recommendations

HSSIB makes the following safety observations

HSSIB suggests safety learning for integrated care boards

Local-level learning

HSSIB investigations include local-level learning where this may help organisations and staff identify and respond to specific patient safety concerns. HSSIB has identified local-level learning for acute providers and psychiatric liaison teams involved in urgent and emergency mental health crisis care. The following prompts may also support local review, improvement and partnership working.

In October 2025 HSSIB launched two investigations into the safety issues for people experiencing a mental health crisis who come into contact with urgent and emergency care services. This report focuses on patient safety concerns for people in mental health crisis who may go to and remain in an emergency department (ED) for an extended period.

1.1.1 Mental health crisis describes when a person feels at breaking point and is experiencing a mental health emergency requiring urgent help. Someone might have a known mental health condition and experience a mental health crisis, or it can be something they have never experienced before. It is usually very distressing for the person experiencing the crisis and those around them. When people are in mental health crisis, they need timely access to support that is compassionate and meets their needs.

1.1.2 Most people experiencing a mental health crisis who go to an ED will be discharged and/or signposted to other services in the community. For some people, however, admission to hospital on a voluntary or compulsory basis is needed.

1.1.3 Patients in mental health crisis who go to an ED can be recommended for admission to a mental health hospital under the Mental Health Act 1983. Section 135 of the Mental Health Act provides a legal framework for the police to intervene when there is concern for an individual’s mental health. It allows police and mental health professionals to enter a person’s home to assess their mental health and, if necessary, remove them to a place of safety. In addition, Section 136 of the Mental Health Act allows a police officer to remove someone (adult or child) from a public place to a place of safety if they appear to be suffering from a ‘mental disorder’ and are in immediate need of care or control. The ‘Mental Health Act 1983: Code of practice’ (Department of Health, 2015 – referred to in this report as the Code of Practice) states that the preferred option is a health-based place of safety where mental health services are provided.

1.1.4 The ED is a ‘place of safety’ as defined by the Mental Health Act within the context of a Section 136/135. In real terms, however, it is only a place of safety if the patient can be kept safe if, and/or when, the police leave.

1.1.5 In the year 1 April 2024 – 31 March 2025, there were 28,450 detentions under section 136 from police forces in England, a 4% decrease on the previous year (Home Office, 2025). Where the place of safety was known, (26,275 Section 136 detentions) Home Office analysis for England and Wales indicates that the proportion of individuals taken to ED as a place of safety has been increasing in recent years while the proportion of detainees taken to a health based places of safety has been declining (Home Office, 2025).

1.1.6 In 2024, there was a change made to the data collection. Previously, if the individual was taken to ED for treatment of physical injuries, police forces were asked to record the place of safety they were taken to after ED. From March 2024, this was amended to recording ED as the place of safety, for example if a person has self-harmed but also needs a mental health assessment. Even accounting for the change to the data collection, the proportion of individuals taken to ED in England continued to increase from 48% in the year ending March 2024 up to 53% in the year ending March 2025. Of the 26,275 detentions under section 136 in England where the place of safety was known, 13,903 (53%) were taken to ED and 11,992 (46%) were taken to health based places of safety. Of the remainder, for 332 (1%) a police station was used as the place of safety and for 48 detainees (>1%) the place of safety was a private home or another setting (Home Office, 2025).

1.2.1 NHS England’s ‘Crisis and acute mental health services policy’ describes a comprehensive pathway across community, ambulance, ED and inpatient settings (NHS England, n.d.a), while the ‘Urgent and emergency care plan 2025/26’ identifies reducing stays of more than 24 hours in an ED for people awaiting mental health admission as a national priority (NHS England, 2025a). However, national data indicates that EDs are managing a substantial and sustained volume of mental health-related presentations (NHS England, 2026b). In 2025/26, there were at least 504,180 mental health-related ED attendances, including a significant number related to self-harm. This demand also generates a high volume of onward referrals for specialist assessment, with over 347,000 referrals to psychiatric liaison services and around 10,500 referrals to paediatric liaison teams.

1.2.2 The NHS England policy (NHS England, n.d.a), establishes a 24-hour, 7-day mental health crisis access and response model. People of all ages experiencing a mental health crisis, or someone concerned about them, can call NHS 111 and select the mental health option to be connected to a trained mental health professional in their local crisis service. The response available may include community crisis assessment, crisis resolution and home treatment, face-to-face support, and alternatives to the ED, such as crisis cafés, safe havens or crisis houses. These services are commissioned and delivered locally, and their configuration, capacity and accessibility vary between local systems.

1.2.3 National evidence shows that people in mental health crisis can wait for prolonged periods in the ED, particularly after a decision has been made that admission to a mental health hospital is required. Limited availability of acute mental health beds can significantly delay a person’s transfer to inpatient care (Getting It Right First Time, 2026). The Care Quality Commission has reported rising demand, inpatient occupancy above safe thresholds, and delayed admission becoming normalised (Care Quality Commission, 2026).

1.2.4 HSSIB’s (2025a) investigations into mental health inpatient settings similarly reported that insufficient locally commissioned beds and fragmented commissioning responsibilities led to prolonged ED stays, distress, suicide risk and 'the exclusion of families from planning and decision making’.

1.2.5 NHS England’s (n.d.a) model for crisis and acute mental health care is based on timely, co-ordinated responses across EDs, liaison psychiatry, community crisis services, ambulance services and wider partners, with patients having access to the right service at the right time and reduced reliance on the ED as the default entry point to care. The NHS England ‘Urgent and emergency care plan 2025/26’ identifies reducing prolonged ED stays for people awaiting mental health admission as a national priority, underlining the patient safety implications of delay and poor patient flow (NHS England, 2025a).

1.2.6 NHS England guidance sets out the core principles and components of high-performing EDs (NHS England, 2026a). Specific guidance on mental health EDs will be published to set out how different components of the wider urgent and emergency care (UEC) system should support improvements in flow.

1.2.7 In August 2026, the government announced £343 million of investment in 100 community mental health centres and 59 mental health EDs, with the first sites expected to open from autumn 2026 and further sites from March 2027 (Department of Health and Social Care, 2026a). The mental health EDs are intended to provide same-day specialist support in calmer, therapeutic settings for people in crisis who are medically fit and do not need treatment in an acute ED. Acute EDs will continue to care for people who require physical health assessment or treatment, statutory assessment under the Mental Health Act, or those who remain in ED while waiting for admission to a mental health inpatient bed.

1.2.8 The need to evaluate emerging models of mental health crisis care is reflected in local evaluations such as Barnardo’s Inner Resilience and Development (BIRD) model. An independent evaluation of BIRD described a partnership model providing timely, tailored support outside hospital for children and young people whose mental health needs did not require a specialist clinical response. Referrals were made through EDs, CAMHS crisis teams and self-referral, with support often provided in the young person’s home. The evaluation found that the service improved wellbeing and had potential to address gaps between ED and specialist mental health provision (Centre for Mental Health, 2025).

1.3.1 NHS England data from April 2025 to March 2026 reported that mental health related presentations consistently represented at least 3% of all ED attendances (NHS England, 2026b) – approximately 9,690 attendances per week – although it is acknowledged that data quality issues could mean the number of people visiting ED in crisis could be higher. Many mental health emergencies also require treatment for injury or physical illness, and people with diagnosed mental illness may visit an ED for unrelated physical health needs. Assessment can be complex because physical illness, injury, intoxication or underlying conditions may co-exist with, mimic or worsen mental health crisis, increasing the time needed for safe care.

1.3.2 People experiencing mental health problems are twice as likely as other patients to remain in the ED for more than 12 hours, with marked regional variation (Getting It Right First Time, 2026). Of the patients waiting for more than 72 hours in the ED, almost a quarter were there for mental health related issues. They included people with dementia, with serious mental illness and in mental health crisis. Other national reports have identified a rise in ED attendances for mental health crisis (Getting it Right First Time, 2021; NHS Confederation, 2024; Royal College of Nursing, 2025).

1.3.3 While an ED should not be the default setting for same-day, non-emergency mental health care, it remains an appropriate and necessary place for joint assessment, care and treatment by emergency medicine and liaison psychiatry, provided patients are discharged or transferred in a timely way. EDs are designed for rapid assessment, stabilisation and discharge or transfer; national liaison psychiatry standards expect psychosocial assessment, transfer arrangements or initiation of Mental Health Act assessment within 4 hours. Delays to onward care can leave people in mental health crisis waiting in environments not intended for prolonged care (NHS England, the National Collaborating Centre for Mental Health and NICE, 2016; Getting It Right First Time, 2026). Performance varies widely across England, and many services face pressures that affect their ability to deliver consistently high-quality, responsive urgent and emergency mental health care (Getting It Right First Time, 2026).

1.3.4 For children and young people, the Royal College of Paediatrics and Child Health’s ‘Facing the Future’ standards set out expectations for the delivery and assurance of safe, high-quality and integrated emergency care, including standards relating to mental health, the care environment, workforce, safeguarding, complex needs and health inequalities (Royal College of Paediatrics and Child Health, 2025).

1.3.5 HSSIB’s (2026a) investigation into mental health crisis care: legislative challenges in emergency departments reported that there is an absence of clear legal powers to lawfully prevent vulnerable individuals from leaving the ED while awaiting assessment or admission. NHS and local authority services are not configured or resourced to be able to consistently arrange emergency Mental Health Act assessment and admission within timescales that respect patients' right to liberty. This legal ambiguity exposes patients to increased risk of harm and/or being unlawfully deprived of their liberty, and places staff in a position of uncertainty when attempting to manage safety. It also reported that for those requiring formal admission to a mental health hospital, an application under the Mental Health Act 1983 cannot be completed until a bed has been identified.

1.3.6 When admission to a mental health inpatient bed is recommended, bed shortages can result in long waits in the ED. Section 140 of the Mental Health Act requires every integrated care board and local health authority board to identify hospitals able to provide urgent admission for people detained under the Act. In practice, beds are not always available for people in cases of ‘special urgency’ or for people under 18 when needed.

1.3.7 For the period January 2024 to March 2026, the Courts and Tribunals Judiciary published at least 27 prevention of future deaths (PFD) reports highlighting coroners’ concerns about the care and treatment of people experiencing a mental health crisis who had visited EDs in England.

1.4.1 National frameworks, including professional guidance on confidentiality and data protection, set standards for lawful and proportionate information sharing. Guidance emphasises the need to follow General Data Protection Regulation principles while meeting safeguarding responsibilities (General Medical Council, 2017; Royal College of Nursing, n.d.; Royal College of Psychiatrists, 2017; Nursing and Midwifery Council, 2018a). National policy also assumes effective transfer of patient information across organisational and digital boundaries.

1.4.2 For children and young people, information sharing must also reflect safeguarding duties under the Children Acts 1989 and 2004 and the statutory guidance ‘Working Together to Safeguard Children’. Relevant information should be shared in a timely, lawful and proportionate way where needed to identify, assess or respond to concerns about a child’s safety or welfare. Data protection law does not prevent appropriate information sharing for safeguarding purposes (Children Act 1989; Children Act 2004; Department for Education, 2026).

1.4.3 The Department of Health and Social Care’s consensus statement on information sharing and suicide prevention emphasises that concerns about confidentiality should not prevent clinicians from sharing relevant information when there is a risk to life (Department of Health and Social Care, 2021).

1.4.4 National Institute for Health and Care Excellence (NICE) guidance on self-harm and adult mental health service user experience supports involving families and carers where appropriate, including in relation to consent, confidentiality, assessment, crisis care, care planning, discharge and aftercare (National Institute for Health and Care Excellence, 2011; 2022).

1.4.5 The Triangle of Care promotes partnership between the person, carers and professionals, including early identification of carers, carer-aware staff and clear protocols for confidentiality and information sharing (Carers Trust, 2013).

1.5.1 There is no universally accepted definition of liaison psychiatry (Welsh and Lee, 2025). However, the Royal College of Psychiatrists told the investigation that ‘Liaison psychiatry deals with the interface between physical and mental health, providing psychiatric treatment to patients attending general hospitals, whether they attend out-patient clinics, EDs, or are admitted to in-patient wards. This includes responsibility for mental health crisis care in the emergency department. Liaison psychiatry is a GMC recognised subspecialty of general adult and old age psychiatry’. Competencies for psychiatric nurses and liaison psychiatrists have been defined (London Liaison Mental Health Nurses’ Special Interest Group, 2014; Royal College of Psychiatrists, 2022).

1.5.2 In England, Core 24 describes a national model for adult and older adult liaison psychiatry in acute hospitals. It sets out the expectation that a multidisciplinary mental health service should be available in hospitals with an ED, 24 hours a day, 7 days a week, to provide timely assessment, treatment and advice for people whose mental health needs arise alongside physical healthcare. NHS England’s ‘Five Year Forward View for Mental Health’ expected all acute hospitals with an ED to have 24/7 liaison psychiatry provision for adults and older adults, and at least 50% to meet the Core 24 model by 2020/21, supported by national transformation funding (NHS England, 2016).

1.5.3 This commitment was reinforced in the NHS Long Term Plan, which set the ambition that 70% of relevant hospitals would achieve Core 24 or an agreed equivalent by 2023/24 (NHS England, 2019; Royal College of Psychiatrists, 2019a; 2019b). Core 24 was intended as a minimum service model, with some hospitals, particularly those with larger bed bases, greater complexity or specialist functions, expected to require enhanced or comprehensive liaison psychiatry provision.

1.5.4 For children and young people under 18, NHS England’s national implementation guidance sets out four core crisis functions: access through the NHS 111 mental health option, crisis biopsychosocial assessment, brief response and intensive home treatment. The assessment and brief-response functions should be available 24 hours a day, 7 days a week through ED, community or combined local arrangements; the guidance does not prescribe a dedicated paediatric liaison psychiatry team in every acute hospital (NHS England, 2024a).

1.5.5 National guidance supports liaison psychiatry for patients of all ages, including children, as an integral part of urgent and emergency care in acute hospitals (NHS England, the National Collaborating Centre for Mental Health and NICE, 2016; Royal College of Psychiatrists et al, 2020). The Royal College of Psychiatrists’ Psychiatric Liaison Accreditation Network standards set expectations for timely assessment, co-ordination and continuity between EDs, liaison psychiatry and onward services (Royal College of Psychiatrists, 2026).

1.5.6 Getting It Right First Time (GIRFT) (2026) clinical operational standards emphasise joint working between emergency medicine and liaison psychiatry, including parallel physical and mental health assessment where this is needed to avoid delay and support timely, safe decision making. This is consistent with joint guidance for acute hospital care, which supports parallel physical and mental health assessment to avoid delay and ensure safe, timely decision making (Royal College of Psychiatrists, Royal College of Emergency Medicine, Royal College of Physicians and Royal College of Nursing, 2020).

1.6.1 NHS England and the Department of Health and Social Care are developing a Modern Service Framework (MSF) on severe mental illness as part of the 10 Year Health Plan for England (NHS England, 2019; Department of Health and Social Care, 2025; Department of Health and Social Care and NHS England, 2026b). It is intended to set long-term outcomes and define what evidence-based, high-quality and equitable care should look like, aligned to the plan’s ‘three shifts’ from sickness to prevention, hospital to community, and analogue to digital.

1.6.2 The MSF is intended to identify outcome-led interventions or areas of activity with greatest potential to improve care where transformation and innovation are most needed. In May 2026, the Department of Health and Social Care and NHS England launched a national call for proposals on interventions and areas of action to support progress towards the current working “moonshot” outcome goal, developed in close collaboration with stakeholders:

1.6.3 NHS England has indicated a particular interest in interventions that span the health and care system, including integrated cross-sector models of care, workforce approaches and digital innovation to improve people’s experience of services, including mental health crisis care. The call for proposals also sought interventions likely to address inequalities and avoidable harms associated with current service delivery.

This section sets out findings from the investigation’s analysis of mental health crisis care in emergency departments (EDs). It draws on evidence from the vignettes in section 2, people with lived experience, families, staff and site visits, and is organised around the following themes:

• People attending emergency departments
• Workforce – knowledge skills and resources
• Oversight and decision making
• Information sharing and systems
• Interagency working and Section 136 pathways.

This section considers safety risks associated with the ED environment and how these may be affected by people’s individual needs, circumstances and characteristics. The investigation identified 10 prevention of future deaths reports, published between January 2024 and March 2026, that raised concerns about ED suitability for people in mental health crisis, including incidents involving worsening agitation and deaths in the ED or shortly after leaving (Courts and Tribunals Judiciary, 2024a; 2024b; 2024f; 2024g; 2025a; 2025c; 2025d; 2025e; 2025f; 2025g; 2026d).

Suitability for people in mental health crisis

3.1.1 The investigation heard about, and observed, patients becoming increasingly agitated in the ED. Some said they needed to pace to regulate distress but were confined to rooms with no natural light and little space to move. Staff described EDs as “noisy, bright, and overcrowded”, with limited calm, privacy or therapeutic engagement. Families and staff described these conditions as intensifying people’s distress during prolonged waits and increasing the likelihood of people trying to leave. National reports have raised similar concerns (Darzi, 2024).

3.1.2 The investigation heard and observed that many patients experienced long waits for a mental health assessment, and many people experiencing a mental health crisis could spend hours in corridors or side rooms with limited therapeutic input or observation. During one site visit, there were 13 people waiting for a mental health assessment in the ED. One parent said that after days in the department their relative was simply “left in a corridor”.

3.1.3 Staff told the investigation that prolonged waits affected basic needs, including access to hygiene facilities, toilets, privacy and a suitable place to sleep. Staff and families associated these conditions with anxiety, frustration, deterioration, aggression, violence and people leaving before being assessed. The investigation’s interim report, ‘Mental health crisis care: legislative challenges in emergency departments’, (Health Services Safety Investigations Body, 2026a) discussed the legal, policy and safety gap when restrictions are needed to prevent immediate harm while a person awaits admission to a mental health inpatient setting, treatment or transfer.

3.1.4 People with lived experience and families described EDs and general hospital wards as unsuitable for prolonged mental health crisis care. They described overstimulating environments, inconsistent observation, unsafe care and, at times, reliance on police for containment and safety.

3.1.5 The investigation was also told by the Faculty of Old Age Psychiatry during report consultation that older adults have distinct needs in EDs, particularly where mental health crisis is associated with dementia, delirium, frailty, cognitive impairment, severe depression, suicidality or co-existing physical illness. These factors can make assessment, communication and risk recognition more complex, and increase the importance of environments that reduce distress, confusion and avoidable deterioration.

3.1.6 The investigation observed older people being cared for in corridors in EDs in confused states and staff were seen trying to encourage them to stay on their trolley when trying to wander off. It was unclear whether their confusion was due to mental health crisis or a physical health issue. Research across three English EDs found that overcrowding, noise, lack of privacy and unsuitable assessment spaces can make it harder for staff to distinguish between physical and mental health needs, increasing the risk of premature diagnostic closure or missed physical illness in people with mental health conditions (Liberati et al, 2026; Jayaprakash et al, 2026). In addition, the investigation was told by the Faculty of Old Age Psychiatry, that in older people, presentation with an apparent mental health crisis should never result in physical illness being assumed absent or secondary. Older adults presenting with agitation, paranoia, psychosis, confusion, behavioural disturbance or suicidal thinking require comprehensive physical assessment before symptoms are attributed solely to mental illness or dementia.

Distressed behaviours and violence and aggression

3.1.7 The Health and Safety Executive (n.d.) defines work-related violence in healthcare as ‘any incident in which a person is abused, threatened or assaulted in circumstances related to their work’. NHS Resolution (2024) has highlighted both the financial cost of negligence claims linked to workplace violence, and the physical and psychological harm experienced by staff.

3.1.8 Staff working in EDs told the investigation that episodes of violence and aggression linked to mental health crisis presentations had caused post-traumatic stress symptoms, physical injury and financial loss when injuries prevented them from working. Some said sickness absence processes were triggered after repeated absences caused by assault at work.

3.1.9 National Institute for Health and Care Excellence (2015) guidance on short-term management of violence and aggression applies to EDs and emphasises prevention, de-escalation, safe restrictive interventions and safeguarding. The investigation found these expectations were difficult to implement consistently during prolonged waits. Staff described limited mental health-specific training, insufficient appropriately trained staff, reliance on security colleagues, and difficulty providing sustained observation and therapeutic engagement in crowded environments. This was consistent with NHS Resolution’s (2024) thematic review of workplace violence claims.

3.1.10 During site visits, the investigation heard that incidents of violence and aggression had increased. Staff reported exposure to physical and verbal abuse, and the investigation observed verbal abuse in EDs. On one visit, the investigation witnessed racist and sexualised comments from a very unwell patient in crisis. Staff and security remained calm, prioritised de-escalation and tried to expedite the patient’s assessment for admission; no restraint was used. The patient remained in the department for many hours, which appeared to intensify their distress.

3.1.11 The investigation was told about serious incidents in which other patients had been harmed in EDs and vital hospital equipment damaged. Risk registers provided to the investigation included examples of harm to staff, patients and visitors linked to insufficient appropriately trained staff to safely support people in mental health crisis.

Children and young people

3.1.12 Several organisations told the investigation that children and young people in mental health crisis remained in EDs for prolonged periods while waiting for an appropriate placement or decision about next steps. One organisation described a young person waiting in an ED for 6 weeks; another described a child remaining in an ED for more than a week ‘while partner organisations disagreed about the appropriate course of action’. Delays arose while assessments continued, admission decisions were disputed, or when specialist inpatient mental health provision for children and young people, known as Tier 4 provision, or an alternative placement could not be secured. National data indicates that bed occupancy rates for children and young people’s inpatient mental health services are relatively low compared with other inpatient services, at an average of 63% in 2025 (Centre for Mental Health, 2025a). This suggests that bed occupancy is not the main barrier to admission.

3.1.13 Staff across several organisations told the investigation that these prolonged stays were unsafe because EDs were not designed, staffed or commissioned locally to provide sustained therapeutic care for children and young people with complex mental health needs, including those liable to be detained under the Mental Health Act. The investigation heard that this left EDs managing risks they were not intended to hold, while children and young people experienced distress, restricted therapeutic support, and loss of peer contact, education and other factors that may support de-escalation.

3.1.14 Organisations described contributory factors for children and young people remaining in the ED. These included gaps in community crisis provision, breakdown of social care or community placements, delayed clinical escalation before hospital presentation, and limited flexibility or availability within Tier 4 pathways.

3.1.15 The investigation also heard about disagreement between approved mental health professionals (AMHPs), child and adolescent mental health service (CAMHS) teams, bed management teams, commissioners, receiving units and social care about responsibility for decision making and care provision. Uncertainty about legal and professional responsibilities once Mental Health Act criteria had been met could lead to repeated professional discussions without timely resolution, leaving children and young people in the ED and creating ongoing safety risks.

Neurodivergence needs

3.1.16 The investigation found that standard ED environments did not consistently meet the needs of people with dementia, a learning disability, autism and other forms of neurodivergence. The investigation reviewed prevention of future deaths reports which described that sensory overload, difficulty processing verbal information, the need for consistency and reliance on trusted relationships were not always considered in ED care (Courts and Tribunals Judiciary, 2024a; 2024b). Staff described sensory overload, uncertainty and limited reasonable adjustment as factors that could increase distress for children and young people, prolong assessment and contribute to repeat ED attendance. National guidance states that assessment and aftercare after self-harm should take account of neurodevelopmental conditions, communication needs and the wider context of the person’s presentation (National Institute for Health and Care Excellence, 2022).

3.1.17 Families described poor communication, inconsistent compassion and limited trauma-informed or autism-informed practice. The investigation found that safety can be affected when neurodivergent people’s communication, sensory distress or behaviour does not align with how services expect distress or risk to present. Families described behaviours linked to autism, overwhelm or distress appearing to be interpreted by some staff as wilful or manipulative. A subject matter advisor told the investigation that recognising these differences and making appropriate adjustments is important to accurate assessment and safe care.

3.1.18 Families described reasonable adjustments, hospital passports and crisis plans being unavailable, ignored or overridden when attending EDs. They also described distress, self-harm or communication differences being attributed solely to autism or personality disorder, rather than considering how neurodivergence, trauma and mental ill health might interact. One parent described autism becoming a “get out of jail card” for services, while another said “staff say all the wrong things” despite written guidance being available.

People with complex social circumstances

3.1.19 People going to EDs in mental health crisis were often affected by deprivation and social inclusion, including homelessness, housing instability, migration-related vulnerability and co-occurring substance use. Staff and site observations indicated that greater social and clinical complexity could be associated with longer waits and reduced continuity.

3.1.20 The investigation heard that housing and social care challenges prolonged people’s ED stays and left staff holding significant moral and practical risk. Some people remained in the ED because there was no discharge address, placement or community support, echoing delays reported in previous HSSIB investigations into mental health inpatient settings (Health Services Safety Investigations Body, 2025a).

3.1.21 During one site visit, the investigation heard about a patient who had spent 5 weeks in the ED across two attendances. During the second attendance they had been in ED for 13 days, had no discharge address and no clear ownership from social care or mental health services beyond daily meeting discussions. Delays in identifying a suitable social care placement meant community services could not be engaged. The person repeatedly said, “I will kill myself if you discharge me” and asked to stay in the ED. Staff said the person’s social challenges were serious and distressing and that the ED is not designed to manage this type of complex mental health crisis.

3.1.22 Staff who had been involved in the care of patients who died immediately after being discharged from the ED described significant shame and guilt, reflecting the moral and emotional impact of managing risk in systems where safe options were limited.

3.1.23 For patients in the ED in mental health crisis, limited access to drug and alcohol services was described as a risk in the patient pathway. ED and psychiatric liaison staff highlighted difficulties with referral routes, restrictive exclusion criteria and reduced out-of-hours provision. As a result, people with co-occurring mental health and substance use needs could fall between services, particularly where mental health services were reluctant to assess or engage when drug or alcohol use was involved.

3.1.24 Prevention of future deaths reports also raised issues about alcohol misuse not being considered as a sign of mental health deterioration, the need for more active treatment and care co-ordination for people with co-occurring mental health and drug or alcohol misuse, lack of training in the dual diagnosis pathway and onward referral, and inadequate procedures for sharing information between mental health and drug and alcohol services (Courts and Tribunals Judiciary, 2024c; 2026b; 2026e).

3.1.25 These findings are supported by research on co-occurring substance use and mental health needs, which describes fragmented care, stigma, exclusion from mental health treatment while using alcohol, and people being passed between services rather than receiving joined-up support. Research on a rapid-access service for people going to an ED with self-harm and excessive alcohol use also suggests that timely, personalised and relational support can help address a care gap for people whose needs do not fit neatly within existing pathways (Jackson et al, 2025; Wigham et al, 2025).

3.1.26 Staff described some people reaching the ED only when their crisis had become severe. They linked this to a range of factors, including stigma, lack of trust, previous experiences of services, barriers to access, unmet social needs and wider disadvantage. ED attendance was therefore often seen as reflecting accumulated unmet need rather than a single, discrete crisis. This is consistent with national evidence showing that people living in the most deprived circumstances, and groups who are often excluded from routine healthcare, experience poorer access to planned care and greater reliance on urgent and emergency services (NHS England, n.d.b; 2023a; Royal College of Emergency Medicine, 2020; Royal College of Psychiatrists, 2016; Welsh et al, 2025).

3.1.27 The investigation identified geographical variation in crisis assessment and onward support from the ED. Access to crisis hubs, liaison psychiatry, transport and approved mental health professional cover differed across areas, with additional challenges in rural and geographically dispersed areas. Published reports note wide variation in crisis care, with inconsistent implementation of evidence-based approaches and use of models without a clear evidence base (Rojas-Garcia et al, 2023; Centre for Mental Health, 2025b). In practice, timely assessment and support could depend on where and when a person made contact with services, and on whether agencies were available to respond.

Equality, discrimination and protected characteristics

3.1.28 Staff working with people in mental health crisis in the ED described language barriers, inconsistent access to interpretation services, and discriminatory responses. These barriers reduced trust in healthcare staff, affected people's willingness to seek help and influenced the quality of assessments. During one site visit, the investigation was able to speak to a patient who did not speak English in their own language, and established that they needed the toilet. The patient also explained that they did not know why they were in the ED, did not know what was happening and felt frightened. National evidence shows that some ethnic minority groups are more likely to enter mental health care through emergency or compulsory pathways and may experience poorer outcomes once in contact with services (Barnett et al, 2019; Care Quality Commission, 2024; UK Parliament, 2022a). The investigation was told by the Royal College of Psychiatrists that ‘where a mental disorder amounts to a disability, failure to provide appropriate care, reasonable adjustments, a safe environment or respectful treatment may amount to disability discrimination’.

3.1.29 The investigation further found that inconsistent recording of protected characteristics limited the ability of local services to identify inequity, monitor access and target improvement (Care Quality Commission, 2024; NHS England, n.d.b; Race Equality Foundation, 2019).

3.1.30 The Patient and Carer Race Equality Framework (PCREF) provides an existing national framework for supporting mental health services to identify and respond to racial inequalities in access, experience and outcomes (NHS England, 2023b). It emphasises partnership with local communities to co-produce services that better meet local needs. The investigation found greater familiarity with PCREF among mental health staff than among acute ED staff. Given that people experiencing mental health crisis may move across mental health, acute, emergency and wider system services, this difference in familiarity may be relevant to how race-equity learning and accountability operate across organisational interfaces. ED staff welcomed having sight of it during site visits. They fed back that this would help inform future planning.

3.1.31 The investigation also identified risks relating to cultural safety, stigma and discrimination. During one site visit, the investigation observed a patient in mental health crisis being filmed on a mobile phone by another person in the ED. Staff said people in mental health crisis were sometimes “mocked” or “laughed at” by other patients and members of the public.

Dedicated mental health spaces

3.1.32 The investigation found significant variation in the availability, design and use of dedicated spaces for people visiting EDs in mental health crisis. Most sites visited had only one or two designated rooms, commonly described as ‘ligature-light’ rather than ligature-free, and these were sometimes unavailable because they were used for other patients when demand for services was high.

3.1.33 Staff described a trade-off between operational demand and providing an appropriate setting for assessment and de-escalation. When dedicated rooms were occupied, people in acute distress were often cared for in areas not designed for therapeutic engagement or observation. This was consistent with evidence from families, prevention of future deaths reports and national reports describing ED environments as unsuitable for people in mental health crisis (Romeu et al, 2024; Royal College of Emergency Medicine, 2021).

3.1.34 The investigation’s site visits looked at the location of specialist rooms, exits, oxygen and electrical points, lines of sight, and observation when specialist areas were full. At some sites, designated rooms were close to exits or difficult to observe, increasing the risk of patients leaving before they had been assessed or treated. At the sites visited, most mental health assessment rooms, or equivalent spaces, did not meet the Psychiatric Liaison Accreditation Network (PLAN) standards (Royal College of Psychiatrists, 2026). This was particularly evident when there was more than one patient at a time with significant emergency mental health needs requiring assessment.

3.1.35 Staff described attempts to improve mental health spaces being delayed by funding limitations, competing estate priorities and older ED layouts. Whilst PLAN is a voluntary accreditation framework and recognised source of quality standards for liaison psychiatry services, the current PLAN standards describe the failure to meet a Type 1 standard as "a significant threat to patient safety, rights, or dignity and/or would break the law".

3.1.36 Staff also described burnout, moral distress (see 3.2.31 & 3.2.32) and unsafe working conditions when caring for people in crisis in environments that were not designed for their needs. National evidence supports these findings, emphasising that environmental safety must be considered alongside privacy, dignity, observation, therapeutic recovery and the availability of timely alternatives elsewhere in the crisis pathway (Office for National Statistics, 2023; Royal College of Emergency Medicine, n.d.).

3.1.37 Emerging mental health ED models are being developed in response to long waits and unsuitable ED environments, but recent national sources highlight the need for clear service definitions, close links with physical health EDs, and environments designed to support dignity, safety, therapeutic care and access to physical health assessment (Centre for Mental Health, 2026; NHS England, 2026a; Royal College of Psychiatrists, 2025).

3.1.38 NHS estates staff told the investigation that there are currently no health building note standards for mental health EDs. Health building notes provide best-practice guidance on the design and planning of healthcare buildings and the adaptation or extension of existing facilities (NHS England, n.d.c). Their absence means there is no clear national design brief or service model that addresses environmental safety, ligature and observation risks, rapid access to physical healthcare, or the safe care of people with urgent and complex needs. This has also been highlighted nationally (Centre for Mental Health, 2026; Design in Mental Health Network, 2026; Getting It Right First Time, 2026).

3.1.39 National stakeholders told the investigation they welcomed the policy focus and funding for mental health EDs (1.2.7), while emphasising that acute EDs will continue to care for people in mental health crisis who need physical healthcare, statutory assessment under the Mental Health Act, or admission to a mental health inpatient bed. They highlighted the importance of evaluating existing and pilot MHED models, clarifying recurrent funding and workforce requirements, and ensuring that the development of MHEDs does not reduce scrutiny of safety, design and governance risks in acute EDs.

3.1.40 The Design in Mental Health Network (2026) has similarly highlighted that the UK has not formally defined what a mental health ED should physically provide, including what the environment should communicate to someone in acute distress, how triage, assessment and de-escalation spaces should relate to each other, the lighting strategy, access to outdoor space, or the threshold between the mental health unit and physical health ED. It warned that there is a risk of new units being built before learning is consolidated, shared and co-produced with clinical teams and people with lived experience.

3.1.41 In addition, stakeholders cautioned against the term “physical health EDs”, as it may imply that mental health emergencies sit outside emergency care and place an unrealistic burden on people to distinguish between physical and mental health needs, which are often overlapping or unclear in ED presentations (Liberati et al, 2026).

Summary

3.1.42 The ED environment itself was a safety risk for people in mental health crisis. Busy, noisy, crowded and overstimulating spaces, together with limited privacy, poor observation, unsuitable rooms and long waits, could increase distress, deterioration, violence and aggression, and the likelihood of people leaving before care was complete. These risks were heightened where environments did not meet sensory, communication, neurodevelopmental, trauma-related or equality needs, and where dedicated mental health spaces were unavailable, unsuitable or inconsistently designed.

3.1.43 The investigation found that the same ED environment may affect people differently according to their sensory, communication, physical, neurodevelopmental, trauma-related and other needs. An environment that is accessible in principle may therefore not provide an equivalent experience of safety for everyone. Designing environments with a range of needs in mind may reduce avoidable distress and complement the reasonable adjustments required by individual patients.

HSSIB makes the following safety recommendation

This section summarises evidence about workforce conditions affecting mental health crisis care in EDs, including workforce capability and confidence, liaison psychiatry capacity, and statutory workforce constraints affecting approved mental health professionals (AMHPs) and Section 12 approved doctors. Fourteen prevention of future deaths reports published between January 2024 and March 2026 identified workforce-related themes including capability, experience, knowledge of guidance, confidence and capacity, and interagency working and communication (Courts and Tribunals Judiciary, 2024c; 2024f; 2024h; 2024i; 2024j; 2024l; 2025d; 2025f; 2025g; 2025h; 2025i; 2025j; 2025l; 2026b).

Workforce capability, confidence and gaps

3.2.1 Staff described gaps in training, specialist skills, de-escalation, trauma-informed care, risk assessment and legal frameworks that affected their ability to respond consistently where people had both mental and physical health needs.

3.2.2 The Nursing and Midwifery Council’s code and standards of proficiency set shared professional standards across nursing fields (Nursing and Midwifery Council, 2018a; 2018b). However, the investigation was told that mental health and adult nurses develop distinct, non-interchangeable specialist skills. Mental health nurses are trained in psychiatric assessment, risk formulation, therapeutic engagement and legal frameworks, while adult nurses are primarily trained in physical health assessment and clinical management. The investigation observed that ED nurses often provide the majority of continuous care during prolonged waits, including risk assessment, therapeutic engagement, escalation and observation, and shared concerns that they did not have access to sufficient mental health nursing support. Acute mental health leads and national stakeholders told the investigation that this specialist mental health nursing contribution is not sufficiently recognised in EDs.

3.2.3 In EDs, limited access to specialist mental health nursing expertise can affect recognition and management of deterioration, escalating self-harm or suicide risk, and other crisis indicators. Staff also described limited understanding of legal frameworks as a factor that could delay or undermine decisions about detention, capacity, consent and safeguarding. Similar issues were described in HSSIB’s investigation report, ‘Mental health crisis care: legislative challenges in emergency departments’.

3.2.4 Staff told the investigation that gaps in therapeutic engagement and de-escalation could contribute to distress, disengagement, behavioural escalation, restrictive interventions or unsafe outcomes for people in mental health crisis. A forum of acute mental health leads described significant variability in how mental health expertise and leadership was resourced in EDs. They said access to trained mental health professionals in EDs, supported by clinical supervision, was essential for patient and staff safety. However, they described variation in access to appropriately qualified people to undertake clinical supervision, as well as time constraints. Variation in supervision has also been raised nationally (Royal College of Nursing, 2022).

3.2.5 Patients and families told the investigation that repeatedly retelling distressing events could contribute to further distress. A subject matter advisor told the investigation that repeated disclosure, fragmented information sharing and repeated questioning could reinforce a sense that “nowhere is safe”.

3.2.6 One family’s story illustrated how an invalidating response in ED could compound earlier trauma and reduce trust in help-seeking. A mother described taking her daughter, who was 17, to the ED following a serious assault, deterioration in eating and drinking, and an overdose. During mental health triage, her daughter was told: “Being a teenager is very difficult. You’ll get over it.” The mother said further overdoses followed and that no effective help was forthcoming. A subject matter advisor said this demonstrated how minimising distress can contribute to trauma by making a place expected to provide safety feel dismissive and unable to respond to escalating risk.

3.2.7 Trauma-informed approaches are increasingly cited in policy and practice, but there remains limited consensus in health and social care about their definition, key principles, and implementation across the health and care system (Department of Health and Social Care, 2022). ED staff told the investigation they did not receive training on trauma-informed care.

3.2.8 A subject matter advisor told the investigation that trauma-informed practice requires organisations to look beyond individual interactions and consider trauma-informed approaches across the whole pathway, including minimising unnecessary retelling, improving information sharing and preserving trusted relationships that promote safety, trust and recovery.

3.2.9 Families and patients said that “continuity mattered”. Practice that felt safe included staff introducing themselves clearly, explaining what would happen next, treating people and relatives with dignity, adapting communication, using prior knowledge, and recognising family insight. One family described a staff member as bringing “humanity”, while another said the police were “calm, transparent and non-judgemental”.

3.2.10 The investigation observed occupational therapists (OTs) providing structured, recovery-focused and therapeutic interventions, with strong patient-centred practice. The investigation was told that a more consistent OT presence across community mental health teams could strengthen recovery planning, therapeutic intervention and alternatives to prolonged ED stays.

3.2.11 The investigation heard that workforce pressures affected service availability and continuity, particularly out of hours. Sites described difficulty sustaining 24/7 psychiatric liaison cover, reliance on temporary staff, and shortages in key roles including AMHPs, medical staff and OTs. This contributed to delayed assessment and fragmented care.

3.2.12 Enhanced therapeutic observation and care (ETOC) teams can support ED staff to provide safer, more person-centred care for people with high levels of distress or risk, by moving observation away from passive monitoring towards therapeutic engagement, de-escalation, dignity and least-restrictive support (NHS England, 2025e). The investigation observed variation in the use of ETOCs in the ED. Some organisations had implemented a robust ETOC model, while others relied on agency nurses and security staff to provide one-to-one supervision of people experiencing mental health crisis with variation in levels of training and oversight from qualified and/or senior nursing staff. Although the ETOC model when fully implemented was observed to benefit people in crisis, implementation was often limited by staffing pressures and recruitment challenges.

3.2.13 The investigation also heard about positive experiences, particularly of lived experience-led services and charities, which were described as compassionate, consistent, proactive, relational and validating.

3.2.14 In addition, patients and families spoke positively about third sector support where they had a named person, active follow-up and practical help during periods of crisis. This meant services stayed alongside the person and helped them hold on to hope, rather than expecting them or their family to manage alone.

Liaison psychiatry capacity and Core 24 delivery

3.2.15 The investigation observed that robust Core 24 (see 1.5) and dedicated enhanced care teams within the ED gave services greater control over clinical decision making, improved patient experience, and reduced reliance on agency and security staff. These findings align with national evidence describing workforce instability, variation in Core 24 delivery, and demand rising faster than capacity (NHS England, the National Collaborating Centre for Mental Health and the National Institute for Health and Care Excellence, 2016; Royal College of Emergency Medicine, 2021; Royal College of Nursing, 2025; Royal College of Psychiatrists, 2025a).

3.2.16 However, the investigation also observed variation in meeting Core 24 liaison psychiatry standards, including differences in how roles and responsibilities were implemented The investigation observed that some psychiatric liaison staff would look after patients while in the ED, some would help with agitation, some would help with difficult capacity issues and some had very little involvement. The investigation observed tensions between ED and liaison psychiatry staff when identifying staff to care for patients who have remained in ED beyond their initial assessment. The investigation was told by the Royal College of Psychiatrists that ‘whilst this should obviously be a core responsibility of a liaison psychiatry team, it is difficult to meet fully given that teams are resourced for initial emergency assessment and care’. National survey data also shows uneven liaison psychiatry provision across England and continuing workforce instability in acute hospitals (Royal College of Psychiatrists, 2016; 2025a). Factors included workforce shortages, funding constraints, fragmented commissioning and low staff morale.

3.2.17 According to NHS England data, between April 2025 and March 2026 there were around 347,405 referrals from EDs to psychiatric liaison teams and around 10,505 referrals from EDs to paediatric liaison teams (NHS England, 2026b). Of the 357,910 referrals to these teams, an estimated 82% received face-to-face contact and 70% of those were seen within the 1-hour access metric, although variation between providers may suggest data quality issues in relation to recording face-to-face contact.

3.2.18 An observational study of liaison psychiatry commissioning found variable and fragmented arrangements, inconsistent standards and outcome measures, limited data on impact, weak integration between mental and physical healthcare commissioning, and unclear accountability across organisational boundaries. These issues constrained strategic planning and sustainable service development (Fossey et al, 2020).

3.2.19 The latest psychiatry staffing survey reported substantial but incomplete progress (Royal College of Psychiatrists, 2026). Although most services are commissioned to meet Core 24 or equivalent standards, fewer deliver this in practice because of workforce instability and vacancies, particularly in consultant roles. The investigation understands that no further staffing surveys are planned.

3.2.20 Stakeholders said liaison psychiatry now carries greater expectations for whole-pathway working, timely assessment, risk formulation, discharge planning and support to ED staff. They said capacity modelling is needed because referral numbers alone do not capture the acuity and complexity of patient needs, repeat assessments, bed delays, or the time required for safe parallel mental and physical healthcare.

3.2.21 The investigation observed more effective Core 24 delivery where liaison psychiatry was supported by wider pathway arrangements that helped move people out of the ED once their acute physical healthcare needs had been addressed. These included “pull” mechanisms from crisis resolution and home treatment, community treatment and assertive outreach teams.

Access to AMHPs and Section 12 doctors

3.2.22 The investigation was told repeatedly that delays in accessing AMHPs and Section 12 approved doctors had a direct and compounding effect on the safety and experience of people going to ED in mental health crisis who may need to be admitted to inpatient mental health care. However, stakeholders also emphasised that the timing of Mental Health Act assessments is only one part of a wider pathway.

3.2.23 The investigation was told that assessment needs to be clinically appropriate, informed by the person’s circumstances and co-ordinated with bed availability, professional decision making and consideration of wider operational pressures. Where no inpatient bed is available, earlier assessment does not necessarily lead to earlier admission or resolution of the person’s wait in ED. The investigation also heard the term ‘therapeutic delay’, reflecting that taking time with a person in mental health crisis may sometimes be clinically appropriate. Limited availability of statutory professionals, particularly out of hours, delayed Mental Health Act assessments and contributed to prolonged stays in environments that were not suitable for therapeutic care. Getting It Right First Time Mental Health Team (GIRFT) also reported concerns about AMHP and Section 12 doctor capacity and the timeliness of Mental Health Act assessments in EDs, health-based places of safety and the community.

3.2.24 The national workforce plan for AMHPs (Department of Health and Social Care, 2019) remains the most recent national workforce plan specific to the AMHP role. This is notable because the Care Quality Commission’s (CQC’s) 2018 review identified workforce fragility, variation in service availability and inconsistent access to timely AMHP assessments, particularly out of hours. These issues contributed to delays, greater reliance on police and EDs, and increased risk for people in crisis, families and practitioners (Care Quality Commission, 2018).

3.2.25 The national service standards for AMHPs provide guidance on governance, service delivery and workforce support but do not replace a funded national workforce strategy (Department of Health and Social Care, 2020). The investigation considered that, without a refreshed workforce plan, the systemic risks identified by the CQC (2018) may persist where local services struggle to sustain safe and resilient 24/7 AMHP cover.

3.2.26 Section 12 approved doctors play a critical statutory role in Mental Health Act assessments, but national oversight focuses mainly on individual approval, training and re-approval arrangements set out in the Mental Health Act approval instructions (Department of Health, 2015; Department of Health and Social Care, 2013). Unlike AMHP services, there is no equivalent national workforce plan or service standard that sets expectations for availability, out-of-hours cover or service resilience. This creates a gap between regulation of individual competence and oversight of system capacity.

3.2.27 National evidence, including CQC’s annual ‘Monitoring the Mental Health Act’ reports and research by King’s College London, continues to identify delays, variation in access and prolonged waits in inappropriate settings associated with difficulty securing timely Section 12 medical input (Care Quality Commission, 2026; Stevens et al, 2022). In view of the evidence about delays in accessing AMHPs and Section 12 approved doctors, the investigation held focus groups with AMHPs and Section 12 doctors to understand their experiences and to seek their views on mental health crisis care in EDs.

3.2.28 Participants described AMHP services as under-recognised, inconsistently supported and frequently required to absorb wider system pressures. AMHPs and Section 12 doctors told the investigation that EDs were often used because no other option was available, rather than because they were clinically appropriate. Delays in accessing mental health inpatient beds meant people remained in the ED long after a decision to admit to mental health inpatient care, increasing distress, repeat assessments and the risk of people leaving before care could be arranged. They also cautioned that pressure to assess quickly, particularly out of hours, could risk Mental Health Act assessments being undertaken with incomplete information or before there was a realistic route to admission to a mental health inpatient bed.

3.2.29 Staff described delays in accessing AMHPs and Section 12 approved doctors as being associated with deterioration in patients’ mental state, distress, agitation and greater use of restrictive interventions. They also described prolonged ED stays increasing reliance on police involvement and informal containment, with risks of de facto detention without timely statutory safeguards. The investigation found that these delays created a recurring patient safety risk across the wider system, consistent with recent ‘Monitoring the Mental Health Act’ reports (Care Quality Commission, 2026).

3.2.30 The investigation also heard that concerns about payment arrangements for Section 12 doctors had affected local assessment arrangements in some areas. The investigation was told this reduced availability of Section 12 approved doctors and placed further pressure on fragile out-of-hours service provision. The investigation was also told of significant variation in payment schedules for Section 12 work in different regions with variation in ICB oversight.

Staff wellbeing

3.2.31 Staff working in EDs and ED-based mental health crisis care frequently told the investigation that they felt overwhelmed, frightened and burnt out. They described the work as intense and demanding, particularly when repeatedly exposed to distress, risk, trauma and suffering without adequate time, resources or specialist support. This is consistent with evidence on compassion fatigue in emergency medicine, which links repeated exposure to traumatic events, high cognitive and emotional demands, workplace stress and limited opportunity to recover with emotional exhaustion, reduced empathy and difficulty sustaining compassionate care (Cocker and Joss, 2016; Jeanmonod et al, 2024).

3.2.32 Staff also described experiences consistent with moral distress and potential moral injury. Moral distress is psychological unease when professionals know, or feel, what ethically appropriate care should be provided but are constrained by institutional, resource or system factors. Moral injury can arise where sustained moral distress leads to longer-term psychological harm or impaired functioning (British Medical Association, 2026; UK Clinical Ethics Network, 2026; Williamson et al, 2021). Staff said they had to contain people at high risk in unsuitable ED environments, sometimes without timely therapeutic intervention, specialist mental health input or onward care. This is consistent with previous HSSIB reports that describe staff fatigue as a potential patient safety risk (Health Services Safety Investigation Body, 2024a).

3.2.33 The investigation was also told that assessment of people experiencing mental health crisis requires staff to interpret behaviour, communication, distress, vulnerability and risk, often in pressured circumstances and sometimes with incomplete information. The investigation found that these interpretations may be affected where communication needs, neurodivergence, trauma, language or other individual circumstances are not sufficiently recognised. Accurate assessment requires time, continuity and therapeutic engagement, but staff described limited capacity to provide this consistently in busy EDs. Repeatedly managing high levels of distress and risk without adequate time, support or safe onward options could contribute to compassion fatigue and moral distress, reducing staff capacity to provide compassionate and relational care.

Summary

3.2.34 In summary, workforce capability, capacity and support across EDs, liaison psychiatry, AMHP services and Section 12 medical provision were central to mental health crisis care. Evidence from staff, focus groups and national reports indicated that gaps in specialist skills, confidence, therapeutic engagement, legal knowledge and statutory assessment contributed to delays, variation and fragmented care.

3.2.35 The evidence indicates that workforce pressures contributed to prolonged ED stays in environments not designed for therapeutic care, with associated risks to patients of distress, deterioration, restrictive intervention and legal uncertainty. Staff described effects on their own wellbeing consistent with literature on compassion fatigue, moral distress and potential moral injury.

Admission and discharge decisions

3.3.1 Family members consistently described that going to the ED was the only option available to them because other crisis pathways were fragmented, with responsibility passing between services and no single team co-ordinating support. One family summarised this as “everybody’s passing you on”. Similar issues have been highlighted in prevention of future deaths reports (Courts and Tribunals Judiciary, 2024a; 2024b; 2026c).

3.3.2 Parents of young people described transitions between ED, CAMHS and adult services, crisis services, social care, and supported accommodation as particularly unsafe. Families said they often found themselves co-ordinating care and managing serious risk with limited guidance or support.

3.3.3 Family members said decisions around admission, discharge or onward support sometimes focused narrowly on whether a person met diagnostic or service eligibility criteria, without sufficient consideration of their wider physical health, neurodevelopmental, trauma-related, emotional or social needs. Rigid criteria could leave people with complex or overlapping needs without appropriate support, and experiences of being excluded or passed between services could increase distress and reduce trust.

3.3.4 People with lived experience and family members told the investigation that questions about mental capacity could lead EDs and psychiatric liaison teams to frame discharge or non-intervention as personal choice rather than as a safety issue requiring active support. One family described hearing that “it’s her choice”, while another reflected that the person “was expected to take responsibility at the very point when they were least able to do so”. This highlights the importance of decision-specific capacity assessments, made at the time decisions are needed and after practicable support has been provided, particularly where crisis, intoxication, distress, psychosis or fluctuation may affect the person’s ability to make or communicate a decision.

3.3.5 The investigation was told that a finding of capacity does not remove safeguarding or duty-of-care responsibilities; where abuse, neglect or self-neglect may be a concern, staff should record safeguarding decisions, agree ownership of immediate risk, and assess risks to families or carers expected to provide support after discharge. Families being left to manage significant risk with limited or no support was also a theme in some prevention of future deaths reports (Courts and Tribunals Judiciary, 2024c; 2025d).

3.3.6 The investigation also heard from families and a voluntary sector organisation that people in acute mental health crisis could be expected to take responsibility for their own safety at a time when their distress, vulnerability or changing risk could limit their ability to do so. This could result in responsibility for managing serious risk being transferred to the person or their family without sufficient safety planning, follow-up or support.

3.3.7 Patients and family members said safety plans were not always considered in ED decisions about admission and/or discharge. These plans included important information about relapse patterns, autism-related communication differences and vulnerability, but families felt their content was not given sufficient weight. This issue was identified in HSSIB’s (2025a) investigation into mental health inpatient settings and safety recommendations were made. Prevention of future deaths reports published between January 2024 and March 2026, and family accounts, showed that families were not consistently involved in assessment, safety planning or decision making (Courts and Tribunals Judiciary, 2024a; 2024b; 2024c; 2025a; 2026a; 2026b).

3.3.8 In addition, confidentiality was sometimes interpreted in ways that limited staff from receiving, using or sharing information with families where this may have supported safety. This issue was identified in HSSIB’s (2026b) investigation into insulin: supporting safe self-administration in vulnerable patient groups in the community. Families told the investigation they were not seeking access to private therapeutic conversations, but wanted to share concerns, contribute to safety planning and understand enough to support their relative. Some described being unable to influence care or obtain basic information about diagnosis, medication or safety planning; one family said confidentiality was used as “a defensive tool”. Families experienced this exclusion as the loss of an essential source of safety, rather than simply a lack of information.

3.3.9 Family accounts described exclusion from decisions around admission and discharge as particularly harmful when paranoia, fear or mistrust were already present. Although national guidance supports lawful information sharing with families and carers where this supports care (see 1.4), staff told the investigation they were uncertain about what could be shared and feared “breaking the rules”.

3.3.10 Families described significant emotional, physical and practical consequences from being left to manage their relative in crisis at home with limited guidance after they had been discharged from the ED. Accounts included monitoring their family member overnight, restricting access to medication or sharp objects, repeatedly contacting services, and having to balance crisis management with employment, parenting and wider caring responsibilities. Some families described becoming fearful in their own home or feeling blamed while carrying the burden of prevention.

3.3.11 The investigation spoke to integrated care board (ICB) leaders and mental health services. Although some areas were expanding crisis response services, 24/7 alternatives to ED attendance and enhanced ED mental health provision, availability, access and integration remained inconsistent. Where alternatives were unavailable, unclear or under-resourced, the ED remained the default place for assessment, waiting and risk management.

3.3.12 Prevention of future deaths reports have raised concerns about decision making in relation to discharge, admission, onward referral or follow-up of patients where there was evidence of significant or escalating risk. Concerns include minimisation of risk, inconsistency in use of available information, adequacy or completeness of assessments or decisions not being escalated (Courts and Tribunals Judiciary, 2024c; 2024j; 2024l; 2025f; 2025g; 2025i; 2025l; 2026b).

3.3.13 Decisions to admit or discharge people in mental health crisis in the ED were influenced by the availability of mental health inpatient beds. Insufficient capacity, particularly in specialist units such as psychiatric intensive care units and beds for children and young people, led to patients at high risk of harming themselves or others experiencing prolonged stays in the ED, increasing distress and disrupting continuity of care. The investigation observed that bed pressure could contribute to discharge back to challenging home environments, changes in plans without clear explanation, siloed working and blame-based language between services. Prevention of future deaths reports have also highlighted concerns about discharge planning and lack of mental health inpatient beds (Courts and Tribunals Judiciary, 2024a; 2024b; 2024d; 2024e; 2024l; 2025b; 2025c; 2025d; 2025e; 2026d).

3.3.14 National practice standards support appropriately skilled liaison psychiatry staff contributing to decisions about admission and the use of community alternatives (Getting It Right First Time, 2026). The investigation findings indicate that unclear decision-making authority, repeated assessments and transfers of responsibility between services can contribute to delay and fragmented care.

3.3.15 NHS England’s operational pressures escalation levels (OPEL) framework sets out roles and actions to manage operational pressure, and GIRFT Mental Health Team has developed a flow-focused approach to inpatient care as part of its clinical operational standards (Getting It Right First Time, 2026; NHS England, 2025b). The investigation observed several operational bed management meetings in EDs. At these meetings, specialty teams discuss reasons why patients in mental health crisis are delayed in the ED and what is preventing admission or discharge to the appropriate setting. The intent is to minimise delays and ensure patients are moved from the ED in a timely way. The meetings observed by the investigation were multidisciplinary and included representatives from social care and mental health inpatient hospitals.

3.3.16 The investigation observed at operational meetings that escalation of safety concerns did not consistently result in timely Mental Health Act assessments, availability of inpatient beds or improved patient flow from the ED, and sometimes no person or team assumed responsibility for the patient. This was often experienced by staff “as shifting responsibility and risk between services”. During a focus group, an acute mental health lead described escalation as “purely an anxiety shift”. The investigation found that escalation did not always improve safety where responsibility was unclear and services were not working together to address the pathway problems causing the delay. Staff frequently described escalation processes as increasing the visibility of underlying capacity pressures, but not resolving them.

3.3.17 The investigation heard about and observed tensions between system partners and concerns that decisions were not always centred on the person in crisis in the ED, particularly where a person’s case was complex or described as “difficult”. Staff working with people in mental health crisis in the ED, and staff who attended focus groups, described challenging operational meetings, fragmentation between acute and mental health providers, and concern that removal of the Care Programme Approach and Assertive Outreach had reduced continuity of care. One participant said services were “fighting in a drama triangle”.

3.3.18 The investigation was told of concerns about speaking up, with leadership teams described as being “pulled into the politics”. One senior leader told the investigation:

3.3.19 ED staff told the investigation that many of the patients they saw were already known to mental health services. National data supports this view, indicating that 80% of people who go to an ED in mental health crisis are already known to mental health services (Care Quality Commission, 2024). ED staff said patients at risk of admission often did not appear to have a co-produced relapse prevention or crisis plan, a concern also reflected in family accounts.

3.3.20 The investigation also heard views that prolonged hospital stays were not always supported because patients could become “too comfortable”. A family member described hearing similar language during a prolonged ED stay, during which their relative did not receive meaningful therapeutic input.

Visibility of delays and harm in the emergency department

3.3.21 The investigation was told that governance structures intended to oversee mental health safety risks in the ED were often fragmented or underdeveloped. This reduced visibility of prolonged waits, of people leaving ED before care was complete and coming to harm, and of exposure to violence and aggression.

3.3.22 Routine national data on mental health-related presentations in EDs mainly measures process and activity. Staff and stakeholders repeatedly highlighted concerns about the limitations and data quality in the Emergency Care Data Set and Mental Health Services Data Set. Harm linked to ED delays, patient and family experience, outcomes after discharge and the use of restrictive practices within ED are not consistently captured or reported, limiting visibility of the impact on people in mental health crisis with implications for ongoing monitoring.

3.3.23 The investigation observed that delays created safety risks including deterioration in the mental state of people in crisis, increased restrictive practices, reliance on non-specialist environments to manage risk, and distress for patients, families and staff (Health Services Safety Investigation Body, 2026a). The investigation was told that some people who needed a mental health bed were admitted to acute hospital wards despite having no physical health need. Staff and families said these “hidden waits” were not routinely understood as being harm caused by delay, and that EDs and acute wards lacked the environmental safety standards and therapeutic provision of mental health wards.

3.3.24 NHS England’s Mental Health Investment Standard categories set out how mental health expenditure should be reported, including spend across children and young people’s mental health, adult mental health, crisis care and inpatient pathways (NHS England, 2025c). However, expenditure categories do not show whether local systems have sufficient capacity, timely access or safe pathways for people in mental health crisis who attend EDs. The investigation was told:

3.3.25 GIRFT Mental Health Team, working with the GIRFT ‘Further Faster’ urgent and emergency care programme, carried out visits focused on mental health-related ED flow. Reviews included dashboard peer review, assessment against clinical operational standards, ED and mental health assessment area walkarounds, staff conversations and agreed improvement actions. Feedback was consistent with the investigation’s findings.

3.3.26 GIRFT Mental Health Team told the investigation that strong relationships between providers and commissioners are critical when considering alternatives to ED attendance, crisis pathway capacity and out-of-hours provision. GIRFT Mental Health Team also highlighted that systems remain heavily reliant on EDs as the default place of safety for people in mental health crisis, especially overnight.

3.3.27 GIRFT Mental Health Team has developed a Mental Health Emergency Department Breach Analysis Tool to help systems understand 12-hour ED breaches for people in mental health crisis, identify where delays arise and support system-wide action to improve safety, patient experience and timely access to appropriate care (NHS England, 2025d).

3.3.28 The tool is intended to bring together patient-level data across ED processes, liaison psychiatry, Mental Health Act pathways, bed availability, transport, workforce capacity, clinical complexity, intoxication and medical fitness. GIRFT Mental Health Team told the investigation that this will support local health and care partners to identify drivers of delay, improve oversight and inform national policy, guidance, operational standards and benchmarking.

3.3.29 The investigation was told that GIRFT Mental Health Team’s nationally developed improvement methods and tools can support local partners to review urgent and emergency mental health crisis pathways in a structured and consistent way. Used alongside patient and family experience, these tools can help integrated care boards, NHS provider organisations and other partners understand unwarranted variation, identify the factors contributing to delay and harm, and agree shared improvement actions across the pathway. This can support more consistent oversight of patient flow, accountability across organisations and opportunities to reduce avoidable harm for people in mental health crisis who attend EDs.

3.3.30 Given this existing national improvement work, HSSIB will not make a safety recommendation in this area. However, the investigation notes the importance of using patient and family experience and reported harm as safety intelligence alongside operational data.

3.3.31 The National Confidential Inquiry into Suicide and Safety in Mental Health (NCISH) does not currently collect data on patients who died by suicide after contact with an ED. NCISH will broaden its health service contact criteria before suicide to include liaison psychiatry services in general hospitals and one-off psychiatric assessments following ED presentation, and will report these numbers in future reports (National Confidential Inquiry into Suicide and Safety in Mental Health, 2026).

Summary

3.3.32 The investigation found that the safety of patients was strengthened when ED care was timely, compassionate, joined up and informed by those who knew them best. Families and carers often held information that could support assessment, safety planning and discharge, but were not consistently involved or enabled to share this effectively.

3.3.33 Decisions about the admission, discharge and onward care of patients were affected by fragmented pathways, limited inpatient bed availability, delays in Mental Health Act assessment, poor information sharing and unclear ownership between services. Escalation processes increased visibility of risk but did not always resolve delay or improve safety. Harm linked to prolonged waits, unsafe discharge, family exclusion and repeated distress was not consistently captured or used as safety intelligence. Where appropriate data are available, examining waiting, restrictive interventions, police involvement, repeat crisis attendance, discharge and patient experience across relevant population groups may help services identify patterns that would not be visible within aggregate measures alone.

HSSIB makes the following safety recommendation

HSSIB makes the following safety observations

HSSIB suggests safety learning for integrated care boards

This section summarises evidence about safety risks associated with fragmented digital systems, poor interoperability between electronic record systems and weak information sharing. Ten prevention of future deaths reports published between January 2024 and March 2026 identified gaps in information sharing, digital systems and organisational boundaries that undermined risk evaluation and mitigation (Courts and Tribunals Judiciary, 2024a; 2024b; 2024f; 2024g; 2024k; 2025a; 2025f; 2025i; 2025k; 2026a; 2026e).

3.4.1 The investigation found that fragmented IT systems, poor interoperability and weak information sharing led to limited timely assessment, risk formulation and decision making. Staff described arrangements across ICBs, NHS England regional teams, acute and mental health trusts, police and third sector organisations as “actively discouraging good practice rather than enabling it”.

3.4.2 ED and liaison psychiatry staff repeatedly told the investigation that they could not easily access information held by community-based services or reliably share assessments with other services. This included risk history, crisis plans, safeguarding information, community contacts and recent mental health assessments. In EDs, where immediate decisions about safety, psychosocial needs, physical health and onward care are made under pressure, missing information limited staff’s understanding of the person’s current risks and needs, and decision making.

3.4.3 The investigation heard that gaps in access to relevant information meant staff were often making decisions with an incomplete understanding of a person’s history, current risks and needs. This increased uncertainty, led to repeated history-taking and duplicated work, and created a risk of inconsistent or delayed responses. A consultant psychiatrist told the investigation that “it is difficult to identify patterns of risk or safety concerns when in the emergency department, which results in missed opportunities for us to make the best decision”. These risks were described as particularly significant for people presenting with self-harm, suicidality, psychosis, severe distress or co-occurring substance use.

3.4.4 Research supports the concerns raised with the investigation. Austin et al (2024) found that improving ED care for people experiencing mental illness depends partly on better communication processes, electronic records and discharge coordination. Li et al (2022) found that electronic health record interoperability is associated with improved safety, quality and data accuracy, while poor interoperability contributes to missing information and workflow inefficiency. Other HSSIB reports have also found that poor information sharing and fragmented records can create patient safety risks (Health Services Safety Investigations Body, 2022; 2023; 2024b; 2025b; 2025c).

3.4.5 Given that HSSIB has made several safety recommendations to national organisations on digital and information-sharing risks, no further safety recommendation has been made. However, the findings may contribute to local-level learning.

3.5.1 The investigation found ambiguity around service ownership for people who go to an ED in mental health crisis. Staff and stakeholders described how overlap between crisis teams, liaison psychiatry, home treatment teams, urgent mental health helplines and police contributed to delay, fragmented care and EDs holding clinical and operational risk.

3.5.2 This was not consistent with the National Partnership Agreement: Right Care, Right Person, which states that people in mental health crisis should receive support from the service best placed to meet their needs, with police involvement limited to immediate policing requirements (Department of Health and Social Care, 2024). Where community crisis pathways, ambulance response and health-based alternatives were unavailable, an ED could become the default setting for assessment, risk management and waiting.

3.5.3 The investigation heard that implementation of Right Care, Right Person can, in some areas, shift risk into EDs when other services do not attend or accept responsibility. This was particularly evident in Section 136 pathways, where limited access to a health-based place of safety and delays in mental health assessment meant people remained in the ED for prolonged periods. National data continues to show substantial use of Section 136 powers (see 1.1.5). In addition, as described in this investigation’s interim report (Health Services Safety Investigations Body, 2026a), concerns were also raised about police refusal to be involved when a person in mental health crisis leaves the ED before being seen or admitted to a mental health hospital. The Mental Health Act Code of Practice supports the least restrictive approach, but this requires a situation-specific judgement about the person’s needs, risks, wellbeing and safety, rather than simply choosing the option involving the least intervention (Department of Health, 2015).

3.5.4 NHS England’s Right Care, Right Person implementation guidance (2024b) states that acute hospitals should have protocols, familiar to staff, setting out how risks will be managed while a person remains in hospital and what action should be taken if they leave before care is complete. This includes risk assessment and appropriate follow-up or escalation, including contacting the police where emergency action is required.

3.5.5 The NHS England High Intensity User programme (NHS England, n.d.d) is intended to support people who attend EDs frequently. The investigation was told that inadequate contingency planning for people who repeatedly visit EDs when in mental health crisis contributed to repeated ED attendance, repeated use of Section 136 and escalating risk. Staff reported that bed and staffing pressures reduced access to health-based places of safety, increasing the likelihood that people detained under Section 136 were taken to, or remained in, an ED for longer than clinically appropriate.

3.5.6 The investigation heard that tensions could arise between police and ED staff when people presented in crisis with high levels of distress, uncertainty or risk. National policy is clear that police involvement remains necessary where there is an immediate risk to life or serious harm. There is also an expectation that this is accompanied by timely healthcare handovers, shared risk management and effective multi-agency working, in line with the Mental Health Crisis Care Concordat (Department of Health, 2014) and Right Care, Right Person (Department of Health and Social Care, 2024).

3.5.7 Relevant case law highlights the importance of safe transfer from police to healthcare, including consideration of the receiving environment and sharing relevant information. The investigation found that handover was more difficult where observation capacity was limited or responsibility between services was unclear (Webley v St George’s Hospital NHS Trust and another, 2014).

3.5.8 Stakeholders identified safety concerns associated with Right Care, Right Person across mental health, ambulance and ED settings. Themes included lack of health capacity when police did not attend, absence of escalation protocols, poor communication between agencies, unclear responsibility when people left health settings, and distress, vulnerability and risk for people in crisis and families.

3.5.9 Some families said police responses felt more reliable, compassionate and accountable than responses from mental health services:

3.5.10 The investigation heard about a local pilot to strengthen support for people attending an ED under Section 136. The pilot brought together an AMHP, mental health liaison staff, adult social care and a voluntary sector practitioner during peak times. It aimed to support person-centred, strengths-based conversations, identify contributors to crisis, consider wider social factors and improve access to community-based options. It was also intended to capture learning about the patterns and drivers of Section 136 presentations, including system gaps and opportunities for earlier support.

3.5.11 Early evaluation of the pilot indicates that people presenting in mental health crisis often have interconnected health, social and practical needs, highlighting the importance of earlier, co-ordinated cross-sector support and community-based alternatives that respond to the whole person, not only the presenting issue. This is consistent with emerging research on solution-focused practice in AMHP interventions, which suggests that purposeful communication, attention to the person’s own voice, and a focus on strengths, hope and practical next steps can support more person-centred and least restrictive crisis responses (Perry, N et al, 2026). This provided a positive example of local partners using learning from ED pathway reviews to improve experience, participation and onward support.

3.5.12 In addition, the investigation went to a local multi-agency forum where anonymised incidents and pathway issues relating to people attending EDs in mental health crisis, Right Care, Right Person and Section 136 were reviewed collectively. This provided an example of a structured, psychologically safe multi-agency forum where partners could discuss difficult pathway issues, review learning collectively, strengthen relationships and identify escalation and improvement opportunities for people presenting to EDs in mental health crisis.

Summary

3.5.13 Right Care, Right Person and Section 136 pathways could increase patient safety risk where implementation was not supported by timely health-based alternatives, clear ownership, shared risk management and effective transfer of care between police, ambulance, mental health and ED services.

3.5.14 The evidence indicates that safer implementation depends on whole-pathway planning, real-time information sharing, timely handover, agreed escalation routes and sustained multi-agency relationships. Local pilots and shared review forums provided examples of how partners can use Section 136 and Right Care, Right Person pathway learning to improve experience, escalation and onward support for people in crisis.

HSSIB suggests safety learning for integrated care boards

During our series of investigations into mental health inpatient settings, we heard concerns about the care of people in mental health crisis which may benefit from a HSSIB investigation. We carried out a range of work to help further understand these concerns, including conversations with stakeholders, reviewing available data, and analysing existing literature, including coroners’ reports to prevent future deaths.

Feedback from stakeholder engagement was supportive of HSSIB work to explore mental health crisis care and suggested that it would add value in terms of a system-focused lens on the safety issues.

As a result of this work we launched two investigations that explore the patient safety issues associated with care pathways for people experiencing a mental health crisis who come into contact with urgent and emergency care services.

• Investigation 1 – Mental health crisis: care of patients in emergency departments. An interim report was published in April 2026 linked to evidence collected in response to a significant legal, policy and safety gap in the care of people who are in mental health crisis in emergency departments (EDs).
• Investigation 2 – Mental health crisis: ambulance service response via NHS 111 and 999. This investigation launched in spring 2026, following completion of substantive work on the first investigation, and is anticipated to be available in spring 2027.

Evidence gathering

The investigation engaged with multiple sites, including a range of hospital types such as district general hospitals, foundation trusts, and university and teaching hospitals. The sites represented a geographical spread covering metropolitan and rural areas. The initial selection below was made on the basis of the trusts that ranked in the top 10 for 2 or more of the 12-hour breach measures (where patients have waited more than 12 hours from arrival in the ED to being seen and either discharged, transferred to another organisation or admitted to hospital) from the NHS Futures urgent and emergency mental health dashboard (NHS England, 2025b) and is based on data from the Emergency Care Data Set.

The three measures used to inform the selection were:

• Average length of delay over 12 hours – the highest ranking trusts being those where patients with mental health issues waited the longest.
• Total excess time that all patients with mental health issues impacted by delays over 12 hours had waited (results presented in minutes and calculated by the volume of people impacted by delays multiplied by the average length of delay over 12 hours) – the highest ranking trusts being those where either the numbers of people experiencing delays over 12 hours and/or the excess waiting times are higher.
• Highest volume of 12-hour breaches – the highest ranking trusts being those where the numbers of people impacted by 12-hour breaches was greatest.

The investigation then profiled the areas served by trusts in terms of the population profile, including ethnic and demographic diversity, levels of deprivation and geography (urban/rural), to inform the final selection.

Stakeholder engagement and consultation

During visits the investigation spoke with a range of staff working in EDs, psychiatric liaison services, crisis services, mental health crisis assessment centres, psychiatric clinical decision units, and environments where people in mental health crisis were being cared for. Speaking to staff was key, as well as seeing the adaptations made and observing ‘work as done’. In addition, focus groups were held to gather views and experiences from those working with people in mental health crisis in EDs.

The investigation engaged with:

• healthcare professionals including medical doctors, registered nurses, registered mental health nurses, AMHPs, Section 12 approved doctors, occupational therapists, psychiatrists, psychologists, social workers, drug and alcohol teams, on-site police, security staff, legal teams in trusts and trust management and executives
• a senior police force representative
• a subject matter advisor specialising in law
• integrated care boards
• national organisations including the Department of Health and Social Care, NHS England, Care Quality Commission, Nursing and Midwifery Council, Royal College of Emergency Medicine, Royal College of Psychiatrists, Getting It Right First Time Mental Health Team (GIRFT), Office of the Chief Allied Health Professions Officer for England, Cygnet, NHS Resolution, Association of Ambulance Chief Executives, Royal College of Paramedics, National Mental Health Forum Council, National AMHP forum, Mental Health Leads in Acute Trust Forum
• charitable organisations including the Centre for Mental Health, Action Against Medical Accidents (AvMA), Black Thrive Global, Suicide Crisis, Making Families Count, Rethink
• many patients, families and carers in person at their home, and patients and/or their family members in the ED or mental health urgent assessment centres.

The investigation engaged with stakeholders to gather evidence and check for factual accuracy, and for overall sense-checking. The stakeholders contributed to the development of the safety recommendations based on the evidence gathered.

Analysis of the evidence

The findings presented in this report were identified through triangulation of multiple evidence sources and consultation with stakeholders involved in the investigation. The AcciMap model (Svedung and Rasmussen, 2002) was used to inform the investigation approach to help understand risks across local, regional and national boundaries. The model focuses on identifying relationships between the different levels of the health and care system, which include government policy and budgeting; regulatory bodies and associations; local area management; physical processes and actor activities (what staff, people, organisations, systems did); and equipment and surroundings. The contributory factors are arranged into a series of levels representing the different parts of the health and care system. The investigation’s analysis focused on identifying connections between the different levels of the system (see figure A).

Figure A The AcciMap model