Guidance

Responding to suicide clusters: guidance for stakeholders and members of cluster response groups

Updated 19 March 2026

Foreword 

Handling a suicide cluster is suicide prevention at its most acute. The risk is immediate, the people affected may be numerous or not yet identified. They may be individuals; they may be whole communities. The watchwords are preparation, vigilance and urgency.

Suicide prevention continues to be a national priority and since the last version of this guidance was published in 2019, we have had a better understanding of the importance of clusters to this wider task, more evidence of which groups are most at risk and how best to respond.

Most of us in suicide prevention believe clusters are more common than we previously realised. The link between one tragic death and another varies. Links can include a close relationship (in-person or virtual), risk factors in common, or the use of a particular suicide method or location. The issue of clusters can therefore merge with other prevention priorities such as online risk and high frequency locations.

Some settings are especially vulnerable. In schools, for example, actions to pre-empt a cluster should start after a single suicide death. In the recent study by my research group of suicides among university students, potential clusters often went unrecognised.

Taking control of a cluster is a skilled task, often falling to a group of people who, though highly experienced, are doing it for the first time. Clear leadership is vital, good data essential, and communications carefully worded.

More than anything a suicide cluster is a human tragedy calling for compassion and care. Devastated families will need support, as will shocked friends and colleagues, some of whom may themselves be at risk. A setting like the NHS presents its own problems - people may work across several sites and be well known to a broad network.

This updated guidance provides a framework for action that local authorities and integrated care boards will adapt to their own needs. Clusters remind us all that suicide prevention is constantly changing, and we have to be constantly responsive.

Professor Sir Louis Appleby CBE

Co-chair of the National Suicide Prevention Strategy Advisory Group

Executive summary

This guidance is intended for those with responsibility for suicide prevention in local authorities and their partner agencies. It focuses on identifying and responding to suicide clusters, primarily through the function of a cluster response group. The information within this guidance may be of relevance following single as well as multiple suicides.

Suicide clusters understandably cause great concern and may lead to hasty, uncoordinated and potentially harmful responses. It is important that plans for such occurrences are prepared in advance to ensure an evidence-based and effective response.

Organisations with a responsibility for suicide prevention need to remain vigilant for potential clusters and possible contagion between deaths by suicide and put strategies in place to forestall this. This resource is designed to support those organisations and has been developed as a contribution to the Suicide prevention strategy for England: 2023 to 2028.

Addressing suicide clusters is the responsibility of cluster response groups, which should be associated with the relevant multi-agency suicide prevention groups, generally led by local authorities.

This guidance covers a range of relevant topics, including:

• how to prepare for a cluster response
• identifying clusters using data
• the management of a cluster response
• standing down a response

Within this, it provides suggestions for identifying:

• who may be at risk of suicidal acts due to the influence of other people’s suicidal behaviour
• the likely mechanisms involved in clustering
• the effects of suicide on other individuals

The steps that need to be taken at local level to respond to a suicide cluster are described. This necessitates the development of a suicide cluster response plan and identification of individuals and agencies that will deliver such a plan. Early indicators of clusters are also described, together with the need to establish the facts and avoid premature and possibly unhelpful responses. This document suggests responses to possible suicide clusters, including:

• preventing unhelpful media reporting
• supporting those bereaved
• identifying individuals and groups who may be particularly vulnerable

It should be noted that whilst this document is based on the best available evidence, this is an emerging field and recommendations are based on good practice, informed by expert opinion and examples of practical experience.

Introduction

Purpose of this guidance

This guidance has been prepared to support the identification, response and management of suicide clusters. It has been designed for use by members of staff who are in a cluster response group. It should be used alongside a local suicide prevention plan, guidance for which can be accessed in Suicide prevention: developing a local action plan.

Summary of changes

Evidence-based approaches to cluster response are crucial. Understanding about risk groups and knowledge of how best to respond to suicide clusters has increased since the 2019 version of this guide was published.

Changes to this updated guidance include:

• refining the target audience of this guidance to professionals who may be directly involved in cluster response as part of a cluster response group
• updated suggested roles and responsibilities for activity associated with cluster response within the remit of a cluster response group
• placing the activity of cluster response within a wider suicide prevention context. This includes suggesting governance structures for cluster response reporting and oversight
• greater consideration of thresholds that may trigger a cluster response, including consideration of high need groups
• updated terminology and refreshed structure to aid simplicity of use

The guidance may be used for any or combinations of the following:

• development of a local cluster response plan (likely to be local authority led)
• implementation of a local cluster response plan and standing up a local cluster response
• support to identify, manage and stand down a cluster response

Introduction

Death by suicide is a leading cause of premature mortality in England. It can affect anyone either directly or indirectly.

According to the Office for National Statistics (ONS), the suicide rate for England and Wales in 2023 was 11.4 deaths per 100,000 people. Suicide rates fluctuate for a variety of reasons but are driven by socioeconomic factors such as age, gender, livelihood and ethnicity.

In 2023, the Suicide prevention in England: 5 year cross-sector strategy was published, outlining over 100 actions that government, the NHS and other partners could take to reduce suicide rates across the country.

Deaths by suicide can influence further suicidal behaviour in others, in a phenomenon known as contagion, which can lead to clustering. Suicide clusters are important public health emergencies, that without timely, evidence-based and data-informed intervention can lead to an escalation of the number of deaths linked to a single event.

It has been estimated that suicide clusters contribute to only a small percentage of total suicides in England and Wales, but they have the potential to have a significant impact on community wellbeing. Early detection and targeted, informed action can prevent further loss of life and mitigate further community impacts. Therefore, continued effective cluster response is required to minimise these risks.

Target audience of this guidance

This updated version of the suicide cluster guidance refines the scope of the guidance to the implementation of cluster responses and will be linked to:

• local cluster response plans
• the formation of cluster response groups if a cluster is identified or suspected

Therefore, this guidance is primarily aimed at:

• directors, deputy directors or consultants of public health
• local authority suicide prevention leads
• data analysts responsible for real time suicide surveillance
• communications leads involved in suicide response
• police involved in suicide response

It is recognised that depending on the specific situation that a variety of other colleagues may be involved in suicide cluster response. This guidance may therefore also be relevant to:

• safeguarding leads involved in suicide response
• representatives of mental health trusts (where deaths are connected to this setting)
• representatives of drug and alcohol services (where deaths are connected to this setting)
• integrated care board (ICB) suicide prevention leads
• representatives of the child death overview panel
• suicide prevention or safeguarding representatives of organisations where deaths are connected to that setting, such as:
  • mental health trusts
  • drug and alcohol services
  • schools
  • universities
  • children’s mental health services

Terminology

The terms ’suicide cluster’ and ‘contagion’ can be anxiety provoking and might also be seen as insensitive to bereaved families. We acknowledge the term contagion may also oversimplify the complex links that put people at risk and should be used with caution. However, these are the terms generally used by professionals working in the suicide prevention field. They have therefore been used in this guide. It is important, however, that agencies which prefer to use other terms ensure consensus and consistency in their use.

Forming a cluster response group

When forming a cluster response group, note that:

• given the likely involvement of the police, you may also wish to consider coordinating in the first instance with local police forces
• the national police chiefs’ council (NPCC) should be involved when:
  • deaths cross multiple police force areas
  • cases involve individuals or agencies outside of the United Kingdom
• cluster response group attendees should be at a sufficient level of leadership to be able to commit their organisation to any required actions, with adequate time to participate in the process
• communications teams from attending agencies should work together to coordinate the management of messaging and reporting across all forms of media, particularly if there is a likelihood of coverage in either local or national media
• the agencies and personnel to be involved should reflect the nature and circumstances of the cluster
• group members need to be able to contribute to the core functions of the cluster response group
• it is important to balance the appropriate level of information with containment of anxiety. Note that too many people included in cluster response groups can escalate anxiety and fear, increasing risk of contagion

Evidence suggests that some settings - particularly NHS and university settings - are more susceptible to clustering of suicidal or self-harm activity. Organisations should have protocols in place that notify the relevant partners in local authorities that will trigger the local cluster protocol.

Glossary

Affected other

An affected other is a person who is, or is at risk of, being impacted negatively by somebody’s death by suicide.

Bereavement

Bereavement is the period of grief and mourning experienced by someone after the death of a connected individual.

Child death overview panel

A child death overview panel is a statutory multi-agency group responsible for reviewing all child deaths in a specific area to identify learning opportunities and prevent future deaths.

Cluster

A cluster is a group of deaths, suicide-related activity or self-harm that are linked in some way by time, death, method, location, geography or demography.

Cluster response group

A cluster response group is a group of professionals with a clear practical role in a suicide cluster response which is active during a confirmed cluster in a local community.

Cluster response plan

A cluster response plan is a locally designed plan which:

• sets out a protocol for responding to predicted, emerging or confirmed suicide clusters
• defines roles and responsibilities of main stakeholders required to deliver that protocol

Contagion

Contagion is the spread of something (for example, suicide deaths, methods or self-harm) from one individual or group to another in an uncontrolled way.

Emerging method

An emerging method is a method used to result in more than one death by suicide that is less prevalent and may be unique.

Health and wellbeing board

A health and wellbeing board is a statutory forum established at local authority level. It is represented by leaders from the health and care system to collaborate on improving the health and wellbeing of the local population and reducing health inequalities.

Multi-agency suicide prevention group

A multi-agency suicide prevention group is a group that typically operates at local authority level. It is:

• likely to be made up of agencies with a responsibility for suicide prevention and other interested organisations
• responsible for delivering the multi-agency suicide prevention plan

Postvention

Postvention is organised support and interventions provided after suicide to affected others.

Potential cluster

A potential cluster is a scenario whereby suicide or suicide related activity appears to be linked but a cluster response group has not yet been stood up and therefore the cluster has not been confirmed.

Real Time Suicide Surveillance

A local system of data collection used to monitor unconfirmed suicide and suicide-related activity within a defined geography.

Self-harm

Self-harm is the act of deliberately injuring oneself, without intent of ending life.

Senior responsible officer

The senior responsible officer is a senior leader in an organisation who is personally accountable for ensuring a project or programme achieves its objectives.

Suicide

Suicide is a means of death whereby the individual has ended their life themselves with intent to do so.

Terms of reference

Terms of reference define the purpose, scope, and structure of a project, committee or any group working towards a shared goal.

Preparing for a suicide cluster

Development of a cluster response plan

Although rare events, the impacts of suicide clusters can be widespread. As such, it is recommended that local areas are covered by a cluster response plan.

This guidance recommends that the cluster response plan is coordinated at upper tier local authority level, with oversight by the corresponding director of public health. It should be developed by the multi-agency suicide prevention group, which in turn ideally feeds into the health and wellbeing board. This recommended structure will:

• provide appropriate oversight of any cluster response
• enable clear communication channels
• create opportunities for learning and reflection of cluster responses to be shared in relevant forums

Essential stakeholders

Suicide clusters can impact a wide range of stakeholders. However, for the purposes of cluster response, this guidance recommends keeping directly engaged partners to a minimum. This minimises the risk of needless escalation. The following are considered essential stakeholders for a cluster response group who should attend cluster response meetings. Additional agencies and personnel to be involved should reflect the nature and circumstances of the cluster.

The recommended minimum stakeholder list for cluster response groups includes:

• director of public health, deputy director of public health or consultant in public health who holds evidenced competency in health protection principles, in which suicide cluster theory is based. Responsibilities may include chairing cluster response meetings and advising on contact tracing of impacted individuals
• local authority suicide prevention lead. This stakeholder is likely to be based in the public health team and will have expertise in cluster response theory and excellent understanding of real time surveillance and local population need. Responsibilities may include conducting contact tracing, supporting partners to conduct their actions and leading reporting back to the cluster response group
• data analyst responsible for real time suicide surveillance, to understand and communicate the data relating to the suicide cluster. They may represent the local police force area or the local authority
• communications lead, to define a comms plan for the cluster response. This will include:
  • managing media enquiries
  • supporting local reporting
  • monitoring social media activity
  • developing and advising on evidence-based comms for partners to use
• police involved in suicide response. The police are likely to:
  • have been involved with the first response
  • have access to real time data
  • be involved with the media response
  • be able to facilitate intelligence sharing where cross-boundary episodes occur
  • direct affected people to support

Components of a suicide cluster response plan

Important topics that should be considered in a cluster response plan include:

• surveillance
• data and information sharing
• media
• contact tracing
• signposting to bereavement support
• governance and reporting
• learning and review
• support to staff

Further guidance and information for each of these topics is available throughout this guide.

Appointing a senior responsible officer

Feedback from local authority suicide prevention leads has noted that previous versions of this guidance have not identified a clear leadership model for suicide cluster responses. This in turn has potentially left the suicide prevention lead without sufficient support to lead the response, define the actions and undertake a lot of the work.

As the theory used in cluster responses lifts heavily from health protection principles - specifically those principles that might be used to limit the spread of infectious disease through a community - it is necessary for an individual with the appropriate training in health protection principles to be involved in an infectious disease outbreak. Therefore, it is recommended that directors of public health should take on senior responsible officer duties for a cluster response, lending their support and expertise to the local suicide prevention lead. This responsibility may be deferred to deputy directors of public health or consultants of public health who also have the necessary training in health protection principles.

Remit of a cluster response group 

The function of a cluster response group is to implement the local cluster response plan in response to real time surveillance or local intelligence that suggests a suicide cluster has occurred, is occurring or is at risk of occurring. The group’s remit should include:

• establishing the facts and confirming (or rejecting) a cluster is present
• information sharing
• continuous monitoring of real time surveillance throughout the cluster response
• identifying vulnerable individuals through contact tracing and reducing the risk of contagion
• communications and media
• reporting of progress to the senior responsible officer and multi-agency suicide prevention group or health and wellbeing board
• advising on the implementation of evidence-based interventions to prevent further contagion of the cluster

The remit should be captured in a terms of reference document, which should be produced in parallel with the cluster response plan.

Workplace support for mental health

Suicide cluster responses are difficult circumstances to operate professionally in, and they may impact the mental health of staff involved. It is therefore recommended that the lead local authority ensures that staff wellbeing offers have adequate provision for staff involved in a cluster response group. This includes:

• backfill of other duties to enable protected time to conduct the response
• line manager and/or clinical supervision
• clear boundaries about out of hours working
• access to an occupational health offer

Information sharing and confidentiality

Although it is important for multiple agencies to work together and share information when planning and implementing suicide cluster response plans, it is crucial that confidentiality and data protection are considered at all stages.

Regardless of which stakeholders form the cluster response, it is important to develop an information sharing agreement and secure the relevant agreement and signatures to allow the cluster process to work. An example of a data sharing agreement is provided in annex A.

Child death overview panels

Existing processes for reviewing and responding to any sudden death, including suspected suicides, of people under the age of 18 years (through the local child death overview panel rapid response team) should be tied in with the suicide cluster response plan. There should be links between the multi-agency suicide prevention group and local bodies responsible for safeguarding children (in England these are local safeguarding children boards) and the local child death overview panel.

Identifying a suicide cluster

Determining if a suicide cluster is occurring

The term ‘suicide cluster’ describes a situation in which more suicides than expected occur in terms of time, place or population, or a combination.

A suicide cluster usually includes 3 or more linked deaths. However, some populations or settings may trigger a cluster response at 2 deaths. These settings and populations should be defined locally based on the needs of the local population, likely to be set out in a joint strategic needs assessment (JSNA). The suicide prevention strategy for England 2023 to 2028 identifies the following groups as being at elevated risk of suicide:

• children and young people
• middle-aged men
• people who have self-harmed
• people in contact with mental health services
• people in contact with the justice system
• autistic people
• pregnant women and new mothers

Note that this is not an exhaustive list and may not represent local population needs.

In addition to general risk of suicide, evidence suggests some population groups are particularly vulnerable to suicide clusters, including young people^{[footnote 1] [footnote 2]}, people with mental health diagnoses^{[footnote 3]}, and prisoners^{[footnote 4]}. Clusters of suicidal behaviour are more common in certain settings, including schools^{[footnote 4]}, universities^{[footnote 5]}, psychiatric facilities^{[footnote 3]}, prisons^{[footnote 4]} and workplaces^{[footnote 6]}. People who share similar characteristics or identify psychologically with individuals who have taken their lives may be vulnerable to the contagious effects of suicide, which may contribute to the development of clusters^{[footnote 7]}. For these groups it is likely that 2 deaths should trigger a cluster response.

It is important that, irrespective of the setting, the director of public health (or equivalent) takes a lead in the cluster response and that the cluster response group is established at an early stage if there are possible connections between suicides. These should be informed by real time suicide surveillance.

Conclusively identifying a suspected suicide cluster should not stand in the way of responding to concerns. If in doubt, convene a cluster response group and follow the process as outlined by the cluster response plan.

Cluster types

Point clusters (or spatial-temporal clusters)

A point cluster is a greater than expected number of suicides that occur within a time period in a specific location. This might be in a community or an institution (for example school, university or psychiatric inpatient setting).

Mass clusters (or temporal clusters)

A mass cluster is a greater than expected number of suicides within a time period which are spread out geographically.

Population cluster

A population cluster is a greater than expected number of suicides within a population group. This may be particularly within groups of higher risk such as children and young people or inclusion health groups.

Echo cluster

An echo cluster is a cluster that occurs in the same location as a previous cluster, but some time later.

Clusters involving a specific method of suicide

Sometimes clustering can involve a particular method of suicide. This can occur in all types of cluster.

Self-harm cluster

While the focus of this guide is mainly on suicides, it is essential to recognise that self-harm can also occur in clusters^{[footnote 8] [footnote 9]}, as can mixed clusters of both suicide and self-harm.^{[footnote 10]} Linked episodes of self-harm may be a precursor to a suicide cluster^{[footnote 11]}. While the term ‘suicide cluster’ is used throughout most of this guide, much of the guidance could apply equally to self-harm clustering.

Causes of a suicide cluster

Suicide clusters may result from ‘contagion’, whereby one or more than one person’s suicide influences another person to engage in suicidal behaviour or increases their risk of suicide ideation and attempts. A variety of mechanisms may be involved, such as modelling, and vulnerable individuals tending to come together in social groups^{[footnote 1]}.

The people involved are likely to already be vulnerable, perhaps because of existing mental illness and thoughts of suicide, or factors such as family discord or previous bereavement. However, it is also possible that exposure to suicidal behaviour may make a person contemplate a suicidal act for the first time. It can also provide information about possible methods or locations.

Suicide contagion may occur:

• horizontally; where a person with similar characteristics to the individual or individuals is influenced to take their own lives. These are more likely to be point or population clusters
• vertically; where the death of a celebrity or other known figure in the public eye influences the suicidal activity of a member of the public. These are more likely to be associated with echo and method clusters

Defining a death as a suicide

In England, determination of whether a death is officially a suicide depends upon a coroner’s inquest. Unfortunately, inquests usually occur a considerable time, often many months, after deaths. Response to possible suicide clusters must occur rapidly to prevent further deaths, and therefore identification of deaths in which suicide is the likely cause must take place at the earliest possible stage, without awaiting coroners’ verdicts.

Defining a cluster

To make a judgement about possible occurrence of a suicide cluster the cluster response group should:

• review available data to understand the timing and circumstances of the suicides. It is useful to record characteristics of the individuals involved and the nature of their deaths in a template covering potentially important factors. This template might include:
  • name
  • gender
  • ethnicity
  • date of death
  • date of birth
  • method
  • location of death
  • home address
  • occupation or workplace
  • faith group
  • GP
  • record of previous suicide attempts
  • record of self-harm
  • history of contact with mental health services
• identify any similarities and possible links between the deaths (for example in terms of method used, possible contagion through exposure to other suicides, community issues, occupation or social connections). Plotting information about suicides on a map can also help identify potential connections (both geographical and social). Mapping can be carried out with the assistance of local public health analysts or local police
• establish a timeline of events surrounding suicides, including factors such as geography, police and healthcare data, coroner’s information and anniversaries of previous suicides
• consider the thresholds that have been set in the cluster response plan, including those for priority groups

If, after a cluster response group is convened in response to a potential cluster, it is determined that the deaths are not linked, the cluster response group should be stood down.

Note that even where a cluster is not occurring, if there is a single death or unconnected deaths, similar consequences to a cluster may occur. This may include escalated media response, appearance of memorials and risk of further use of an identified method or location.

The role of the media

Media may be a significant influence in the development of a potential suicide cluster. This includes traditional media, such as print, newspapers and television, as well as the internet through social media platforms. The latter have become increasingly influential, especially as these can spread inaccurate and harmful information ^{[footnote 12] [footnote 13] [footnote 14]}. The media has a role to play in minimising the possibility of contagion through safe reporting.

Responding to a suicide cluster

Standing up a cluster response

If intelligence indicates a potential cluster (at least 2 deaths that are possibly linked), the cluster response group should convene within 72 hours. An example agenda for the group’s first meeting is at annex B. The group’s purpose is to help prevent the development or further contagion of a cluster.

In the first meeting, the cluster response group should agree:

• the frequency of meetings
• the frequency of data review
• an initial schedule of meeting dates that indicates how long the work of the cluster response group is expected to continue

The group will also be responsible for co-ordinating activity, which can include:

• an overarching review assessment of impact from the response plan
• checking progress against specific actions, milestones and outcomes
• monitoring media activity and the impact of the media strategy

The group will also be responsible for monitoring any further suicides, attempted suicides or incidents of self-harm that are consistent with the characteristics of the cluster.

The remit of a cluster response group includes:

• establishing the facts and confirming a cluster is present
• information sharing
• continuous monitoring of real time surveillance throughout the cluster response
• identifying vulnerable individuals through contact tracing and reducing the risk of contagion
• communications and media
• reporting of progress to the senior responsible officer and multi-agency suicide prevention group or health and wellbeing board
• advising on the implementation of evidence-based interventions to prevent further contagion of the cluster

Establishing the facts and confirming a cluster is present

In the first stages of a response, it is important to collect as much relevant information about the cases involved in the suspected cluster.

This information can be used to:

• consider possible connections and links (such as method, community issues or social connections) and relevant actions that need to be taken
• determine whether a cluster is present. The cluster response group should consider all the data available to determine whether the deaths are connected in a way that puts further people at risk whether that be due to the method used, the location or the population group impacted

If any death considered was in a child (under 18 years old) members of the group must ensure that a link has been made with the local child death overview panel, including the child safeguarding partnership or local safeguarding board.

Information sharing

If there are clear concerns that a suicide cluster may be occurring, information should be shared across relevant professional groups, subject to appropriate confidentiality. Relevant agencies (police, mental health trusts or bereavement services) should be given factual information about suicides.

The affected family or families should be consulted and advised with sensitivity, so they understand what information is to be shared and why. There may be certain facts families do not want to be shared and they may be uncomfortable about inference that a death has been a suicide before a coroner’s inquest has been completed. This should be respected as much as possible.

Continuous monitoring of real time surveillance

It is likely that the cluster has been identified through local surveillance. This surveillance will need to continue throughout the response to monitor any further deaths that could be associated with the initial cases, including suicide attempts and self-harm.

Identifying vulnerable individuals through contact tracing

Contact tracing is the identification of individuals who may be impacted by a suicide (or suicides) in an attempt to provide prevention intervention to minimise risk. The cluster response group should use contact tracing to identify potentially vulnerable individuals and prevent contagion. Examples of people to consider when contact tracing may include:

• family and friends
• witnesses to the suicide or individuals who found the body
• work colleagues
• people from the same or similar demographic

It is important that the cluster response group identifies everybody who may have been affected by the deaths and refers them to support to reduce the risk of further harms.

The circles of vulnerability theory can help cluster response groups identify people who may be vulnerable following a suicide. Vulnerability can be thought of in terms of interlocking concentric circles representing 3 themes:

• geographical proximity - the physical distance between a person and the incident. For example, people discovering the body of someone who has died by suicide or those exposed to the immediate aftermath may be more at risk. Extensive or sensationalised news or social media coverage may extend the geographic boundaries of people who may be vulnerable
• social proximity - the social closeness or distance to the person who has died by suicide. Family members and close friends are likely to be particularly vulnerable. It is also important to consider individuals in communities such as schools, faith groups and wider friendship groups, including online social networks
• psychological proximity - how close or distant someone relates psychologically to the person who has died by suicide. Some people may identify with the deceased, for example, those of a similar age, sex, faith or sexuality

People bereaved by suicide should be given information about available support at the earliest opportunity. Ideally this should be provided by the police who attend the scene of a suspected suicide and followed up by bereavement support agencies. Additional consideration should be given to those who may have additional vulnerabilities. People who might be considered vulnerable include:

• young people
• people with mental illness
• people experiencing substance misuse
• people who self-harm
• people who feel responsible for the suicide or who are facing allegations as a result
• people who share similar characteristics with the individual who has died (for example ethnicity, people with a learning disability, people with autism)
• people who identify psychologically with the person who has died
• people with previous experience of suicide or self-harm in people close to them
• people who are socially isolated and lack family or social support
• communities with historical cultural trauma such as the Gypsy, Roma or Traveller community

Minimising risk of contagion

Suicide can be devastating for families, friends, work colleagues, teachers and others, and any suicide usually affects a large number of people. Suicide bereavement can itself increase suicide risk^{[footnote 15]}. While this may apply especially to people close to the person who died, there may be particularly vulnerable individuals across all categories of those who are exposed to suicide due to mental health problems, isolation, history of self-harm, and other psychological, psychosocial or environmental factors. These individuals may be known or unknown to the deceased and known or unknown to services. Undertaking contact tracing and offering timely bereavement support to individuals identified as being at risk is therefore essential.

The range of individuals who may be affected by suicide^{[footnote 16]} include:

• suicide bereaved long term, including:
  • family
  • close friends
• suicide bereaved short term, including:
  • friends
  • peers
  • work colleagues
  • teachers
• first responders: those directly involved, for instance those who have found the body or were called to the incident. This group may include:
  • paramedics
  • police officers
  • train drivers
  • health and social care staff
• suicide exposed, including:
  • local groups and communities
  • passersby
  • social groups
  • faith groups
  • acquaintances
  • wider peer groups, including social media

Some of these individuals may already be vulnerable. Support After Suicide’s support for witnesses provides useful guidance.

Communications and media

It is essential that there is a single point of media contact. This should be the media communications lead of the suicide cluster response group. The members of the suicide cluster response group should carefully consider a media communication plan, including making statements to the media when developing the cluster response plan. Reporting is more likely in the event of suicide clusters, where a suicide involves a young person or where an unusual method is used^{[footnote 17]}.

There may be concerns about how a suicide is reported in the media. To support appropriate reporting, see:

• the independent press standards organisation’s media guidance on reporting suicide
• Samaritans’ guidance for media on how to report suicide responsibly

It has been suggested that more positive media coverage such as reports of individuals who experienced suicidal ideation and overcame it may be associated with preventing suicide. In particular, digital media has the potential to provide crisis support, reach a broad range of people during a cluster including more marginalised groups, deliver therapy and give information about sources of support and advice^{[footnote 12]}.

When dealing with the media, the cluster group should:

• have a single point of contact within the cluster response group
• encourage media outlets to consider the Samaritans’ guidance
• avoid sensationalising information
• respect confidentiality preferences of friends and family of the deceased
• avoid disclosing information about locations and methods
• avoid speculation about a ‘single trigger’ for a suicide. Suicide is complex and seldom the result of a single factor. It is likely to have several interrelated causes and this complexity should be portrayed in news reporting where possible
• avoid dramatic language, such as ‘suicide epidemic’ and ‘hot spot’, and sensationalist pictures or video (for example of a suicide location)
• refrain from publishing the content of suicide notes

The communications lead should work with local suicide bereavement services and people with lived experience to support and advise families of the deceased about:

• possible media interest
• how to respond to media enquiries
• how to engage with social media

The communications lead should also share any press statements or responses with the families before publication.

Reporting

While a suicide cluster response is ongoing, regular updates should be provided by the suicide prevention lead to the senior responsible officer, who may also be the chair of the cluster response group. The frequency of updates should be determined in the first cluster response meeting by the chair. The group should also ensure that updates are reported to the multi-agency suicide prevention group and health and wellbeing board. Guidance around confidentiality should be considered in these updates.

Updates should cover:

• the number of deaths linked to the cluster
• progress made with regard to contact tracing and identification of vulnerable individuals and groups
• the postvention support that has been provided to these groups
• other interventions that are being considered for implementation
• what interest the media has taken in the cluster, including reporting by the media and enquiries made to organisations affected by the cluster

Evidence-based interventions to prevent further contagion of the cluster

Once the cluster response group has identified at-risk individuals, groups and populations using contact tracing, the group should consider what interventions are appropriate at a whole population, targeted population and individual level. It is not the cluster response group’s responsibility to implement these interventions but to give advice back to the multi-agency suicide prevention group for consideration.

Population approaches

The approaches should promote community awareness of mental health and wellbeing, suicide prevention and health-seeking behaviour. Examples of local outreach interventions include: 

• drop-in sessions
• print and online media to share stories of people who have sought help, coupled with appropriate signposting to sources of support

Targeted approaches

The population approaches listed above should be refined to only reach the populations identified as high-risk.

Individual approaches

Individuals identified as being at risk should be supported to access, or be referred to, postvention services for additional support.

Cross-boundary clusters

Where suicide clusters cross local authority areas, there must be close collaboration with neighbouring authorities, and all affected areas should be represented on the cluster response group.

Suspected clusters can even cross national and international borders. Where local real time suicide surveillance determines that this has happened, the directors of public health (or local equivalent) should coordinate with their counterparts in the relevant areas.

Role of social media

During a cluster, social media can be a way to reach a huge number of people, including vulnerable people, and promote a sense of social connection and support. It is likely that communities affected by a cluster will use social media to inform people about the death or deaths, create virtual memorials and post messages, often of sympathy. However, users of social media may also sometimes blame people or organisations for the death.

At this time, it is useful for those involved in responding to suicide clusters to work with communities or organisations to:

• use social media to let people know where they can go for help and support
• use social media to promote suicide prevention messages
• issue advice on how to intervene or who to contact if people are aware of, or concerned about, messages they see online in the aftermath of a suicide
• organise an appropriate member of staff or other professional within an organisation to meet with relevant individuals to discuss what is being shared online
• let people know that negative comments are often left on memorial pages but acknowledge that these sites are often created to honour individuals who have died

Supporting staff throughout the response

Being professionally involved in a suicide cluster response can be challenging with the potential to negatively impact on mental health and wellbeing. All staff involved in the response should be:

• reminded that this is a challenging area of work and feelings of stress and anxiety are normal, and therefore seeking help is expected and encouraged
• provided with training and information by their workplace to enable them to recognise the signs and symptoms which indicate that they may need support
• enabled to access organisational support if required

Professionals and other involved staff can be directed towards the resources produced by Mind, which support staff who are working in this challenging area.

Standing down a cluster

Timing

Determining when to stand down a cluster response can be challenging, particularly as there may be a fear of standing down too early and exposing communities to further risk of harm. The decision to stand down should always be informed by the data provided by local surveillance systems.

There is little evidence to guide groups on determining when to stand down a cluster. Groups - guided by their senior responsible officer and real time surveillance - should therefore use their best judgement to take a proportionate approach. Cluster response groups may want to consider:

• time since last suicide (or self-harm) related activity
• scale of contact tracing undertaken
• emergence (or lack of emergence) of new intelligence that would inform action

Lessons learned

A crucial part of cluster response is reflection and review. This practice should first start within the cluster response group and then be fed back to the multi-agency suicide prevention group where there is learning relating to the broader prevention plan (for example, restricting access to high-risk locations such as bridges or train stations). Guidance on this is available at Suicide prevention: suicides in public places.

The purpose of this exercise is to consider what could have been done differently to prevent similar circumstances again. This may include consideration of:

• what could have prevented the initial death or deaths from happening in the first place
• what could have prevented the subsequent clustering of deaths
• whether the cluster response plan is fit for purpose

These reflections may relate to:

• actions directly in the cluster response group’s control, such as effective contact tracing
• wider factors that should be escalated and shared with the multi-agency suicide prevention group and health and wellbeing board, such as interventions at high-risk locations

Ongoing monitoring

Ongoing monitoring associated with a cluster should revert to already established surveillance arrangements such as those provided by real time suicide surveillance systems.

Particular attention should be paid to important dates and milestones that may be associated with the cluster, for example, the anniversaries of deaths.

Supporting staff after the response

Being professionally involved in a suicide cluster response can be extremely challenging and has potential to negatively impact on mental health and wellbeing. All staff involved in the response should be:

• asked if they have any connections to the deaths in the first instance
• supported throughout the response and be able to remove themselves without judgement from the cluster response group
• provided with the training and information to enable them to recognise the signs and symptoms which indicate that they may need support
• enabled to access organisational support if required. It is anticipated that organisations will have the relevant support services available to their workforce

A date should be set for a structured staff debrief at the earliest opportunity. The purpose of the debrief, once the cluster response group is stood down, is to identify learning from the process in operational terms but also to assess staff wellbeing and support needs, given the challenging nature of this work. If the cluster response continues for an extended period, then consideration should be given to limiting the length of tenure for individual representatives on the group.

It should also be noted that some individuals may not feel comfortable disclosing their mental health needs in a group setting. Therefore, routes to approach senior members of staff (such as the chair of the cluster response group or their own line manager) should also be made available.

Continuing wider suicide prevention activity

A cluster response is a time limited piece of work in response to a sudden change in suicide related activity, informed by real time surveillance. Once a cluster response has concluded, it is anticipated that a local area will be covered by wider suicide prevention activity through a local multi-agency suicide prevention plan. Further guidance is available at Public Health England’s Suicide prevention: developing a local action plan.

Annex A: example of an information sharing agreement

Note that this is a specific local example which will need to be adapted to individual local circumstances and agreed with the necessary input and review from local data protection officers.

Suspected suicide cluster

Data sharing agreement

Version: 1

Next review: [insert date]

This data sharing agreement is dated [insert date]

Parties

Such organisations or parties who share data for the purpose and who agree to the terms of this data sharing agreement.

1. Background

1.1 One or more of the parties have identified a suspected suicide cluster. This data sharing agreement is to facilitate and govern the lawful exchange of information between all parties and third party organisations and individuals in order to identify, assess and respond to the suspected cluster and any associated safeguarding.

1.2 This data sharing agreement should be read in conjunction with the [insert name of local area] multi-agency suicide cluster response plan and any applicable legislation including the Data Protection Act 2018, but where there is any inconsistency between the protocol and this data sharing agreement this agreement will take precedence.

1.3 Unless otherwise set out the words and phrases used in this data sharing agreement have the same meanings as are set out in the Data Protection Act 2018 and the UK GDPR.

1.4 This data sharing agreement sets out the framework for the sharing of personal data in connection with the suspected cluster. As the parties have their own statutory responsibilities, they have a responsibility to use their own professional judgment in relation to assessments, risk and actions in respect of any shared personal data, notwithstanding they may collaborate and contribute to a multi-agency assessment and response. Therefore, once shared personal data is received by a party the retention and processing of such shared personal data will be as an independent data controller and subject to and in accordance with any applicable statutory or professional obligations and that party’s own policies and procedures and decision making.

2. Purpose

2.1 The purpose of this agreement is to help and promote the identification of individuals at risk or associated with the suspected cluster who require safeguarding care and support and to provide such safeguarding care and support. Processing, including sharing, of personal data is necessary as such individuals may be in different areas and different information will be held by the parties. The parties or third party organisations may have different safeguarding and other statutory responsibilities and any decisions relating to identification, risk and safeguarding actions as a result may occur either alone or in a multi-agency setting.

2.2 The legal basis for sharing personal data is:

• Article 6(1)(d) processing is necessary in order to protect the vital interests of the data subject or of another natural person
• Article 6(1)(e) processing is necessary for the performance of a task carried out in the public interest or in the exercise of official authority vested in the controller

In respect of special category data:

• Article 9(2)(c) processing is necessary to protect the vital interests of the data subject or of another natural person where the data subject is physically or legally incapable of giving consent
• Article 9(2)(g) processing is necessary for reasons of substantial public interest
• Article 9(2)(h) processing is necessary for the purposes of the provision of health or social care or treatment

In respect of criminal offence data the sharing complies with section 10(5) of the Data Protection Act 2018 as it meets the conditions in Paragraph 18 of Part 2 Schedule 1 and Paragraph 30 Part 3 Schedule 1 of that Act.

2.3 The sharing of the personal data is necessary and proportionate as receivers of shared information will need the personal data in order to identify statutory responsibilities towards the data subjects or others and then take safeguarding action to protect the data subjects or other natural persons. The aim of the data sharing is to ensure those parties or organisations with responsibility in those areas have relevant and appropriate information to inform decision making to identify and provide an individual or co-ordinated service or support to individuals who are or may be part of the suspected cluster or who are associated with individuals who are or may be part of the suspected cluster for their safety or wellbeing. The data sharing is fair as it will benefit the individuals as it will contribute to their safety and health and intervention and prevention of death or injury and is in the public interest.

2.4 The parties agree to share personal data for the purpose of:

(a) identifying individuals who may be part of or at risk of being part of the suspected cluster, assessing the risk to such individuals, responding to or carrying out safeguarding actions in relation to that risk either as regards the individuals at risk or other individuals, family, relatives or those known to the individuals believed to be at risk and may include but is not limited to the following specific purposes:

• identify and assess individuals, groups and populations at different levels of need
• help to understand the circumstances around the death or serious self-harm incident and the people affected
• promote wellbeing
• prevent death or serious harm
• coordinate effective and efficient responses to assessed need
• enable early interventions to prevent the escalation of risk of harm
• prevent abuse and harm that may increase the need for care and support
• maintain and improve good practice in safeguarding adults, children or young people
• reveal patterns of abuse or harm that were previously undetected and that could identify others at risk of abuse or harm
• identify low-level concerns that may reveal people at risk of harm, declining mental health, self-harm, and suicide
• help people to access the right kind of support to reduce risk and promote wellbeing
• help identify people who may pose a risk to others and, where possible, work to reduce offending behaviour

The parties shall not process the shared personal data in a way that is incompatible with the above purposes unless otherwise allowed by law.

2.5 Each party shall appoint a single point of contact (SPOC) who will work together to reach an agreement with regards to any issues arising from the data sharing and to improve the effectiveness of the data sharing including reviewing this data sharing agreement. The points of contact for each of the parties are: [insert names, addresses, and emails for each contact].

3. Compliance with data protection laws

3.1 All parties confirm that they will only share personal data in accordance with applicable data protection legislation where they consider it necessary and lawful to do so for the purposes set out above and will only process any personal data shared with them in accordance with applicable data protection legislation and such processing will be fair and lawful.

3.2 Each party as an independent controller will provide clear and sufficient information to data subjects, in accordance with such legislation, of the purposes for which it will process their personal data, the legal basis for such purposes and such other information as is required by the data protection legislation as soon as possible.

3.3 Information will be shared on a ‘need to know’ basis and therefore different information may be shared with different parties.

3.4 It is acknowledged by all parties that in cases of urgency and emergency and where appropriate personal information may be shared with third party organisations in order to carry out real time monitoring, checking or for safeguarding purposes. Any sharing will be in accordance with the above purpose and in compliance with the remainder of this agreement.

3.5 All parties confirm that they have such valid registrations as are required by the Information Commissioner or legislation which covers the data sharing which is the subject of this data sharing agreement.

3.6 All parties confirm that shared data will be subject to the requirements of the Data Protection Act 2018 including data subjects’ rights, retention, security and further processing. Each party as an independent controller confirms that any shared data received will be held, protected, processed and deleted in accordance with relevant legislation.

3.7 Where any request is received from a data subject in relation to their data protection rights as regards shared data the parties will use reasonable endeavours to liaise with the other parties if appropriate to ensure the data subjects rights are complied with.

4. Shared personal data

4.1 The following types of personal data may be shared between the parties if considered necessary for the purpose:

• gender
• ethnicity
• date of death
• date of birth (and age)
• mode of death
• location of death
• home address
• clubs associated with the individual (for example, sports clubs)
• GP practice registered at
• previous medical history (including previous suicide attempts, history of self-harm or known mental health diagnosis)
• contact with mental health services
• social media presence
• possible triggers for suicide (such as financial or family issues)
• any other information considered relevant

Note that some of this information will in some cases also be special category personal data within clause 4.2 below.

4.2 The following types of special categories of personal data may be shared between the parties if considered necessary for the purpose:

• racial or ethnic origin
• political opinions
• religious or philosophical beliefs
• genetic or biometric data used to uniquely identify a natural person
• data concerning a natural person’s physical or mental health or condition, sex life or sexual orientation

4.3 Criminal offence data may be shared between the parties if considered necessary for the purpose.

4.4 All sharing of personal data under this data sharing agreement will take place using secure methods as follows:

Data sent by email must be sent using a secure email system (for example GCSX, PNN, GSX, Gist, CJSM, NH net and N3), or encrypted or password protected where secure email is not available.

4.5 The parties will use all reasonable endeavours to ensure any shared data is accurate and up to date.

5. Miscellaneous

5.1 The parties will ensure all staff and officers involved with the processing of any shared data have received appropriate training.

5.2 In complying with the data protection legislation the parties acknowledge that any information shared is limited to the amount of information required, is only shared with a limited number of people on a need to know basis, is only used for the purposes and is shared using secure emails or portals or is otherwise password protected and encrypted at all times.

5.3 This data sharing agreement does not require any personal data to be shared but where a party considers it is appropriate to share personal data for the purposes will do so in accordance with this agreement. As the shared data and any actions taken by the parties will depend on circumstances at the time the parties will use their professional judgement to decide what and when to share in all the circumstances and having appropriate regard to data protection obligations and the rights of data subjects and compliance with this agreement and the data protection requirements by the other parties.

5.4 If appropriate the parties will review this data sharing agreement, including its effectiveness, every 12 months or once there is considered to be no further need to share data for the purposes, whichever is earlier.

5.5 If any party considers there has been a data breach, whether by themselves or another party, in respect of shared data or the data subjects in relation to the purposes, if appropriate the other parties will be notified as soon as possible and will provide such assistance as is reasonable bearing in mind the circumstances of the breach and any safeguarding concerns.

5.6 Written confirmation that a party will follow the contents of this data sharing agreement will be provided. Where an organisation or individual not listed as a party at the beginning of this agreement provides written confirmation that they will follow the contents of this data sharing agreement at the next review of the agreement they will be added as a party for future reference.

Annex B: suggested agenda for initial suicide cluster response group meeting

1. Identify an individual to take notes of the meeting and provide ongoing administrative support.

2. Confirm purpose of meeting is to:

• confirm leadership and co-leadership of the suicide cluster response group
• establish facts surrounding suspected suicides and determine possible links between deaths
• agree appropriate level of escalation and intervention
• review membership of the suicide cluster response group, identify additional members (those required to attend meetings and those with whom links should be established) and clarify roles
• agree and record actions and timescales

3. Surveillance: consider the following points:

• details and circumstances of deaths: data from real time monitoring
• understanding the events in the context of the organisation concerned; for example, is there something going on in the wider community that is affecting the organisation, or vice versa?
• identify anniversaries of previous suicides
• identify any reported increases in non-fatal self-harm
• systematically record details and circumstances to aid identification of similarities and possible connections between deaths
• information sharing
• identify further investigations - what additional information is required from whom?
• identify who needs to be informed of concerns and agree communication processes
• agree confidentiality and information-sharing processes

4. Media: consider the following points:

• identify media communication lead of suicide cluster response group
• identify media communication leads within affected organisations
• agree communication strategy between organisations and with internal and external media
• agree content for media and a press statement
• identify possible social media concerns
• agree process for monitoring known Facebook memorial sites and responding to concerning posts with supportive statements and signposting

5. Bereavement support: consider the following points:

• identify bereavement support that is already in place or has been provided
• identify additional support needs (using circles of vulnerability)
• identify bereavement support agencies
• identify any shortfall and make contingency plans
• identify any additional support requirements for bereavement support providers

6. Monitoring of plan: consider the following points:

• agree monitoring (including record keeping and storage) and evaluation strategies
• agree frequency of suicide cluster response group meetings

1. Haw C, Hawton K, Niedzwiedz C, Platt S (2013) Suicide clusters: a review of risk factors and mechanisms. Suicide and Life-Threatening Behaviour; 43: pages 97-108 ↩ ↩²

2. Hawton, K, Hill, N T, Gould, M, John, A, Lascelles, K and Robinson, J (2020) Clustering of suicides in children and adolescents. The Lancet Child and Adolescent Health, 4(1), pages 58-67 ↩

3. Haw CM (1994) A cluster of suicides at a London psychiatric unit. Suicide and Life-Threatening Behaviour; 24: pages 256-66 ↩ ↩²

4. Niedzwiedz C, Haw C, Hawton K, Platt S (2014) The definition and epidemiology of clusters of suicidal behaviour: A systematic review. Suicide and Life-Threatening Behaviour; 44: pages 569-81 ↩ ↩² ↩³

5. MacKenzie DW (2013) Applying the Anderson-Darling test to suicide clusters: evidence of contagion at U.S. universities? Crisis; 34: pages 434-7 ↩

6. Cheng Q, Chen F, Yip P (2011) The foxconn suicides and their media prominence: is the Werther effect applicable in China? BMC Public Health; 11: page 841 ↩

7. Mesoudi A. (2009) The cultural dynamics of copycat suicide. PLoS ONE; 4: e7252 ↩

8. Gould MS, Petrie K, Kleinman MH, Wallenstein S. (1994) Clustering of attempted suicide: New Zealand national data. International Journal of Epidemiology; 23: 1185-9 ↩

9. Benson, R, Rigby, J, Brunsdon, C, Cully, G, Too, L.S and Arensman, E, (2022) Quantitative methods to detect suicide and self-harm clusters: a systematic review. International journal of environmental research and public health, 19(9), page 5313 ↩

10. Too LS, Pirkis J, Milner A, Spittall MJ (2017) Clusters of suicides and suicide attempts: detection, proximity and correlates. Epidemiology and Psychiatric Science; 26: pges 491-500 ↩

11. Hacker K, Collins J, Gross-Young L, Almeida S, Burke N (2008) Coping with youth suicide and overdose: One community’s efforts to investigate, intervene, and prevent suicide contagion. Crisis; 29: pages 86-95 ↩

12. Marchant A, Hawton K, Stewart A, Montgomery P, Singaravelu V, Lloyd K, and others (2017) A systematic review of the relationship between internet use, self-harm and suicidal behaviour in young people: The good, the bad and the unknown. PLoS One; page 12 ↩ ↩²

13. Singaravelu V, Stewart A, Adams J, Simkin S, Hawton K (2015) Information seeking on the Internet. Crisis; 36: pages 211-19 ↩

14. Sisask M, Värnik A. (2012) Media roles in suicide prevention: A systematic review. International Journal of Environmental Research and Public Health; 9: 123-38 ↩

15. Pitman A, Osborn D, King M, Erlangsen A (2014) Effects of suicide bereavement on mental health and suicide risk. Lancet Psychiatry; 1: pages 86-94 ↩

16. Cerel J, McIntosh JL, Neimeyer RA, Maple M, Marshall D. (2014) The continuum of “survivorship”: definitional issues in the aftermath of suicide. Suicide and Life Threatening Behaviour; 44: 591-600 ↩

17. Marzano L, Fraser L, Scally M, Farley S, Hawton K (2018) News coverage of suicidal behaviour in the United Kingdom and the Republic of Ireland. Crisis; 27: pages 1-11 ↩