Annual report 2026: UK patient and general population data, 2013-2023

Date of publication: February 2026

About the report

The 2026 annual report from the National Confidential Inquiry into Suicide, Homicide and Safety in Mental Health (NCISH) provides findings relating to people aged 10 and above (in line with the Office for National Statistics) who died by suicide between 2013 and 2023 across the UK (England, Northern Ireland, Scotland and Wales) and Jersey.

Additional findings are presented on the number of people under mental health care who have been convicted of homicide, and those in the general population in the UK. Mental health care refers to care provided by both the NHS and the main private providers. The majority of patients are under NHS care. Complete details of our methodology are provided in our previous reports and on the NCISH resources page.

On this page:

• About the report
• Our database
• What the report covers
• Changes in the NCISH methodology
• Key findings
• Clinical messages
• Suicide in the general population
• Mental health patient suicide
• Real-time surveillance of mental health patient suicide
• Themes in this report
• Homicide in the UK
• Links to additional online data by UK country

Download the full report (PDF)

The NCISH database has been established for more than 28 years. It includes a national case series of suicide by patients who had been in recent (in the previous 12 months) contact with mental health services.  

The current suicide database stands at over 178,700 deaths by suicide in the general population, including over 45,500 patients. This internationally leading database allows NCISH to make recommendations for clinical practice and policy that will improve safety locally, nationally and internationally.

On this page:

• About the report
• Our database
• What the report covers
• Changes in the NCISH methodology
• Key findings
• Clinical messages
• Suicide in the general population
• Mental health patient suicide
• Real-time surveillance of mental health patient suicide
• Themes in this report
• Homicide in the UK
• Links to additional online data by UK country

Download the full report (PDF)

< Back to top

On this page:

• About the report
• Our database
• What the report covers
• Changes in the NCISH methodology
• Key findings
• Clinical messages
• Suicide in the general population
• Mental health patient suicide
• Real-time surveillance of mental health patient suicide
• Themes in this report
• Homicide in the UK
• Links to additional online data by UK country

Download the full report (PDF)

< Back to top

Restart of the Homicide Inquiry

Homicide by patients with mental illness is one of the most sensitive subjects in mental health and we recognise how much staff, friends and families are affected by these tragic incidents.  

We have been commissioned to resume the National Confidential Inquiry into Homicide in England, to reflect current public concerns regarding patient care. Detailed data collection on homicides ended in 2018. We will restart data collection in 2026. Our aim is to obtain information to improve patient care and respond to any emerging concerns about patient safety. 

On this page:

• About the report
• Our database
• What the report covers
• Changes in the NCISH methodology
  • Widening the NCISH definition of contact with services prior to suicide
  • Reporting of suicide in older age groups
  • Restart of the Homicide Inquiry
• Key findings
• Clinical messages
• Suicide in the general population
• Mental health patient suicide
• Real-time surveillance of mental health patient suicide
• Themes in this report
• Homicide in the UK
• Links to additional online data by UK country

Download the full report (PDF)

< Back to top

Patient suicide numbers and rates

• Over 2013-2023, there were 18,602 suicides by patients in the UK and Jersey, an average of 1,691 deaths per year, 26% of all general population suicides.
• The number of patient suicides increased in 2018 following a change in the standard of proof for suicide at inquest, with lower figures since then. The rate of suicide among patients under mental health care, which takes into account the total number of people in contact with mental health services, has fallen in England.

Social and clinical characteristics

• Nearly half of all patients who died by suicide lived alone (47%) or were unemployed (46%), and 17% had recently experienced serious financial problems.
• The majority (61%) had a history of self-harm and nearly a third (31%) had self-harmed in the previous 3 months. Alcohol (46%) and drug (38%) misuse, and comorbidity (56%), meaning more than one mental health diagnosis, were common.

Clinical care

• There were 4,735 (27%) patients who died by suicide in acute mental health care settings, including as in-patients (5%), in post-discharge care (13%) or under crisis resolution/home treatment teams (14%) (with overlap between the latter two groups), an average of 430 deaths per year.
• There were an estimated 66 suicides by mental health in-patients in 2023, around 4% of all patient suicides in that year. Overall, the number and rate of in-patient suicide have fallen over the report period, though figures have not decreased further in 2020-2023. The proportion of in-patients who died on the ward has increased from 36% in 2013-2016 to 47% in 2020-2023, in part driven by deaths in patients aged under 25 and by an increase in strangulation as suicide method.
• There were an estimated 205 deaths by suicide in the 3 months after discharge from mental health in-patient care in 2023, 12% of all patient suicides in that year. Based on our predicted figures, the number and rate of post-discharge suicide may have risen in 2021-2023. The highest risk was in the first 1-2 weeks after discharge, and of those who died in the first week, the highest number occurred on days 3 and 4 when taking day 1 as the day of discharge (54 and 55 patients respectively, or 18%). However, in 2020-2023, the highest number occurred later in the week, on days 4 (20, or 21%) and 6 (19, or 20%).

Suicide under crisis care settings

• In 2013-2023, there were an estimated 2,465 suicides by patients receiving treatment in crisis care settings, including crisis resolution/home treatment and other crisis care services (e.g.such as crisis houses), 14% of all patient suicides, an average of 224 deaths per year with no recent change. There were also 207 (6%) patients in 2019-2023 whose last contact was with liaison psychiatry services, an average of 41 deaths per year.
• Patients who died by suicide while under crisis care were older than other patients, with more aged 45-64 (45% compared with 38%). Nearly a third (32%) were married, 29% were employed and 44% lived alone.
• The most common primary diagnosis was depressive disorder (42%); overall, 34% had been ill for less than a year. A quarter (25%) died within 3 months of discharge from in-patient care. Self-harm in the previous 3 months was common (40%), as were adverse life events (58%) including serious financial difficulties (21%) and relationship breakup (14%).

Suicide by people aged 75 and above

• In the general population in 2013-2023, there were an estimated 5,184 suicides by those aged 75 and above, 7% of all suicides, an average of 471 deaths per year.
  There was an increase overall, driven by those aged 75-79 and by men aged 90 and above. A fifth (1,029, or 20%) were mental health patients. This meant that they had been in contact with mental health services in the year before death, an average of 94 deaths per year. This is a lower proportion in contact with services than the younger age groups.
• Half (51%) of older patients were living alone. 7% were in living in a nursing/care home. 12% had been bereaved in the previous 3 months.
• The most common primary diagnoses were depressive disorder (49%) and dementia (14%). Over half (55%) also had a major physical illness. Overall, they had lower rates of common suicide risk factors such as alcohol (12%) and drug misuse (2%). 41% had a history of self-harm and 19% had self-harmed in the previous 3 months.

Suicide and anxiety disorders

• In 2013-2023, there were an estimated 1,337 suicides by patients with a primary diagnosis of anxiety disorder, 7% of all patient suicides, an average of 122 deaths per year. The average number increased from 98 in 2013-2016 to 147 in 2020-2023. The increase was seen in men and women, and in all adult age groups.
• Most patients with anxiety disorder were in the 45–64 age group (41%). Over a third (38%) were married. 41% were unemployed; 10% were on long-term sick leave. The majority (68%) had an additional mental illness, most often depression (39%). Three-quarters (76%) were receiving antidepressants, in line with NICE anxiety guidance, but only a quarter (25%) were receiving psychological therapy. Previous self-harm (54%) and alcohol (35%) and drug (29%) misuse were less common among patients with anxiety. Short-term suicide risk was viewed by clinicians as not present or low in 83%.
• Around half of the patients under 25 with anxiety disorders were female (49%). Suicide-related internet use (32%) was more common among these young patients compared to other young patients who did not have anxiety disorders (18%).

Suicide and recent migrants

• In 2013-2023, there were 838 suicides by patients known by clinicians to have resided in the UK for less than 5 years, of whom 33 were seeking to stay in the UK (for example, asylum seekers and refugees seeking to continue living in the UK after visa expiration). These patients represented 5% of all patient suicides, an average of 76 deaths per year. The number has increased since 2017, overall and particularly in women. These figures are confirmed and not estimated; there may be a further increase in these figures as completeness of our data increases.
• Recent migrants were younger than other patients who died by suicide, with more aged 25-44 (43%). The majority (87%) were white. Almost half (46%) lived alone.
• Affective disorder (38%) was the most common diagnosis (including 28% with depressive illness and 10% with bipolar disorder); a quarter (26%) had a short (<12 months) history of mental illness. Recent migrants were more likely to have died after discharge from mental health in-patient care (17%). A higher proportion were discharged to socioeconomic adversity such as housing, financial or employment problems (34%) and poor social support (28%).

On this page:

• About the report
• Our database
• What the report covers
• Changes in the NCISH methodology
• Key findings
  • Patient suicide numbers and rates
  • Social and clinical characteristics
  • Clinical care
  • Suicide under crisis care settings
  • Suicide by people aged 75 and above
  • Suicide and anxiety disorders
  • Suicide and recent migrants
• Clinical messages
• Suicide in the general population
• Mental health patient suicide
• Real-time surveillance of mental health patient suicide
• Themes in this report
• Homicide in the UK
• Links to additional online data by UK country

Download the full report (PDF)

< Back to top

The following clinical messages are intended for clinicians, mental health services, Integrated Care Boards (ICBs) and Health Boards.

Acute mental health care settings

There has been a change in the pattern of suicide among mental health in-patients, in that a rising proportion of deaths now occurs on the ward itself. This rise is particularly seen in in-patients under the age of 25. In part, it reflects an increase in strangulation as a method of suicide. In the last two decades, wards have substantially reduced suicides by hanging. The clinical challenge now is to extend this to deaths in which ligatures, but not ligature points, are used, particularly by younger patients.

There are also signs of a change in the pattern of suicides in the first week after in-patient discharge, with the peak risk occurring later in the week. This may reflect the recent requirement in England for follow-up within 72 hours and it raises a concern that early post-discharge suicides have been postponed rather than prevented. It is important that these early post-discharge contacts anticipate any imminent deterioration.

Suicide under crisis care settings

The features of patients who die by suicide under crisis teams point to an area of future priority: depressed, a short history of illness, living alone, adverse life events and recent self-harm. Crisis services should review how well they are able to support this group. Similarly, these are the patients at risk for whom the new mental health emergency departments, as outlined in the recent 10-year health plan for England (PDF), should be designed.

Suicide by people aged 75 and above

The risk profile of older patients who die by suicide is one in which depression, physical illness, isolation and bereavement are prominent. These factors should be the target of prevention. Recent self-harm in older patients should be a warning sign for significant risk. The low rate of mental health service contact suggests the need for improvements in access for this age group.

Suicide and anxiety disorders

A primary diagnosis of anxiety disorder is an increasing feature of patient suicide. This contrasts with a view of anxiety in clinical practice and in the wider public, where it can be seen as a less severe condition, mainly affecting younger people. Our evidence indicates that, for some, it is a serious disorder that carries significant risk, despite fewer conventional risk factors such as self-harm and substance misuse. Services need to be alert to the risk, treat co-existing depression and make psychological therapies available.

Suicide and recent migrants

Our findings highlight social adversity in mental health patients who die by suicide after living in the UK for a relatively short time. Many face economic difficulties and have insufficient social support. This may exacerbate suicide risk, requiring multi-agency care planning at key points in the clinical pathway, such as in-patient discharge.

On this page:

• About the report
• Our database
• What the report covers
• Changes in the NCISH methodology
• Key findings
• Clinical messages
  • Acute mental health care settings
  • Suicide under crisis care settings
  • Suicide by people aged 75 and above
  • Suicide and anxiety disorders
  • Suicide and recent migrants 
• Suicide in the general population
• Mental health patient suicide
• Real-time surveillance of mental health patient suicide
• Themes in this report
• Homicide in the UK
• Links to additional online data by UK country

Download the full report (PDF)

< Back to top

Between 2013 and 2023, NCISH was notified of 71,519 deaths in the general population in the UK and Jersey that were registered as suicide or “undetermined”, an average of 6,502 deaths per year (Fig. 1). These are referred to as suicides throughout the report.  

Fig. 1: Number of suicides in the general population and by mental health patients, by UK country in 2013–2023.

There were:

• 2,470 suicide deaths in Northern Ireland, 722 (29%) were by patients;
• 4,087 suicide deaths in Wales, 915 (22%) were by patients;
• 8,279 suicide deaths in Scotland, 2,453 (30%) were by patients;
• 56,577 suicide deaths in England, 14,471 (26%) were by patients.

Additionally, there were 106 suicide deaths in Jersey; 41 (39%) were by patients.

In England and Wales there were higher rates of suicide following the lowering of the standard of proof at inquest introduced in 2018, and then a plateau (Fig. 2). However, figures in 2023 are expected to rise once late inquests are added.

In Scotland there were higher rates from 2018 compared to those in 2014-2017. Northern Ireland rates have increased in 2021-2023 compared to lower rates in 2015-2020 following a change in how some deaths are classified (see details of the review of suicide statistics in Northern Ireland). In Jersey the rates fluctuated, being based on small numbers; the average suicide rate in Jersey in 2021-2023 was 8.9 per 100,000 population.

Fig. 2: General population suicide rates by UK country (2013-2023)

Note: In Northern Ireland data prior to 2015 are not directly comparable with those in 2015 onwards due to a review of suicide statistics by the Northern Ireland Statistics Agency (NISRA). Rates in Jersey not shown due to small numbers.

There was some variation in the peak age in England, Scotland and Wales, but overall, the pattern was equivalent, with highest suicide rates in middle-aged groups, especially the 40-44- and 45-49-year age groups. (Fig. 3). There was also a relatively high rate in those aged 90 and above. In Northern Ireland the highest suicide rates were in younger people, especially in the 20-24- and 25-29-year age groups.

Fig. 3: Suicide rates in the general population by age-group, by UK country (2013–2023)

Note: Rates in Jersey not shown due to small numbers.

On this page:

• About the report
• Our database
• What the report covers
• Changes in the NCISH methodology
• Key findings
• Clinical messages
• Suicide in the general population
  • Number of suicides in the general population and by mental health patients, by UK country in 2013-2023
  • General population suicide rates by UK country (2013-2023)
  • Suicide rates in the general population by age-group, by UK country (2013-2023)
• Mental health patient suicide
• Real-time surveillance of mental health patient suicide
• Themes in this report
• Homicide in the UK
• Links to additional online data by UK country

Download the full report (PDF)

< Back to top

There were 18,602 mental health patients who died by suicide (this means people in contact with mental health services within 12 months of suicide) in the UK and Jersey in 2013-2023. This represents 26% of all suicide deaths, an average of 1,691 deaths per year (see Fig. 1: map showing number of suicides in the general population and by mental health patients, by UK country in 2013-2023 above).

There was an increase in the number of patient suicides occurring from 2018 in line with the change in the standard of proof (Office for National Statistics) for suicide that began in 2018. We did not see an increase in 2020 when COVID-19 began but are estimating slightly higher figures in 2021-2023 (Fig. 4), driven by an increase in England during these years (Fig. 5). However, this increase in England was not reflected in the rate of suicide. Taking into account the total number of people under mental health care, which was 35.5 per 100,000 mental health service users in 2023, this suicide rate has continued to fall long-term (see additional online data for further details). The number in Scotland has fallen since 2018 whilst recent numbers in Northern Ireland show an increase. In Wales, the numbers have generally been stable (Fig. 5).

Fig. 4: Number of mental health patients who died by suicide, by sex in the UK and Jersey (2013-2023)

Notes: The number of males and females in 2020 and 2022 do not total the overall figure due to rounding. Dashed lines represent estimated numbers in 2021-2023.

Fig. 5: Number of mental health patients who died by suicide, by UK country (2013–2023)

Notes: The number of males and females in 2020 and 2022 do not total the overall figure due to rounding.

Northern Ireland data prior to 2015 are not directly comparable with those in 2015 onwards due to a review of suicide statistics by Northern Ireland Statistics Agency (NISRA). Data from Jersey are not shown due to the low number of patient deaths by suicide.

Dashed lines represent estimated numbers in 2021–2023.

Method of suicide by mental health patients

The most common methods of suicide were:

• hanging/strangulation (9,306, or 50%)
• self-poisoning (4,022, or 22%)
• jumping/multiple injuries (2,431, or 13%)

Hanging/strangulation increased by 16% during 2013–2023, especially after 2017 when the standard of proof for suicide was lowered (Fig. 6). The increase was especially seen in women, rising from an average of 41% of all female deaths in 2013–2016 to 47% in 2020–2023. Deaths by jumping or multiple injuries fell by 27% between 2013–2020 but estimates show an increase in 2021–2023.

After a rise between 2015 and 2020, we estimate lower figures for self-poisoning deaths in 2021–2022 but a possible rise in 2023 (Fig. 6).

Opiates (including opioid compounds) were the most common substances used, accounting for nearly a third (1,058, or 32%) of deaths by self-poisoning; however, deaths using opiates or opioids fell by 35% between 2013 and 2023.

Fig. 6: Main suicide methods by mental health patients in the UK and Jersey (2013–2023)

Note: Dashed lines represent estimated numbers in 2021-2023.

Social characteristics of mental health patients who died by suicide (UK and Jersey, 2013-2023)

Socio-demographic characteristics

• The majority were male (12,048, or 65%), unmarried (11,739, or 73%), unemployed (7,361, or 46%) and nearly half (7,723, or 47%) lived alone (see additional online data for further information).
• 1,694 (10%) were aged under 25, including 283 (2%) aged under 18, 1,411 (8%) aged 18-24 and 979 (6%) aged 75 and above. Of all patients aged under 25, 218 (14%) had a history of local authority care.

Our report Suicide by Children and Young People identified common themes in the lives of young people who die by suicide. Information on suicides by children is also presented by the National Child Mortality Database (NCMD) Programme in their annual reports, alongside recommendations for prevention

• In 2016-2023, 440 (5%) patients were known to identify as lesbian, gay or bisexual and 104 (1%) were within a trans (including transgender, transsexual, non-binary) group. Of all patients, 1,206 (7%) were from an ethnic minority group.

Financial problems

There were 2,229 (17%) patients who had recently (within 3 months) experienced serious financial problems.

Bereavement

1,328 (9%) patients had been recently bereaved. Overall, 1,196 (11%) patients died on or near a date of significance (e.g., the patient’s birthday, on or near the anniversary of a family member of friend’s death, the birthday of a deceased family member).

Internet use

Suicide-related internet use (for example, visiting “pro-suicide” websites) was reported in 875 (9%) of all patients.

Clinical characteristics of mental health patients who died by suicide

• Over half (56%) of patients who died by suicide had a comorbid (or additional) mental health diagnosis. The proportion with comorbidity increased from 52% in 2013-2016 to 59% in 2020-2023.
• Rates of alcohol misuse (46%) and drug misuse (38%) were high (see additional online data for further information); 56% of patients had a history of alcohol or drug misuse.
• The majority (10,176, or 61%) of patients had a lifetime history of self-harm – this proportion fell in 2020-2023 (59%) compared to in 2013-2016 (65%). Nearly a third (4,832, or 31%) had self-harmed in the 3 months before suicide; in 3,439 (27%) instances, the patient had been seen by the emergency department for self-harm.

Diagnosis of mental health patients who died by suicide

The main primary mental health diagnoses are shown in (Fig. 7). Suicide by patients with depressive illness has generally been falling since 2016 but estimated figures show a rise in 2021-2023 (Fig. 8). The average number of patients with bipolar disorder increased from 121 in 2016-2019 to 144 in 2020-2023. Suicide by patients with schizophrenia or other primary psychotic disorders has not changed in 2019-2021, though the estimated number in 2022 shows a rise.

The number of suicide deaths in patients given a diagnosis of personality disorder may have fallen in 2021-2023 after an increase in 2018-2020. Suicide by patients with anxiety disorders rose between 2013 and 2023. The number of patients with alcohol or drug dependence or misuse who died by suicide fell after 2013 and figures in 2021-2023 are the lowest over the report period.

There were 288 (2%) patients with any diagnosis (primary or secondary) of an eating disorder. The average number increased from 21 in 2013-2016 to 30 in 2020-2023. There were 499 (3%) patients with any diagnosis of autism and 312 (2%) with attention deficit hyperactivity disorder (ADHD) who died by suicide. These numbers have increased since 2018 and estimated figures in 2021-2023 show together they account for approximately 139 deaths each year. There were differences in the younger age groups. In patients aged 18-24, 10% had a diagnosis of autism and in those aged under 18 this was 21%. 5% of those aged 18-24 and 8% of those under 18 had a diagnosis of ADHD.

Fig. 7: Primary diagnoses of mental health patients who died by suicide in the UK and Jersey (2013-2023)

Notes: The sum of all diagnoses totals 101% due to rounding. ADHD = Attention Deficit Hyperactivity Disorder. *includes both primary and/or secondary diagnosis. †other diagnoses include drug induced psychosis, dementia, learning disability, conduct-dissocial disorder, somatisation disorder, organic disorder, and other specified.

Fig. 8: Main primary diagnoses of mental health patients who died by suicide in the UK and Jersey (2013-2023)

Note: Dashed lines represent estimated numbers in 2021-2023.

Contact with services by mental health patients who died by suicide

Nearly half (8,075, or 46%) of all mental health patients had been in contact with services in the week before death (Fig. 9). At the final service contact, the immediate risk of suicide was viewed as not present or low in the majority of patients (11,750, or 81%); this proportion has not changed over the report period. There are now initiatives to move away from stratified assessment of suicide risk towards a more personalised approach to assessment and management. These include the NHS guidance on Staying Safe from Suicide and the Culture of Care programme.

Fig. 9: Timing of last contact with mental health services by patients who died by suicide in the UK and Jersey (2013-2023)

Clinical settings of mental health patients who died by suicide

During 2013-2023, there were 4,735 patients (27%) who died by suicide in acute care settings (in-patients, under crisis resolution/home treatment, recently discharged from in-patient care), an average of 430 deaths per year (Fig. 10). The proportion under acute care has fallen in 2020-2023 (26%) compared to 2013-2016 (29%). There was a fall in the number of patients who missed their last service contact, from an average of 350 in 2013-2016 to 322 in 2020-2023; the number of patients who were non-adherent with medication has not changed over the report period.

Fig. 10: Service characteristics of mental health patients who died by suicide in the UK and Jersey (2013-2023)

Note: These categories are not mutually exclusive.

There were 248 patients who died by suicide having been subject to a Community or Compulsory Treatment Order (CTO) at some time in 2013-2023. This accounts for 1% of all patient suicides, an average of 23 deaths per year. Two-thirds of these patients (157, or 65%) were under a CTO at the time of suicide. Overall, there were 299 patients under the care of an assertive outreach service, 2% of all patient suicides, an average of 27 deaths per year.

In-patient suicide

In-patient deaths include those that occur physically on the ward and those that occur off the ward (e.g. during authorised or unauthorised leave).

There were 953 in-patient deaths by suicide in 2013-2023, representing 5% of patient suicides overall during this time period. This percentage has decreased since 2016, dropping to 4% in 2023. 22 (2%) were aged under 18, representing 2% of all in-patients or 8% of all patients aged under 18; and 91 were aged 18-24, representing 10% of all in-patients or 7% of all patients aged 18-24. The average rate of in-patient deaths over the report period was 5.7 per 10,000 admissions.

There was a 45% fall in the number of in-patients who died by suicide between 2013 and 2023, although figures in 2020-2023 have not fallen (Fig. 11). We also found rates of in-patient suicide per 10,000 admissions fell by 35% in 2013-2023, considering the total number of in-patient admissions in the UK, but recent rates have not changed (Fig. 11).

343 (41%) died on the ward, half (413, or 49%) had left the ward with staff agreement, and 79 (9%) had left the ward without staff agreement or left with agreement but failed to return. Overall, a third (306, or 35%) had been detained under Mental Health Act (MHA) powers and 128 (14%) died within 7 days of admission.

There was an increase in the proportion of in-patients who died on the ward in 2020-2023 compared to in 2013-2016 (47% compared to 36%). The increase over these time periods was seen in men (57, or 48% compared to 78, or 32%) and in those aged under 25 (19, or 66% compared to 15, or 38%), especially women and girls aged under 25 (14, or 74% compared to 10, or 43%).

The majority (311, or 91%) of deaths on the ward were by hanging/strangulation/asphyxia. The number of these deaths fell in 2015-2017 but have since not changed, and account for an average of 28 deaths per year in 2020-2023. When separating hanging/strangulation/asphyxia, there has been a rise in deaths by strangulation on the ward over the report period, from 25 in 2013-2016 to 32 in 2020-2023.

Fig. 11: Number and rate of mental health in-patients who died by suicide in the UK and Jersey (2013-2023)

Notes: Rates of suicide exclude Jersey due to unavailable denominator data. Dashed lines represent estimated figures in 2021-2023.

Mental health patients who died by suicide after recent discharge

There were 2,250 patients who died by suicide within 3 months of discharge from in-patient care, 13% of all patient suicide deaths, an average of 205 deaths per year. 13 (1%) were aged under 18 and 172 (8%) were aged 18-24. The number and rate of suicides by patients within 3 months of discharge have risen in 2021-2023 (Fig. 12) In the UK, the average rate of suicide over the report period was 13.9 per 10,000 discharges.

Post-discharge suicide deaths were most frequent in the first 1-2 weeks after leaving hospital (Fig. 13). Of patients who died in the first week after discharge, the highest number occurred on days 3 (54, or 18%) and 4 (55, or 18%) after leaving hospital (day one = day of discharge) (Fig. 14). Over the report period, the peak days of suicide after discharge have become later in the week, e.g. in 2013-2016 the highest number were on days three (30, or 24%) and five (22, or 17%) and in 2020-2023 they were on days four (20, or 21%) and six (19, or 20%).

Of all post-discharge suicides, 190 (10%) died before the first follow-up appointment.

There were 311 (23%) patients known to have been discharged to housing, financial or employment problems. 255 (19%) were reportedly discharged to poor social support.

Overall, 188 (9%) had initiated their own discharge. This figure was higher in those who died within a week of discharge (43, or 14%). 247 (12%) died after being discharged from an in-patient unit which was out of their local area.

Fig. 12: Number and rate of mental health patients who died by suicide within 3 months of in-patient discharge in the UK and Jersey (2013-2023)

Note: Rates of suicide exclude Jersey due to unavailable denominator data. Dashed lines represent estimated figures in 2021-2023.

Fig. 13: Number of mental health patient deaths by suicide per week following discharge in the UK and Jersey (2013-2023)

Fig. 14: Number of mental health patient deaths by suicide per day in the week following discharge in the UK and Jersey (2013-2023)

On this page:

• About the report
• Our database
• What the report covers
• Changes in the NCISH methodology
• Key findings
• Clinical messages
• Suicide in the general population
• Mental health patient suicide
  • Number of mental health patients who died by suicide, by sex in the UK and Jersey (2013-2023)
  • Number of mental health patients who died by suicide, by UK country (2013-2023)
  • Methods of suicide by mental health patients
  • Social and clinical characteristics of mental health patients who died by suicide
  • Diagnosis of mental health patients who died by suicide
  • Contact with services by mental health patients who died by suicide
  • Clinical settings of mental health patients who died by suicide
  • In-patient suicide
  • Mental health patients who died by suicide after recent discharge
• Real-time surveillance of mental health patient suicide
• Themes in this report
• Homicide in the UK
• Links to additional online data by UK country

Download the full report (PDF)

< Back to top

We have been developing a national real-time surveillance (RTS) of suspected suicide deaths (which were submitted before confirmation by inquest) of patients under the recent care of mental health services in England. 

Our initial focus is on in-patients and those who died within 2 weeks of discharge – patients in close proximity to services. 

The aim is to collect information early to support suicide prevention efforts.

This includes information about: 

• emerging novel suicide methods 
• suicide-related internet use  
• travelling to a specific location 
• potential clusters 
• systemic problems in care, and specific problems that patients face. 

We are taking a staged approach to establishing this real-time data collection. First, we piloted the data collection in 8 NHS trusts in south-east England. In April 2024, we expanded the data collection to all mental health trusts across England, asking for clinicians to notify us of any suspected suicide death of an in-patient or a patient recently (within 14 days) discharged from in-patient care, where the death occurred from 1 January 2024. Clinicians can therefore complete an online questionnaire without waiting for the inquest or a request from us. 

Numbers of suspected suicide deaths by mental health patients 

To date, we have been notified of 104 suspected suicide deaths by patients under mental health care in England, the majority (66, or 63%) by in-patients.  

We were notified of 47 suspected suicide deaths which occurred between 1 January 2025 and 2 December 2025. Of these, over half (27, or 57%) were men and most were aged between 25 and 44 (14, or 30%) and 45-64 (18, or 38%). 89% were white.  

The most common method of death was hanging/strangulation (17, or 36%), followed by jumping/multiple injuries (9, or 19%). Adverse experiences in the 3 months prior to death included isolation or loneliness (7, or 15%) and workplace and/or financial problems (11, or 23%)​.

Clinical characteristics  

Of the 47 patients we were told about, most were in-patients (25, or 53%) with the remaining 22 (47%) patients dying within 2 weeks of discharge from in-patient care. The most common diagnoses were affective disorder (19, or 40%) and schizophrenia or other primary psychotic disorders (9, or 19%). 

In-patient deaths

Most patients (23, or 92%) were admitted locally. At the time of death, two-thirds (15, or 60%) had been detained under Mental Health Act (MHA) powers and 10 (40%) had been admitted voluntarily. 6 (24%) patients self-harmed within 24 hours prior to death. 10 (40%) patients died on the ward itself, and over half (13, or 52%) were on agreed leave at the time of death.

Post-discharge patient deaths

For 9 (47%) patients who had been recently discharged, the last admission lasted less than a week. 4 patients had been detained under the MHA during this last admission. The majority (16, or 76%) of patients had their first follow-up within 3 days of discharge; in most cases (11, or 52%) this was face-to-face contact. 6 (29%) patients had experienced financial and/or housing problems following discharge.

On this page:

• About the report
• Our database
• What the report covers
• Changes in the NCISH methodology
• Key findings
• Clinical messages
• Suicide in the general population
• Mental health patient suicide
• Real-time surveillance of mental health patient suicide
  • Numbers of suspected suicide deaths by mental health patients
  • Clinical characteristics
  • In-patient deaths
  • Post-discharge patient deaths
• Themes in this report
• Homicide in the UK
• Links to additional online data by UK country

Download the full report (PDF)

< Back to top

In this section we provide more detailed data on specific topics, reflecting high or increasing risk. In this year’s report the themes are: those in crisis care settings, older (≥75 years) patients, those with a primary diagnosis of anxiety disorders, and those who were recent migrants to the UK.

Suicide in crisis care settings

In 2013-2023, there were an estimated 2,465 suicides by patients receiving treatment in crisis care settings. This represents 14% of all patient suicides, an average of 224 deaths per year.

These included 2,377 (14%) patients under crisis resolution/home treatment services (CRHT), an average of 216 deaths per year, and 88 (1%) whose last contact was with other crisis care teams (for example, crisis houses), an average of 8 deaths per year.

The number of patients under crisis care has not changed in recent years following a peak in 2018 and a subsequent fall (Fig. 15). There were also 207 (6%) patients whose last contact was with the related setting of liaison psychiatry services (data from 2019 onwards), an average of 41 deaths per year – these patients are not included in the following sections.

Fig. 15: Suicide by patients under crisis care settings in the UK and Jersey (2013-2023)

Note: Dashed lines represent estimated numbers in 2021-2023.

Characteristics of patients who were under crisis care settings and died by suicide (UK and Jersey, 2013-2023)

The majority (1,605, or 65%) of patients under crisis care were male. More were aged 45-64 compared to other patients who died by suicide (1,079, or 45%, compared with 5,589, or 38%); 181 (8%) were aged under 25, including 30 (1%) aged under 18. Around a third were married (736, or 32%) and 655 (29%) were employed (see: additional online data for further information).

1,006 (44%) patients under CRHT lived alone. Although for most (1,116, or 60%), the care plan included additional social support from outside the home, such as from a relative, friend or neighbour, those living alone were less likely to receive additional support (368, or 47% compared with 729, or 70%).

Patients under crisis care were more likely to have depressive disorder (982, or 42%, compared with 4,166, or 29%); fewer had schizophrenia or other primary psychotic disorders (246, or 10%, compared with 2,159, or 15%). Over half (1,310, or 56%) had a comorbid (additional) mental illness and almost a quarter (537, or 23%) a comorbid physical illness. A third had been ill for less than a year (729, or 34%, compared with 2,309, or 19%) including 245 (12%) who had been ill for less than 3 months, more than in other patients in other settings (518, or 4%).

Recent (previous 3 months) self-harm was more common among patients under crisis care (925, or 40%, compared with 3,693, or 29% for those in other settings); 617 (70%) had been seen by services (mostly an emergency department) for this episode of self-harm. Overall, 1,362 (58%) had experienced recent adverse life events, including serious financial difficulties (416, or 21%, compared with 1,735, or 16%), relationship breakup (284, or 14%, compared with 1,093, or 10%) and bereavement (228, or 10% compared with 1,076, or 9%). In 2020-2023, nearly half (214, or 48%) had reported insomnia, more than other patients (646, or 33%).

A quarter died within 3 months of discharge from in-patient care (594, or 25%, compared with 1,534, or 11%); 248 (10%) died within 2 weeks of discharge; 144 (6%) within a week. Of the patients specifically under CRHT, 813 (39%) had been receiving this care for less than a week, 217 (27%) of whom died within 3 months of discharge. Immediate (61% compared with 84%) and long-term risk (40% compared with 60%) of suicide were less likely to be viewed as not present or low in patients under crisis care settings compared to other patients.

Suicide in people aged 75 and over

Age categories for defining older adults have varied across settings and over time. In this report, we have chosen to describe people aged 75 and over, because they have a different clinical profile to those aged 65-74 years (who, for example, are less likely to have a physical illness or dementia).

In 2013-2023, there were 5,184 suicides in the general population in those aged 75 and over, 7% of all suicides, an average of 471 deaths per year. The number increased overall and specifically in men and women aged 75-79, and in men aged 90 and above.

An estimated 1,029 (20%) were suicides by mental health patients – people who had been in contact with mental health services in the previous 12 months – indicating an average of 94 deaths per year. This group represented 6% of all patients who died by suicide:

• 479 (3% of all patients who died by suicide) were aged 75-79
• 291 (2%) were aged 80-84
• 174 (1%) were aged 85-89
• 78 (0.4%) were aged 90 and over.

The number of patients aged 75 and over fell after a peak in 2018, but has since been increasing overall and in men (Fig. 16).

Fig. 16: Suicide by patients aged 75 and over in the UK and Jersey (2013-2023)

Notes: The number of men and women in 2021 and 2023 do not total the overall figure due to rounding. Dashed lines represent estimated numbers in 2021-2023.

Characteristics of patients aged 75 and over who died by suicide (UK and Jersey, 2013-2023)

The majority (654, or 64%) of patients aged 75 and above were men. Most were married/co-habiting (383, or 42%) or widowed (340, or 37%) (see additional online data for further information). Around half were living alone (489, or 51%) and 67 (7%) were in a nursing/care home. Fewer older patients were from an ethnic minority group compared to younger patients (33, or 4%, compared with 1,173, or 7%).

The most common primary diagnoses were depressive disorder (457, or 49%) and dementia (130, or 14%). Two-thirds (605, or 66%) were receiving care under older people’s mental health services. Older patients were more likely to have had a short (<1 year) history of mental illness compared to younger patients (262, or 32% compared with 2,888, or 20%).

Over half had major physical illness (488, or 55% compared with 3,699, or 24% for younger patients); the most common were cardiovascular disease (50%), impaired mobility (28%), musculoskeletal disease (27%), diabetes (19%) and respiratory disease (18%).

384 (41%) had a history of self-harm and 171 (19%) had self-harmed in the previous 3 months. 101 (12%) had experienced a recent bereavement. Factors commonly associated with suicide were less common in older patients, including alcohol misuse (111, or 12% compared with 7,520, or 48%, for younger patients), drug misuse (19, or 2% compared with 6,277, or 40%) or recent adverse life events (including financial difficulties) (392, or 43% compared with 7,671, or 51%).

A quarter died by self-poisoning (246, or 25%) and, of the drug types used in self-poisoning, paracetamol and paracetamol/opiate compounds were the most common (28%).

Both short-term (716, or 85% compared with 11,034, or 80%) and long-term (508, or 65% compared with 7,277, or 56%) risk of suicide were more often viewed as not present or low compared to younger patients.

Suicide in mental health patients with anxiety disorders

In 2013-2023, there were an estimated 1,337 patients who died by suicide who had received a primary diagnosis of anxiety disorder (includes anxiety, phobia, obsessive compulsive disorder, panic disorder and post-traumatic stress disorder). This represents 7% of all patient suicides, an average of 122 deaths per year. The average annual number increased from 98 in 2013-2016 to an estimated 147 in 2020-2023 (Fig. 17). The increase was seen in men and women, and in all age groups except those aged under 18.

Fig. 17: Number of mental health patients with a primary diagnosis of anxiety disorder who died by suicide in the UK and Jersey (2013-2023)

Notes: Male and female numbers in 2019 and 2022 do not total the overall figure due to rounding. Dashed lines represent estimated numbers in 2021-2023.

Characteristics of patients with a primary diagnosis of anxiety disorder who died by suicide (UK and Jersey, 2013-2023)

832 (62%) patients with anxiety disorder were male (see additional online data for further information). Most patients (514, or 41%) were in the 45–64 age group; 110 (9%) were aged under 25, including 26 (2%) aged under 18. More patients with anxiety disorder were married (442, or 38% compared with 3,871, or 26%). 465 (41%) were unemployed and 112 (10%) were on long-term sick leave. 469 (40%) were living alone.

They were more likely to have a comorbid (additional) psychiatric illness (853, or 68% compared with 8,672, or 55% of those without a primary diagnosis of anxiety disorder), most commonly depressive illness (488, or 39%). Nearly a quarter (266, or 23%) had been ill for less than a year.

Fewer patients with anxiety disorder had a history of common risk factors, including self-harm (654, 54% compared with 9,327,or  61%), and alcohol (423, or 35%, compared with 7,047, or 47%) or drug (353, or 29%, compared with 5,801, or 38%) misuse.

The majority (898, or 76%) were receiving antidepressants, in line with NICE guidance on the management of anxiety, but only a quarter were receiving psychological therapy (278, or 25%). A third (296, or 34%) were receiving benzodiazepines.

Short-term risk of suicide was viewed as not present or low in 83%.

Patients aged under 25 with anxiety disorders

Of the 110 patients aged under 25 with anxiety, around half (54, or 49%) were female. Over a third were full-time students (34, or 37%), 20 (19%) were from an ethnic minority group, and a fifth (12, or 20%) identified as LGBT. Suicide-related internet use (such as visiting pro-suicide websites) was reported in more younger patients with anxiety compared to other patients aged under 25 (17, or 32% compared with 148, or 18%).

A similar proportion had a history of self-harm (78, or 73%, compared with 1,126, or 77%) and alcohol and/or drug misuse (54, or 51%, compared with 879, or 61%). Over a third were receiving psychological therapy (38, or 38%). Long-term risk of suicide was more likely to be viewed as not present or low in young patients with anxiety disorder compared to other young patients (64% compared with 50%).

Suicide in mental health patients who were recent migrants to the UK

In 2013-2023, there were 838 patients known by clinicians to have resided in the UK for less than 5 years, of whom 33 were seeking to stay in the UK (for example, an asylum seeker, a refugee, or a patient seeking to continue living in the UK after their visa had expired). This group represented 5% of all patient suicides, an average of 76 deaths per year. The number has increased since 2017, especially in women (Fig. 18). These figures are confirmed and not estimated, and may increase substantially further in 2021-2023 as NCISH figures become more complete.

Fig. 18: Number of mental health patients who died by suicide and were recent migrants to the UK and Jersey (2013-2023)

Note: The numbers shown are confirmed and not projected. As data collection is not yet complete for 2021-2023, it is likely that these numbers will increase.

Characteristics of patients who were recent migrants and died by suicide (UK and Jersey, 2013-2023)

548 (65%) patients who were recent migrants were male. Overall, they were younger than other patients, with most aged 25-44 (362, or 43%, compared with 5,955, or 36%); 95 (11%) were aged under 25 and 11 (1%) were aged under 18. In 678 (87%), their ethnicity was reported as white (see additional online data for further information). Of those from an ethnic minority, the most common groups were South Asian (Indian, Pakistani, Bangladeshi) (23, or 3% of recent migrants) and Black African (19, or 2%). Overall, half were unemployed (371, or 50%) and 355 (46%) were living alone.

A history of self-harm (58%), alcohol (47%) and drug (39%) misuse were common. More recent migrants had a history of childhood abuse compared to other patients (230, or 40%, compared with 4, 156, or 34%).

The most common primary diagnoses were affective disorders (308, or 38%, including depressive illness (225, or 28%) and bipolar disorder (83, or 10%)) and schizophrenia or other primary psychotic disorders (137, or 17%). Over half (54%) had a comorbid (additional) mental illness and a quarter had a recent (<12 months) history of illness (177, or 26%). 175 (23%) had missed their last contact with services.

A higher proportion died after recent (past 3 months) discharge from in-patient care (133, or 17%, compared with 1,977, or 13%), including within a week of discharge (26, or 14%, compared with 271, or 8%). Around a third had been discharged to housing, financial or employment problems (27, or 34%, compared with 275, or 22%) and a quarter discharged to poor social support (22, or 28%, compared with 228, or 18%).

Overall, half had experienced recent adverse life events (404, or 53%), including serious financial problems (21%), workplace problems (8%) and legal problems (4%). Clinicians estimated short-term suicide risk to be not present or low in 81%.

On this page:

• About the report
• Our database
• What the report covers
• Changes in the NCISH methodology
• Key findings
• Clinical messages
• Suicide in the general population
• Mental health patient suicide
• Real-time surveillance of mental health patient suicide
• Themes in this report
  • Suicide in crisis care settings
  • Suicide in people aged 75 and over
  • Suicide in mental health patients with anxiety disorders
  • Suicide in mental health patients who were recent migrants to the UK
• Homicide in the UK
• Links to additional online data by UK country

Download the full report (PDF)

< Back to top

In 2013-2023, NCISH was notified of 5,733 homicide convictions in England, Wales and Scotland, an average of 521 per year. There were 6,097 victims, an average of 554 per year.

There were an estimated 622 patients in recent (<12 months) contact with mental health services who were convicted of a homicide offence, 11% of general population homicides, an average of 57 per year. There were 642 victims, an average of 58 per year. Around 1 in 4 patients convicted of homicide were under 25 (151, or 25%), and 41 (7%) were under 18.

The number of convictions fell after 2013 and remained similar in 2014-2019 (Fig. 19), but estimated figures in 2021-2023 show an increase. Figures in 2020 may be low because of delays in court prosecutions and outcomes due to COVID-19. These estimated figures should be treated with caution. Homicides by people in contact with mental health services based on year of offence showed no increase (Fig. 19).

Across the UK, patients represented 11% of people convicted of homicide (Table 1). This figure continues to be higher in Scotland (15%) and Wales (15%), and where the general population homicide rates are also higher.

Northern Ireland homicide convictions data were available only until 2014. There were 39 homicide convictions in 2013-2014, and fewer than 3 homicides were by patients.

Fig. 19: Homicide by patients in recent (<12 month) contact with mental health services in England, Wales and Scotland, by year of conviction and year of offence (2013-2023)

Note: Dashed lines represent estimated numbers in 2021-2023.

Table 1: Number of homicide offenders in the general population and by patients in recent (<12 month) contact with mental health services, by UK country excluding Northern Ireland (2013-2023)

On this page:

• About the report
• Our database
• What the report covers
• Changes in the NCISH methodology
• Key findings
• Clinical messages
• Suicide in the general population
• Mental health patient suicide
• Real-time surveillance of mental health patient suicide
• Themes in this report
• Homicide in the UK
• Links to additional online data by UK country

Download the full report (PDF)

General population (N)

Patients under mental health care (N%)

England: 4,996

Scotland: 492

Wales: 245

UK (excluding NI): 5,733

England: 509 (10%)

Scotland: 76 (15%)

Wales: 37 (15%)

UK (excluding NI): 622 (11%)

< Back to top

Further data can be found by using the links below:

1. UK additional online data
2. England additional online data
3. Northern Ireland additional online data
4. Scotland additional online data
5. Wales additional online data
6. Projecting suicide figures
7. Evidence reference guide

On this page:

• About the report
• Our database
• What the report covers
• Changes in the NCISH methodology
• Key findings
• Clinical messages
• Suicide in the general population
• Mental health patient suicide
• Real-time surveillance of mental health patient suicide
• Themes in this report
• Homicide in the UK
• Links to additional online data by UK country

Download the full report (PDF)

< Back to top