Hostname: page-component-cb9f654ff-65tv2 Total loading time: 0 Render date: 2025-08-01T02:40:55.381Z Has data issue: false hasContentIssue false

Liaison between psychiatrists and police siege negotiation teams

Published online by Cambridge University Press:  27 June 2025

Harry G. Kennedy*
Affiliation:
A consultant forensic psychiatrist and Professor of Forensic Psychiatry, Academic Department of Psychiatry, Trinity College Dublin, Ireland, Department of Clinical Medicine, Aarhus University, Denmark, and Interdisciplinary Department of Medicine, University of Bari Aldo Moro, Italy. He established the support and training for police siege negotiators in Ireland. His research interests include forensic psychopathology, therapeutic security, structured professional judgement and judgement support frameworks, and designing and evaluating models of care.
*
Correspondence Harry G. Kennedy. Email: kennedh@tcd.ie
Rights & Permissions [Opens in a new window]

Summary

Barricaded incidents, hostage-taking and sieges occur in the community, where police negotiators are usually called on to bring about a peaceful resolution. They occur also in prisons and psychiatric hospitals, where they will be managed by the institution’s staff, with police support if needed. Psychiatrists and other mental health professionals have been involved in providing training and on-call support for negotiators and decision makers in these crisis situations. This article describes definitions and goals in relation to such incidents, and outlines a five-phase framework for their management (training; first responders, preliminary interventions and inquiries; negotiations; resolution; aftercare), indicating the psychiatrist’s role during each phase. Ethical issues are also discussed.

Information

Type
Article
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2025. Published by Cambridge University Press on behalf of Royal College of Psychiatrists

LEARNING OBJECTIVES

After reading this article you will be able to:

  • understand a framework for the managed course of sieges and hostage-taking incidents in terms of goals, processes, therapeutic interventions and evaluation

  • navigate processes of engagement in five phases: preparation, preliminary inquiries, negotiation, resolution, aftercare and evaluation

  • appreciate the ethical and clinical challenges of working in a system that is not medically controlled.

Psychiatrists are involved in training and supporting specialist police teams who negotiate with those who barricade themselves into strongholds with threats, with or without hostages. The role of the psychiatrist is not to negotiate directly with the person concerned or to take decisions: that is the role of the specialist police team. The psychiatrist’s role is to provide expert advice and assistance to bring about a safe and peaceful resolution while maintaining their medical ethical boundaries. The role discussed here is that of supporting the siege negotiators, not that of the treating psychiatrist, where there is one. If the psychiatrist engaged directly with the barricader, a normal doctor–patient relationship would apply. But the psychiatrist in the role of support for hostage negotiators will be expected to consider not only the needs of the person concerned, but also the needs of hostages, first responders, neighbours and police personnel. In a hospital setting there are also competing interests for other patients and other clinicians. Consent, confidentiality, therapeutic rapport and trust would also be engaged differently.

Reviewing the psychiatry and psychology of sieges, Bahn (Reference Bahn, Bluglass and Bowden1990: p. 611) observed: ‘the common element is the individual(s) behind a barricade or barrier, refusing to come out peacefully […] The essence of the modern view is that sieges are more psychological than physical confrontations’. He went on to outline a progression in the response to sieges, from storming to containment and negotiation. Bahn’s views and approach still hold today.

Lipsedge (2004) defined hostage-taking as ‘a dramatic, time-limited public event aimed at coercing a third party’. Planned, deliberative and instrumental hostage-taking is rare and occurs in the course of interrupted organised crime such as bank robberies and for political goals. The taking of hostages most commonly occurs as an extension of the expressive, performative events described above by Lipsedge.

Scott (Reference Scott1974) described acute hostage-taking episodes in terms of primary victims, who are subjected to direct threats, and secondary victims, who are subjected to demands. The primary victim is the person taken hostage. In the case of a man who takes his child hostage, the secondary victim is the mother of the child, who is subjected to demands. People outside the stronghold are subjected to demands in almost all cases, even when there is no hostage in the stronghold.

This article is structured like a model of care, first describing the definitions and goals; then pathways and processes in five phases (training, first responders and preliminary inquiries, negotiations, resolution and aftercare), with therapeutic options integral to each phase; and finally evaluation. Ethical issues are discussed as a reflection on this unique pathway.

Definitions and goals

Domestic sieges and scripts

Subjective, cultural and moral justifications may explain the behaviour of protagonists across the range of instrumental or expressive sieges. There is often a subjective moral dimension even in the course of robberies (Mullen Reference Mullen1992) and political and grievance-driven acts. The most common ‘script’, domestic crisis confrontations and demands in which a father takes his own child hostage, is usually heavily laden with moral complaints regarding loyalty and betrayal, coercive control and dependence (Kennedy 1992; Lipsedge 2004). All three domestic siege and child hostage cases described by Kennedy & Dyer (1992) ended without injury. Substance misuse, a family history of domestic violence and fears of rejection were prominent, and the recent birth of a child was an added precipitant. Lipsedge (2004), considering the sort of cases described by Kennedy & Dyer (1992), described this as coercive communication that is more expressive than instrumental. This ‘impotent reprisal’ aimed to inflict emotional harm on the primary victim – the child’s mother – as a punishment for perceived injury. The ‘deprived’ father invoked the power of theatre to achieve his ends, prompted by culturally determined informal ‘scripting’ (Lipsedge 2004) or ‘ritual performances’ (Lipsedge 1997). There may also be some tendency towards clustering in time and place (Kennedy 1992).

Barricading and hostage-taking in prisons and psychiatric hospitals

A special concern arises regarding barricading and hostage-taking in prisons and secure psychiatric hospitals. Hughes et al (Reference Hughes, Ireland and Ireland2018) reported similarities and differences between the predisposing and triggering factors for riots and hostage-taking in prisons. Riots were driven by needs to communicate or to secure power, rights, control or freedom, and these were the basis for transactional negotiations. Hostage-taking added the needs to remove negative emotions, to inflict pain, to punish or gain revenge (a form of moral motivation), to effect a release, to manage boredom and to promote positive emotions (Hughes Reference Hughes, Ireland and Ireland2018). These are less obviously negotiable or transactional.

A Canadian prison study found that in prison sieges more than one-third of women hostages were sexually assaulted, with convicted rapists mainly responsible (Mailloux Reference Mailloux and Serin2003). Völlm et al (Reference Völlm, Bickle and Gibbon2013) found four hostage-taking incidents in one high-security hospital over a 25-year period. The hostage victims were all staff. None of the incidents were driven by acute symptoms of mental illness; all incidents were planned; three patients had a history of hostage-taking; two incidents involved a weapon but none resulted in serious injury. Remarkably, none resulted in prosecution. In another study from a high secure hospital, all hostage takers and barricaders had a history of such behaviours, with personality factors distinguishing them from other patients (Ireland Reference Ireland, Halpin and Ireland2015).

Psychiatric support for police negotiators

Early documented examples of psychiatric engagement with police negotiators in crisis situations in the UK include the Spaghetti House siege of September 1975 in London, at which a consultant forensic psychiatrist advised police negotiators to allow time to pass while building rapport with the barricaders and led to a successful resolution (Manwaring-White Reference Manwaring-White1983), the Balcombe Street siege of December 1975, also in London, in which the same sort of psychiatric and psychological advice and tactics again led to successful outcomes (Waddington Reference Waddington1991), and the Iranian Embassy siege in 1980 in London, in which negotiation was superseded and a dramatic tactical intervention was considered necessary after a hostage was killed (Lyons Reference Lyons1980).

These high-profile sieges involved organised gangs and hostages, but the great majority of sieges involve a single person who barricades and is mainly at risk of suicide. In Ireland in April 2000 a man with bipolar affective disorder discharged a shotgun when a police vehicle attended his house, leading to a 24 h siege before he exited his stronghold and was fatally shot by the police. The subsequent Barr Tribunal in 2006 recommended mental health training for all police recruits, with an emphasis on de-escalation, specially trained police negotiator units and the continuously available support of mental health professionals (Barr Reference Barr2006; Garda Inspectorate 2007). Similar recommendations have been made in many other places, with a more recent shift towards providing mental health-based crisis services rather than police-based services (Watson Reference Watson and Compton2019).

Box 1 outlines the most common types of incident involving barricading or hostage-taking.

BOX 1 Types of siege or hostage-taking incident

Community

  • Barricaded incident, with a single person threatening suicide and harm to anyone who intervenes; may have demands

  • Barricaded incident with hostage:

    1. 1 hostage a child or a current or former intimate partner

    2. 2 hostage a current or former work colleague

    3. 3 hostage a care professional

Institutional

  • Barricaded, with single person making demands

  • Barricaded, with a group or gang, with or without hostages

Terrorist, criminal

  • Single person or group in a building (e.g. with demands or because of interrupted robbery)

  • Single person or a group moving about (mass killings)

  • Hijack of an aircraft or other vehicle (often while stationary)

Goals of psychiatric interventions in armed or barricaded incidents

The primary goal is to ensure no loss of life, injury or harm to any of those involved. Next is to minimise harm, including psychological harm to hostages and other victims. The resolution of the incident should be a planned sequence of release of hostages, safe disposal of all weapons and then the safe exit of the person or persons concerned from the stronghold into the care or custody of the siege management team. An enabling goal therefore is to engage the person concerned, ascertain their needs and wants, advocate for a peaceful resolution from the start and communicate the pathways to peaceful resolution.

The goals of the psychiatrist are to facilitate a peaceful resolution where there are mental health factors that should be known with a view to facilitating the process. This requires an awareness of professional boundaries. When providing training and when first called to a scene with an unfamiliar team, it is important to emphasise that the psychiatrist is not in charge and can only offer advice – the on-scene police commander makes decisions. The psychiatrist does not engage directly with the person in the stronghold or with hostages: that is exclusively a matter for the police negotiators.

Pathways and processes: stages of engagement for the psychiatrist

Phase 1: Training, preparation and operational procedures

The specialist police units concerned in the management of sieges and hostage-taking incidents invest heavily in training. Negotiating skills are the primary emphasis, although only within a larger structure of training for tiered command, scenario planning and tactical support. The negotiators are usually experienced police officers, often volunteering as a first step, then selected for empathic and listening skills and further trained in negotiating within the structure of specialist crisis scenario planning from a policing perspective. An important part of the role of the psychiatrist is to be involved in the regular training exercises and education sessions with the police negotiating team prior to any incident.

Training of negotiators should include an awareness of simple screening questions for mental health problems (McKinnon Reference McKinnon and Grubin2013) (Box 2).

BOX 2 Mental health screening questions

  • Have you ever had any contact with mental health services?

  • Have you ever tried to harm yourself? (Have you any scars?)

  • Are you taking any medication for anxiety or depression or any similar problem?

  • Have you had any problems with alcohol or other drugs? Are you currently using any drugs?

(Adapted from McKinnon & Grubin Reference McKinnon and Grubin2013)

An on-call rota of trained psychiatrists or mental health professionals is usually available to the coordinators of the police negotiating team. This rota should not overlap with the ordinary on-call rota for a hospital and community service, since the call-outs may be very prolonged and may be outside the normal catchment area.

Indemnity cover should be ensured by the employing agency – in the UK, this is usually a National Health Service (NHS) trust or board. A written standard operational procedure or protocol may be required at this stage, as part of the forensic mental health model of care (Scott 2020; Kennedy 2022).

Phase 2a: First responder level

The first responder in the community may or may not be a member of the police, a member of a community mental health team, or – in a hospital – a nurse, healthcare assistant or psychiatrist. The first responder may also be at risk of being taken hostage, so training and induction are essential for all working in such settings.

Most such incidents conclude very quickly, resolved by a first responder. There is a body of recommendations about how best to engage at this stage (Noblett Reference Noblett and Ikin1997; Chandley Reference Chandley2001), mainly emphasising de-escalation skills and active listening.

Box 3 outlines the role of a first responder to an incident.

BOX 3 Role of a first responder at a barricading or hostage-taking incident

Goals of the first responders should be:

  • violent or dangerous behaviour should stop

  • hostages should be released

  • weapons should be thrown out of reach, followed by

  • peacefully exiting the stronghold.

First responder’s processes:

  • the first responder should maintain their own personal safety at all times

  • they should maintain a safe distance and, in a barricaded incident, they should not enter the stronghold

  • they should raise the alarm, if possible engaging others to obtain help (in the community help will come from the police; in secure institutions and hospitals there may be an operational procedure for local ‘alarm’ response and escalation to external assistance):

    1. 1 the first responder should ascertain ‘who, what, where, when’ – in particular, whether there are hostages and if so whether they are in imminent danger; asking and observing for weapons is also essential

    2. 2 a factual account should be passed on to others, including police responders.

First responder’s interventions:

  • the first responder should interact with the person concerned, aiming to de-escalate the situation

  • they should open a dialogue, actively listening for the concerns or demands of the person concerned

  • they should use first names for barricader, hostages and themselves, to humanise and build rapport

  • make no promises other than a foreseeable resolution

  • they should try to make contact with any hostage or others also in the stronghold.

Phase 2b: Preliminary interventions and inquiries

Where early communications and negotiations do not bring an incident to a close, the next stage is for a mobile team response. In a secure institution or hospital, this immediate response to an alarm will be from other care and security professionals in the area, and then from the alarm response team in accordance with policies and standard operational procedures and training (Davoren Reference Davoren, Turner, Davoren and Kennedy2024). The agenda for opening a dialogue and aiming to achieve goals is the same.

Local police crisis team intervention

Incidents that do not resolve quickly may trigger a response from the local police force proportionate to the seriousness of the danger presented to the public, to professionals and to the person concerned. The decision to call for a police intervention in a hospital or secure institution means that the hospital or prison management has given up control of the area concerned to the police. This is a decision that can be taken only by an authorised senior manager in accordance with a policy and standard operational procedure.

Whether the incident is in the community or an institution, at an early stage, a secure perimeter should be established with an inner secure boundary and outer perimeter for privacy to ensure the safety of the public. An on-scene command post will be established at the edge of this outer boundary, a hub for communications, for the negotiators and team who establish communication with the person concerned and the psychiatrist.

Calling in a specialist police negotiating team

If the police or crisis team cannot end the situation and their report to their off-scene command indicates that a serious incident is underway a specialist unit will then be deployed. The specialist police team typically includes two trained police negotiators and a negotiator-coordinator, information gatherers, a scribe, an on-scene commander, technical support and tactical support. The on-scene police commander (‘bronze’) reports to a senior police commander who is not on site (‘silver’) and a more senior commander (‘gold’) who may have responsibility for more serious or sensitive decisions.

The specialist team will work to improve communication and if necessary the means of communication, using technology such as field telephones. In a serious incident it is usually necessary to control communication with the person in the stronghold to avoid uncontrolled or unplanned communication with third parties that may be harmful. An essential goal is to prevent movement of hostages, hostage takers or single barricaded persons. The safety of hostages, hostage takers, responders (including police) and members of the public cannot be maintained if the hostage taker or an armed barricader moves from place to place.

The psychiatrist referral

The on-scene commander or coordinator of the negotiating team may call a trained psychiatrist for a number of purposes (Box 4). This will not occur in all cases but will be triggered by indicators. A referral will only arise if the incident is regarded as a serious threat to life and safety, if there are indications of a mental disorder and the incident has already lasted more than the typically short duration of such incidents. The person concerned may have given warnings, they may have a known psychiatric history, they may appear mentally disordered or have a history of grievances and resentments. The incident itself may have a level of complexity and may not be progressing through the usual processes of negotiation. The referral should be authorised by a senior officer (the ‘bronze’ on-scene commander or ‘silver’ tactical commander) and agreed by the siege negotiating coordinator.

BOX 4 Why call in a psychiatrist to a barricading or hostage-taking incident?

The psychiatrist can give advice about matters outside ordinary knowledge regarding the person’s behavioural abnormality, which might be due to their:

  • mental state

  • mental illness

  • capacity to negotiate

  • intoxication and withdrawal

  • fatigue

  • relationships and the origin of the incident

  • interaction with/reaction to team roles, relationships and communication.

In addition to advising and supporting the negotiators (Box 5), the psychiatrist will have a role in advising on the mental health and mental state of hostages as well as hostage takers or barricaders, and the extent to which the hostages or hostage takers are capable of continuing. The psychiatrist should be acting within the model of care of a formally commissioned public mental health service and a written service level agreement with the specialist police unit.

BOX 5 The advisory role of the psychiatrist in a barricading or hostage-taking incident

The psychiatrist’s role includes:

  • supporting and advising the negotiating team

  • supporting and advising for the on-scene commander

  • information gathering

  • doctor-to-doctor liaison with general practitioners and mental health services involving:

    1. 1 disclosure on a need-to-know basis where necessary and proportionate

    2. 2 aftercare planning and liaison.

The psychiatrist may also give advice on:

  • the person’s mental state and capacity to negotiate

  • psychiatric risks:

    1. 1 suicide

    2. 2 ‘suicide by cop’

    3. 3 suicide–homicide

    4. 4 homicide/trauma to hostages

  • hostage issues

  • third-party negotiators.

Note that the psychiatric expert advises – it is the on-scene commander who decides:

  • psychiatric advice is only one of the factors the on-scene commander must weigh up when making their decisions

  • the on-scene commander may have other priorities that must legitimately be considered

  • the on-scene commander will have final responsibility.

Note also that the psychiatrist never communicates directly with the person in the stronghold, because:

  • that would change the nature of the interaction to a doctor–patient relationship of a different type

  • psychiatrists have no training or experience in ‘field craft’.

The psychiatrist should commence a contemporaneous time log of all communications, advice and events from the time of first contact and should continue this through to the end of the incident.

The psychiatrist may have to elicit a brief from the coordinating negotiator or on-scene commander. At its most basic, this will include the name, age and address of the person concerned, their general practitioner and, in some cases, their psychiatrist or mental health team. If these are not already known then the psychiatrist will prompt the gathering of this information.

Information about the events leading up to the current situation may offer an understanding of how the situation arose and may suggest other informants who could be of assistance.

Following the preliminary telephone contact and briefing, there is a phase of telephone information gathering by the psychiatrist prior to or while travelling to the scene. At this early stage, preparations for aftercare may begin with local mental health services and police if necessary. Preliminary advice to the on-scene commander and negotiators may be immediately helpful – for example passing on information about relevant illnesses on a need-to-know basis. Often, this preliminary telephone advice may be all that is required as the incident ends soon afterwards.

Phase 3: Negotiations

By the time the psychiatrist arrives on scene, the possibility of early resolution will have passed. A prolonged, sometimes interminable negotiation may follow. The tone of communication from the person concerned may be angry and abusive and may be unreasonable. Demands may be made that are either practical or impractical (Scott 2018). Ultimatums may be issued. These demands and ultimatums will be passed to the on-scene commander. There may be periods of prolonged silence. These may be due to sleep or rest, which can be helpful particularly where intoxication is a factor. However, periods of silence may also be used for preparation and lead on to some new action.

The negotiation team will be familiar with the many guides to the form and content of negotiation. These include active listening, reflective talk, exploring and naming feelings, and, where suicide is a declared issue, talking openly about the finality of death. Negotiators should address hostages by name to make it harder to de-humanise them and make it harder to harm them. Some boundaries must be set explicitly and may have to be set by the negotiator as firm statements in the course of an otherwise non-judgemental interchange – for example not harming hostages and not leaving the stronghold except in an agreed and safe way (Lanceley Reference Lanceley2003).

Therapeutic assessment in phase 3

The psychiatrist will spend the next period telephoning to gather more information, for example from the general practitioner, treating psychiatrist or community mental health nurse. These interviews involve complex negotiations regarding confidentiality and its limits. In an urgent and life-threatening situation, a doctor-to-doctor disclosure is possible on a need-to-know basis in the best interests of the person concerned. Information can be disclosed in the public interest where it is necessary to prevent a serious and imminent threat to public health, national security, the life of the individual or a third party, or to prevent or detect serious crime. The disclosure should be both necessary and proportionate (British Medical Association 2024). The psychiatrist should then exercise professional judgement in making disclosures to the negotiator-coordinator or on-scene commander on the same basis, where this appears necessary and proportionate for the prevention of crime, a serious and imminent threat to the life of the individual or a third party and bring about a safe and peaceful resolution (MDU 2024).

Information gathering progresses with enquiries made of family, friends, neighbours and work or school colleagues to discover the context for the incident. Records (medical, psychiatric, criminal) are another source of reliable information. The personal ‘output’ of the person concerned – notes, diaries, computer, emails, etc. – should be examined for plans or themes. Finally, observation and active listening will always be informative even in the absence of a formal psychiatric interview. This process of biological and social history taking and fact-checking (collateral information, triangulation) is a psychiatric skill and enables the persons concerned to be humanised, understood and appreciated as people. The information gathered is fed back to the negotiators and on-scene commander. For a guide to how information may be shared with the police, see Thomson (Reference Thomson, Davoren and Kennedy2024).

Boxes 69 very briefly outline the situation and personal information relating to four fictitious scenarios involving hostage-taking or sieges.

BOX 6 Vignette: domestic siege in a women’s refuge

The situation

  • Person concerned (P): male, age in his 40s, with limited literacy

  • Highly dependent on wife (Q), who herself has low self-esteem

  • Violent to wife Q

  • Q leaves, taking their children with her

  • P enters women’s refuge and barricades himself in, with Q and several other hostages

Mental health factors

  • Script (if any) taken from heroic narratives in popular media

  • Rigid, moralistic

  • Limited vocabulary

  • Needs ego-support

  • Not a good negotiator

  • Wants to be restored to his old role, with his wife tending to his needs

BOX 7 Vignette: domestic siege in the family home

The situation

  • Person concerned (P): male, aged in his 20s, has a history of panic attacks

  • Alcohol and benzodiazepine misuse and dependence

  • Panic and rage attacks when hung over or withdrawing

  • Partner Q threatens to leave

  • P takes baby hostage upstairs in family home

Mental health factors

  • Unlikely to have a script

  • Drugs and alcohol problems

  • Likely to be emotionally labile and unpredictable

  • Wants his old role back with wife tending to his needs

  • Needs to fend off feelings of despair

BOX 8 Vignette: school hostage-taking

The situation

  • Person concerned (P): male, in his late teens

  • Few friends, withdrawn, some drug and alcohol use

  • Incident starts with a specific victim hostage – P’s ex-girlfriend

  • P orders everyone out of his school, claims to have cross-bows and knives

Mental health factors

  • P has an elaborate written script

  • Preoccupied with revenge and grievances

  • Suicide highly likely, may try to provoke ‘suicide by cop’

  • Craves some form of romantic celebrity

BOX 9 Vignette: psychosis-induced single-person siege

The situation

  • Person concerned (P): male, in his 30s, lives with parents

  • History of schizophrenia, stopped taking medication 3 months ago

  • Has expelled parents from the family home

  • Has a quantity of firearms and refuses to come out

Mental health factors

  • Delusional beliefs, hallucinations

  • Concrete thought processes

  • May not believe anything he is told

  • May or may not respond well to promises of medical treatment

Capacity to negotiate

A useful focus for the mental health professional and the trained negotiator is to assess the person’s capacity to negotiate or to make competent decisions about any specific matter (Grisso 2006). Negotiators (and the psychiatrist) may need to take into account the person’s capacity to appreciate the importance and relevance of such decisions. For example a sense of a foreshortened future or a suicidal intent may make appeals to medium- or long-term best interests less attractive. It is often possible to assess ability to negotiate by making small concessions to demands, for example for cigarettes or food. Repeating these exercises builds trust and prepares for an eventual negotiated resolution plan. Rigidity, pervasive anger or mood swings will indicate little capacity for negotiation. Intoxication complicating mental illness or personality disorder will typically magnify these difficulties and is reported to be a factor in 40–90% of cases (Lanceley Reference Lanceley2003).

Therapeutic interventions in phase 3: negotiators skills and specialist responders

Negotiation itself follows a trained pattern of active listening and communicating in which the negotiator allows the person concerned to express their needs, follows up on matters raised directly or indirectly and expresses empathy to build rapport while continuously aiming to motivate the person towards the goals of peaceful resolution and the pathways to achieve that. As for earlier stages, it is important to make no promises that cannot be fulfilled. The negotiator should focus on gaining time to allow further information to be gathered, to give the person concerned the opportunity to change their intentions or to accept the terms of a resolution, and, if a resolution is agreed, to make preparations for it.

Unreasonable or impossible demands are dealt with by temporising, clarifying, engaging – neither making promises that cannot be kept nor denying the demand outright.

Supported third-party negotiators

Third-party negotiators such as family or friends may be introduced but only on the assessment that this would be helpful. Third-party negotiators should be assessed for reliability, for a positive relationship with the person concerned, for being able to accept the goal of a peacefully negotiated ending of the incident and for having a known rapport and trust with the person concerned in a positive non-judgemental way. The third party should be able and willing to accept support and direction from the police negotiators before and during the communication. They communicate over the team’s communication line, not by going into the stronghold or placing themselves in danger. The communication must be monitored by the negotiation team, who may decide to end it at any point.

Therapeutic interventions and assessments – suicide and ‘suicide by cop’

‘Suicide by cop’ may be overt or inferred. This term refers to incidents in which a person intentionally engages in life-threatening behaviour towards police officers or civilians with a lethal weapon or what appears to be a lethal weapon with the intention of provoking officers to shoot the suicidal individual in self-defence or to protect civilians. Indicators of potential ‘suicide by cop’ include refusal to negotiate, having just killed a significant other, demanding to be killed, setting a deadline, a fatalistic view of the outcome or describing plans or intentions for own death, or wanting to ‘go out in a blaze of glory’ (die) with maximum media coverage (Mohandie Reference Mohandie and Meloy2000).

In a series of ‘suicide by cop’ cases in California, 39% involved domestic violence (Hutson Reference Hutson, Anglin and Yarbrough1998). All cases started as crisis interventions, with or without barricading and a hostage present, usually a family member or intimate partner. There is evidence from the USA of racial and geographical disparity as well as increased risk of fatal shooting by police if there is evidence of mental illness (Thomas Reference Thomas, Jewell and Allen2021). An excess rate of fatal shooting cannot be taken as evidence of an excess of intended ‘suicide by cop’. This should prompt a greater emphasis on delay, negotiation and de-escalation. Although the majority of shootings by police officers in the USA occurred early in the incident, late exits from the stronghold can also appear to be attempts at ‘suicide by cop’.

Phase 4: Resolution

Police negotiators or supported third-party negotiators may persuade the person to come out of the stronghold with extensive face-saving and reassurances of support. Negotiators may persuade the person to come out and seek help for their problems. These may have been revealed in the course of the incident to be relationship problems, depression or alcohol-related dysphoria, other drug-related mental state changes or suicidal impulses. Offers of medical or psychiatric help may or may not be acceptable as part of the resolution. Advice that ‘no crime has been committed’ may also be a part of the resolution, where true.

The end stages of the negotiation will include a negotiated plan for exit from the stronghold. There will be a requirement that hostages are first released. Weapons are thrown out and are unreachable. Then the person themselves exits in a planned and agreed manner.

Phase 5: Aftercare

First aid and emergency technicians and an ambulance will usually be standing by, in case of injuries. Hostages should be offered the most immediate assistance, for both physical and mental well-being, and the psychiatrist might be involved in providing the latter.

A Mental Health Act assessment of the person (protagonist) may be carried out either at the scene or in a local place of safety if this appears likely to be relevant. The psychiatrist should have liaised with the local mental health service from the earlier stages to ensure readiness for this eventuality, and the psychiatrist should provide a handover, doctor-to-doctor. The psychiatrist who supported negotiators should not carry out the Mental Health Act assessment. This is best performed by a doctor who has not been involved in the incident but who is given a full factual handover. Reasons for this include fatigue; possible bias arising from what has gone before; a possible dual obligation to victims; and because the psychiatrist’s support for negotiators and involvement in debriefing with the negotiators and others would either have to stop or would be changed by direct patient–doctor obligations.

The protagonist is liable to be charged with any offences that have been committed. A Mental Health Act assessment and hospital admission, as a voluntary or involuntary patient, should never prevent this criminal justice process, even when the precipitating event for the incident has been a threat of suicide. A brief mental health admission may be followed by a repeat barricaded incident, with learned impunity (Opotow Reference Opotow2002; Martinez-Solares Reference Martinez-Solares, Morales and Alfaro2022). Domestic violence or coercive control (Tanha Reference Tanha, Beck and Figueredo2010; Callaghan Reference Callaghan, Alexander and Sixsmith2018; Stark Reference Stark and Hester2019) are often part of the scenario leading to the episode, and re-victimisation is likely in repeated episodes, with escalation to femicide possible (Campbell Reference Campbell, Webster and Koziol-McLain2003). The criminal justice process may provide space and opportunity for motivational and ‘cycle of change’ work to commence with family or intimate partner victims, with sensitivity for coercive control.

Diversion of those with severe mental illness from the criminal justice system to psychiatric treatment reduces rates of recidivism generally, from about 50% to 25% (Soon Reference Soon, Singh and Greenberg2024). Recidivism therefore remains common even after diversion, and where serious violence is at issue, a forensic specialist disposal is necessary. Any psychiatric admission should be at a level of therapeutic security to prevent hostage-taking or barricading in the psychiatric hospital. A court appearance with a view to an order under a forensic section of mental health legislation may be necessary. If the person is remanded in prison custody the psychiatrist should ensure a handover to the prison in-reach and court diversion mental health services.

Box 10 offers a vignette outlining the phases of managing a barricading incident, from first response to aftercare.

BOX 10 Vignette: managing a barricading incident, from first response to aftercare

The situation

  • Person concerned (P): male, in his 30s, has moved from an inner city to a remote rural area to get away from his drug suppliers

  • He is living in an isolated farm house

  • He meets police officers on the road near his home and demands they help him obtain benzodiazepines at once

  • When police officers do not comply he stabs one and flees back to his house

Phase 2a: First responders

  • A prolonged barricaded incident develops

Phase 2b: Early information gathering

  • Psychiatrist, while en route, contacts P’s general practitioner, who advises to speak to his consultant psychiatrist, who advises to speak to his community mental health nurse (CMHN): P has a long history of psychiatric treatment

  • CMHN adds that P has unstable insulin-dependent diabetes and is prone to comas when stressed or relapsing

  • Psychiatrist advises on-scene commander that any silence or sleep may be a diabetic coma and is dangerous to P

Phase 3: Negotiations at the scene

  • Briefing from an appropriate officer

  • Speak to on-scene commander

  • Stage of progress – communicating, risks, persons present in the stronghold, weapons, intoxicants, supplies, etc.

  • Listen to and acclimatise to tenor of the negotiation

  • Briefing from intelligence gatherers

  • Mother (Q) wants to be allowed to negotiate with/speak to her son:

    1. 1 P has been expressing strongly negative feelings for her

    2. 2 both parents are interviewed – mother not recommended to speak to him

    3. 3 father (R) is coached to negotiate, establishing dialogue.

Phase 4: Resolution

  • Prolonged silence and P is seen to be either asleep or comatose on sofa

  • Siege ends after a tactical intervention and P is detained with no injury

Phase 5: Aftercare

  • P taken to emergency department by ambulance for treatment of diabetic hypoglycaemia

Debriefing

A joint debriefing exercise for staff should be held after all hostage-taking or barricading episodes. This has operational benefits to identify learning points for future training and revision of standard operational procedures. Debriefing should aim to foster resilience, anticipate and guard against moral injury, positively connote the contributions of all team members and advise against dysfunctional coping.

Evaluation

Continuous evaluation is neccessary to ensure that the process of five phases is being implemented with fidelity to the model and is effective in achieving the overarching goals. This should include a review of the logic model – the relation between resources allocated and outcomes achieved from a psychiatric point of view. A complete series of barricading incidents involving the police in Ireland found 73 incidents over a 6-year period (Garda Inspectorate 2007). Of these, 63% resolved within 4 h, a further 21% resolved in 4–9 h and only 7% went on for 24 h or more. Most (73%) began outside normal working hours. Longer incidents were the more serious. Of the 73 incidents, 29% involved hostages, 22% of the protagonists had mental health problems, 16% involved firearms and 21% involved other weapons (knives, crossbows, hoax explosives). A third-party negotiator was used in 41%. In more recent times these incidents have become more common, with approximately 20% of call-outs for specialist police negotiators also involving an on-call psychiatrist, amounting to psychiatric support once or twice a month, for a population of 5 m.

Conclusions

Areas of psychiatric practice such as hostage-taking and barricading are easily overlooked. Responsible and reliable research is difficult to do, requiring careful ethical planning as well as review. The routine publishing of activity statistics aggregated for large populations or over multiple years would allow audit and service planning and preparation for research on new approaches.

On reflection, the crisis negotiation team, the tactical intervention teams and the media interest surrounding such incidents represent a legally necessary intrusion into the private domestic space in the majority of such incidents, as uninvited guests. Although this intrusion may seem unwelcome, it is often a repeated part of an unrecognised script played out within a family, sometimes repeated over generations. The psychiatrist is not there to passively accept a role and a script, either from the person concerned or from the police or other agencies. We work collaboratively, within our professional boundaries. We are (uninvited) guests in the houses where these incidents occur.

Reflective practice and ethics

The psychiatrist is not in a normal doctor–patient relationship with the person concerned or with hostages. They never communicate with these individuals during an incident and no doctor-to-doctor referral has occurred; instead, the engagement has come from the police as the state’s crisis interventionists. Medical ethics always apply – a doctor is always a doctor. The psychiatrist must act in the best interests of the person concerned, but also in the best interests of hostages, other members of the public, responders (including police and clinicians) and, in a hospital setting, in the best interests of other patients present. Scrupulous care is required concerning medical confidentiality, with careful disclosures to negotiators and decision makers only on a need-to-know basis, where necessary and proportionate to prevent serious harm. The psychiatrist is not in a treating relationship with the person concerned but must nonetheless observe medical ethical rules and guidance regarding, for example, doing no harm and preventing cruel or inhumane treatment. Indeed, the involvement of the psychiatrist as a registered medical practitioner should be a guarantee of respectful rights-based practice in the crisis team, one of the purposes of the Istanbul Protocol (Office of the United Nations High Commissioner for Human Rights 2022).

Debriefing and reflective practice are essential to ensure that no harm has been caused by involvement and that no opportunity for better practice has been missed. This is particularly important when decision-making, resource deployment and the intervention of tactical teams are under police control not medical control, as is also the case in the day-to-day work of forensic psychiatrists in prisons, courts and other criminal justice settings. It is better to be there than not to be there, it is better to do some good than to stand back. Being able to navigate this ethical environment of dual duties and the role of advisor rather than decision maker with firm and assertive boundaries may be more familiar to forensic psychiatrists than to others. When working with the police, or prisons or courts we are guests in their house.

MCQs

Select the single best option for each question stem

  1. 1 Concerning hostages:

    1. a first responders are never taken hostage

    2. b one-third of female hostages in prisons are sexually assaulted

    3. c hostages seldom have psychological sequelae

    4. d hostages seldom have somatic health sequelae

    5. e hostages should not be addressed by name.

  2. 2 Hostage takers:

    1. a most commonly take a family member, intimate partner or child

    2. b kidnap for ransom or political reasons more often than for domestic reasons

    3. c are rarely intoxicated during the incident

    4. d are commonly mentally ill

    5. e make demands that are either instrumental or expressive, never both.

  3. 3 In specialist police teams:

    1. a the on-scene commander has full decision-making autonomy

    2. b the psychiatrist has a veto over any planned intervention

    3. c achieving a safe peaceful resolution is the goal

    4. d the negotiators follow through to the resolution, however long it takes

    5. e there is no risk to negotiators themselves.

  4. 4 In barricaded incidents or hostage-takings, the psychiatrist:

    1. a should engage directly with suicidal barricaders

    2. b should engage directly with hostages where possible

    3. c should be willing to undertake every task requested by the on-scene commander

    4. d should avoid training with the specialist team in order to remain independent

    5. e should be acting within the model of care of the respective public mental health service.

  5. 5 Barricaders:

    1. a often threaten suicide

    2. b are seldom intoxicated

    3. c are usually mentally ill

    4. d seldom act as a result of a recent separation or family crisis

    5. e seldom have personality traits of emotional instability or rigidity.

MCQ answers

  1. 1 b

  2. 2 a

  3. 3 c

  4. 4 e

  5. 5 a

Data availability

Data availability is not applicable to this article as no new data were created or analysed in this study.

Acknowledgement

I acknowledge and greatly appreciate the extensive learning and experience of working closely with An Garda Siochana’s siege negotiators and senior coordinators.

Funding

This research received no specific grant from any funding agency, commercial or not-for-profit sectors.

Declaration of interest

H.G.K. gave expert evidence to the Barr Tribunal (2006), provided the training and established and took part in the on-call rota for forensic psychiatric support for police negotiators in Ireland.

References

Bahn, C (1990) Sieges and their aftermath. In Principles and Practice of Forensic Psychiatry (ed Bluglass, R, Bowden, P): 611–6. Churchill Livingstone.Google Scholar
Barr, J (2006) Report of the Tribunal of Inquiry into the Facts and Circumstances Surrounding the Fatal Shooting of John Carthy at Abbeylara, Co. Longford on 20th April, 2000: Set Up Pursuant to the Tribunal of Inquiry (Evidence) Acts 1921–2002. Stationery Office, Dublin (https://archive.org/details/reportoftribunal0000trib).Google Scholar
Callaghan, JE, Alexander, JH, Sixsmith, J, et al (2018) Beyond ‘witnessing’: children’s experiences of coercive control in domestic violence and abuse. Journal of Interpersonal Violence, 33: 1551–81.CrossRefGoogle ScholarPubMed
Campbell, JC, Webster, D, Koziol-McLain, J, et al (2003) Risk factors for femicide in abusive relationships: results from a multisite case control study. American Journal of Public Health, 93: 1089–97.CrossRefGoogle ScholarPubMed
Chandley, M (2001) Before the experts arrive. Journal of Psychosocial Nursing and Mental Health Services, 39: 1220.CrossRefGoogle ScholarPubMed
Davoren, M, Turner, P (2024) Ward milieu and the management of in-patient violence: use of seclusion and other restrictive practices. In Seminars in Forensic Psychiatry (2nd edn) (eds Davoren, M, Kennedy, HG): 229–45. Cambridge University Press.CrossRefGoogle Scholar
Garda Inspectorate (2007) Review of Practices and Procedures for Barricade Incidents: Report of the Garda Síochána Inspectorate. Garda Inspectorate.Google Scholar
Grisso, T (2003) Evaluating Competencies: Forensic Assessments and Instruments (2nd edn). Kluwer Academic/Plenum Publishers.Google Scholar
Hughes, CJ, Ireland, JL, Ireland, CA (2018) Examining the functions of prison critical incidents: a preliminary qualitative analysis of public reporting. Journal of Criminological Research, Policy and Practice, 4: 101–10.CrossRefGoogle Scholar
Hutson, HR, Anglin, D, Yarbrough, J, et al (1998) Suicide by cop. Annals of Emergency Medicine, 32: 665–9.CrossRefGoogle ScholarPubMed
Ireland, CA, Halpin, L, Ireland, JL (2015) Exploring individual factors associated with critical incidents in a secure psychiatric setting: a preliminary study. Psychiatry, Psychology and Law, 22: 259–72.CrossRefGoogle Scholar
Kennedy, HG, Dyer, DE (1992) Parental hostage takers. British Journal of Psychiatry, 160: 410–2.CrossRefGoogle ScholarPubMed
Kennedy, HG (2022) Models of care in forensic psychiatry. BJPsych Advances, 28: 4659.CrossRefGoogle Scholar
Lanceley, F (2003) On-Scene Guide for Crisis Negotiators. CRC Press.Google Scholar
Lipsedge, M, Littlewood, R (1997) Psychopathology and its public sources: from a provisional typology to a dramaturgy of domestic sieges. Anthropology & Medicine, 4: 2543.CrossRefGoogle Scholar
Lipsedge, M (2004) Hostage-taking and domestic sieges. Psychiatry, 3: 24–6.CrossRefGoogle Scholar
Lyons, H (1980) Accident and emergency: a hospital under siege. Health and Social Service Journal, 90: 699701.Google ScholarPubMed
Mailloux, DL, Serin, RC (2003) Sexual assaults during hostage takings and forcible confinements: implications for practice. Sexual Abuse, 15: 161–70.CrossRefGoogle ScholarPubMed
Manwaring-White, S (1983) The Policing Revolution: Police Technology, Democracy and Liberty in Britain. Harvester Press.Google Scholar
Martinez-Solares, V, Morales, A, Alfaro, AM, et al (2022) Mexico Case Study: Gender-Based Violence and Impunity for Femicide and Transfemicide. US Agency for International Development.Google Scholar
McKinnon, IG, Grubin, D (2013) Health screening of people in police custody – evaluation of current police screening procedures in London, UK. European Journal of Public Health, 23: 399405.CrossRefGoogle ScholarPubMed
Medical Defence Union (MDU) (2024) Confidentiality and Disclosure in the Republic of Ireland. MDU (https://www.themdu.com/guidance-and-advice/guides/confidentiality-and-disclosure-in-the-republic-of-ireland?Region=Ireland).Google Scholar
Mohandie, K, Meloy, JR (2000) Clinical and forensic indicators of ‘suicide by cop’. Journal of Forensic Sciences, 45: 384–9.CrossRefGoogle ScholarPubMed
Mullen, PE (1992) Psychopathy: a developmental disorder of ethical action. Criminal Behaviour and Mental Health, 2: 234–44.CrossRefGoogle Scholar
Noblett, S, Ikin, P (1997) Hostage incidents: first on the scene training for healthcare professionals. Psychiatric Care, 4: 279–82.Google Scholar
Office of the United Nations High Commissioner for Human Rights (2022) Istanbul Protocol: Manual on the Effective Investigation and Documentation of Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment. United Nations (https://www.ohchr.org/sites/default/files/documents/publications/2022-06-29/Istanbul-Protocol_Rev2_EN.pdf).Google Scholar
Opotow, S (2002) Psychology of impunity and injustice: implications for social reconciliation. In Post-Conflict Justice (ed MC Bassiouni): 201–16. Brill Nijhoff.Google Scholar
Scott, PD (1974) Battered wives. British Journal of Psychiatry, 125: 433–41.CrossRefGoogle ScholarPubMed
Scott, R, Shanahan, R (2018) Man Haron Monis and the Sydney Lindt café siege – not a terrorist attack. Psychiatry, Psychology and Law, 25: 839901.CrossRefGoogle Scholar
Scott, R (2020) The Sydney Lindt café siege: the role of the consultant psychiatrist. Australian & New Zealand Journal of Psychiatry, 54: 244–58.CrossRefGoogle ScholarPubMed
Soon, Y-L, Singh, S, Greenberg, D, et al (2024) Impact of mental health court diversion on reoffending: a direct comparison of diverted and undiverted groups. International Journal of Forensic Mental Health, 23: 484–98.CrossRefGoogle Scholar
Stark, E, Hester, M (2019) Coercive control: update and review. Violence Against Women, 25: 81104.CrossRefGoogle ScholarPubMed
Tanha, M, Beck, CJ, Figueredo, AJ, et al (2010) Sex differences in intimate partner violence and the use of coercive control as a motivational factor for intimate partner violence. Journal of Interpersonal Violence, 25: 1836–54.CrossRefGoogle ScholarPubMed
Thomas, MD, Jewell, NP, Allen, AM (2021) Black and unarmed: statistical interaction between age, perceived mental illness, and geographic region among males fatally shot by police using case-only design. Annals of Epidemiology, 53: 429.e3.CrossRefGoogle ScholarPubMed
Thomson, L (2024) Community forensic psychiatry including liaison with health, criminal justice and public protection agencies. In Seminars in Forensic Psychiatry (2nd edn) (eds Davoren, M, Kennedy, HG): 246–61. Cambridge University Press.Google Scholar
Völlm, BA, Bickle, A, Gibbon, S (2013) Incidents of hostage-taking in an English high-secure hospital. Journal of Forensic Psychiatry & Psychology, 24: 1630.CrossRefGoogle Scholar
Waddington, PA (1991) The Strong Arm of the Law: Armed and Public Order Policing. Oxford University Press.CrossRefGoogle Scholar
Watson, AC, Compton, MT (2019) What research on crisis intervention teams tells us and what we need to ask. Journal of the American Academy of Psychiatry and the Law, 47: 422–26.Google ScholarPubMed
Submit a response

eLetters

No eLetters have been published for this article.