Home > Care pathway for the short-term...

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

National Collaborating Centre for Nursing and Supportive Care (UK). Violence: The Short-Term Management of Disturbed/Violent Behaviour in In-Patient Psychiatric Settings and Emergency Departments. London: Royal College of Nursing (UK); 2005 Feb. (NICE Clinical Guidelines, No. 25.)

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.

6Care pathway for the short-term management of disturbed/violent behaviour

Flowchart Icon

Overview algorithm for the short-term management of disturbed/violent behaviour (PDF, 34K)

6.1. Introduction

This guideline considers the short-term management of disturbed/violent behaviour in adult psychiatric in-patient settings and emergency departments (for mental health assessments). It considers a number of interventions and related issues. Although separate from one another, each of the interventions and related issues described here form part of an integrated pathway of care. It is hoped that the order in which these topics are discussed will facilitate this pathway of care.

The algorithm on page 19 represents an overview of this integrated pathway of care from a starting point of predicting violence, to its prevention and if necessary to the selection of interventions for the continued management of disturbed/violent behaviour. Emphasis is placed on the importance of maintaining risk assessment and de-escalation techniques throughout the care pathway process. Also this guidance focuses on the importance of staff training and service user perspectives.

This full version of the guideline presents the methodology and results of the systematic reviews of the evidence on which the recommendations have been based, in conjunction with expert review and consensus techniques. The structure of this version of the guidance begins from the pretext that prevention is the most desirable management strategy for the short-term management of disturbed/violent behaviour. Firstly, the following areas are examined: the environment, organisation and alarms, and then prediction, which is sub-divided into three areas, namely: antecedents of disturbed/violent behaviour, warning signs and risk assessment. Since none of the interventions discussed in this guideline can be safely practiced without adequate training, the guideline then turns to staff training needs. This is followed by an examination of service user perspectives, and issues raised in relation to black and minority ethnic groups, gender and other related concerns, all of which staff need to be conversant with before employing the interventions described in this guideline. The guideline then turns to preventative psychosocial interventions for continued management of disturbed/violent behaviour: de-escalation techniques and observation, before examining the other interventions: physical intervention, seclusion and rapid tranquillisation. It then considers post-incident reviews. Finally the guideline considers special issues relating to the short-term management of disturbed/violent behaviour in emergency departments for those requiring mental health assessments only, and the issue of searching.

The following background information is offered to contextualise the issues addressed in the evidence reviews, the recommendations, and good practice points that follow.

6.1.1. Legal preface

This takes place within a multi-faceted legal framework. Compliance is a core measure of quality and good practice. For example, the management of disturbed/violent behaviour frequently involves interventions to which an individual does not – or cannot – consent. It is especially important that such interventions are in accordance with best practice.

Failure to act in accordance with the guideline may not only be a failure to act in accordance with best practice, but in some circumstances may have legal consequences. For example, any intervention required to manage disturbed behaviour must be a reasonable and proportionate response to the risk it seeks to address.

The service should ensure access to competent legal advice when required, in relation to the management of disturbed/violent behaviour.

The law provides the authority to respond to disturbed/violent behaviour in some circumstances, and it sets out considerations that are extremely important when service providers have to decide what action they may take. The contribution of the law to the management of disturbed/violent behaviour should be recognised as positive and facilitative.

All those involved in the short-term management of disturbed/violent behaviour in psychiatric in-patient settings and emergency departments should:

  • be familiar with, in particular:
    • the relevant sections of the Mental Health Act 1983 and its current Code of Practice
    • the principles underlying the common law doctrine of ‘necessity’
    • the requirements of the relevant articles of the European Convention on Human Rights, including Article 2 (right to life) and Article 3 (the right to be free from torture or inhuman or degrading treatment or punishment), Article 5 (the right to liberty and security of person save in prescribed cases) and Article 8 (the right to respect for private and family life), and the principle of ‘proportionality’
    • the Health and Safety at Work Act 1974, which place duties on both employers and employees, and applies to the risk of violence from patients and the public
    • the Management of Health and Safety at Work Regulations 1992, which places specific duties on the employer to ensure suitable arrangements for the effective planning, organisation, control, maintenance and review of health and safety (these duties include ensuring that the risk assessments are undertaken and implemented)
  • receive regular training on the legal aspects of the management of disturbed/violent behaviour
  • ensure that a comprehensive record is made of any intervention necessary to manage an individual's disturbed/violent behaviour, including full documentation of the reason for any clinical decision
  • ensure or contribute to ensuring that all aspects of the management of disturbed/violent behaviour are monitored on a regular basis, and that any consequential remedial action is drawn to the attention of those responsible for implementing it
  • be aware of the obligations owed to a service user while their disturbed/violent behaviour is being managed, and of parallel obligations to other service users affected by the disturbed/violent behaviour, to members of staff, and to any visitors
  • ensure or contribute to ensuring that any service user who has exhibited disturbed/violent behaviour should not be the subject of punitive action by those charged with providing them with care and treatment, and that where the disturbed/violent behaviour is thought to warrant criminal sanction, it is drawn to the attention of the proper authority.

6.2. Prevention

6.2.1. Environment and alarm systems

Environmental factors are believed to be important determinants of disturbed/violent behaviour in psychiatric in-patient settings. A therapeutic environment is one that allows individuals to enjoy safety and security, privacy, dignity, choice and independence, without compromising the clinical objectives of the service. Comfort, noise control, light, colour and access to space will all have an impact on the well-being of both staff and service users. However, to date there has been very little research conducted to ascertain how the environment affects staff and service users of in-patient psychiatric settings.

The little existing research in this area has suggested that high traffic areas in in-patient units are the location of the largest number of assaults. Several studies have indicated that the highest proportion of assaults occur in either the day room/communal room or in the corridors (Carmel 1989; Coldwell and Naismith 1989; Lanza et al. 1993; Rosenthal et al. 1992), suggesting that assault frequency is related to either a chance encounter or that crowding (service user population density) is a significant factor. Studies of temporal variation show that most assaults occur during mealtimes and afternoons and increase in frequency until late evening (Carmel 1989; Lanza et al. 1993; Manfredini et al. 2001).

Recent national guidance documents have highlighted the need for in-patient psychiatric settings to not only be safe and secure for staff and service users, but further have recommended that the quality of design and finish should also be a prime consideration. Indeed, recent audit reports have indicated that many UK psychiatric in-patient facilities have failed to meet basic standards for a decent working or residential care environment and these wards are rated by staff, service users and visitors as noisy, hot, smelly and dirty (College Research Unit 2000; 2001).

Alarm systems are also an essential environmental safety feature in psychiatric in-patient settings. The report Violence and aggression to staff in health services, outlines three types of alarm system:

Panic buttons

Panic button systems are hardwired systems operated by strategically placed buttons installed throughout the area where a threat exists. When they are activated, an audible or visual alarm is triggered on a monitoring console. […] panic buttons may be useful in treatment and consulting rooms, where their location is known only to members of staff (Health & Safety Commission 1997, p21).

Personal alarms

Personal alarms may be of the simple ‘shriek’ type or may form part of more complex systems. […] They are most effective in situations where other people may hear them and can respond (Health & Safety Commission 1997, p21).

More complex personal alarms

More complex systems may be suitable in particularly high-risk areas. They include personal alarms linked to fixed detection systems by infra red or radio systems (Health & Safety Commission 1997, p23).

The RCPsych guideline suggested that personal and institutional alarms and communication devices are a useful means of pre-empting disturbed/violent behaviour and of protecting staff when instances of disturbed/violent behaviour arise. However, there is a paucity of research in this area.

6.2.2. Prediction: antecedents, warning signs and risk assessment

While most service users in psychiatric in-patient settings are not disturbed/violent, a small minority place health care professional and other service users at serious risk of assault. Therefore the prediction of short-term disturbed/violent behaviour is not an outcome that is measured for its own sake, but is part of a risk management plan that works towards minimising disturbed/violent behaviour and aggression, allowing both service users and staff to feel safe. As a consequence, risk assessment must be seen as an essential intervention, possibly the single most important intervention, in the therapeutic management of disturbed/violent behaviour. Worryingly, a survey conducted in 1999 by the Standing Nursing and Midwifery Advisory Committee (SNMAC) found that risk management, which should logically follow from risk assessment, is poorly defined and practice is highly variable (SNMAC, 1999). Furthermore, they found that risk assessment was not regarded as an essential aspect of clinical practice (SNMAC, 1999). While nothing can ever be predicted with 100 per cent accuracy, prediction of short-term disturbed/violent behaviour and risk assessment is integral to the management of disturbed/violent behaviour in psychiatric in-patient healthcare settings. The recent UKCC – now the Nursing and Midwifery Council (NMC) – report stresses:

While it is absolutely clear that violence is often unpredictable, the use of comprehensive risk assessment materials, followed by a properly developed plan is an absolute pre-requisite for the recognition, prevention, and therapeutic management of violence (The recognition, prevention and therapeutic management of violence in mental health care (2002) London: United Kingdom Central Council for Nursing, Midwifery and Mental Health Visiting, p15, p22).

Much of the research pre-1995 (the cut-off point for the majority of the searches underlying the original RCPsych guideline) suggested that risk factors of short-term psychiatric in-patient disturbed/violent behaviour can be identified. Key risk factors appear to include a history of disturbed/violent behaviour, young age and number of admissions. However, Stein (1998) argues that the real challenge is not their identification, but in how they should be combined and weighted. He states:

The prediction of […] harm to others is a complex and unreliable synthesis of observed past behaviour (both inside and outside of hospital […]). The key predictors are well understood but there is much less agreement about how they should be weighed […]

Therefore the issue that faces mental health care professionals is how the best predictive validity can be attained. Three main approaches have been adopted:

  • the clinical approach (‘first generation’)
  • the actuarial approach (‘second generation’), which includes actuarial tools or checklists
  • structured clinical judgement (‘third generation’).

Most of the literature prior to 1995 suggests that clinical judgement has poor positive predictive validity of around 33 per cent (Doyle and Dolan 2002). Therefore a ‘second generation’ of risk assessment studies adopted actuarial measures, in an attempt to raise the positive predictive validity of short-term psychiatric in-patient disturbed/violent behaviour. This actuarial approach depends on ‘assessors reaching judgements based on statistical information according to fixed and explicit rules’ (Doyle and Dolan 2002). Actuarial checklists have been created to enhance this process. Both the use of checklists and this general approach have been suggested to improve predictive validity (Doyle and Dolan 2002). However, there are noticeable disadvantages to this approach, in particular the tendency it generates to focus on static factors, such as history of disturbed/violent behaviour, demographic information and diagnosis, without taking individual service user needs into consideration.

Most recently, it has been suggested that prediction needs to be carefully slotted into a more holistic approach, which places emphasis on the empirical or static factors isolated by the actuarial approach, whilst combining it with clinicians' judgements. This ‘third generation’ approach, described by Doyle and Dolan (2002) as ‘structured clinical judgement’ has the advantage of placing emphasis on the service user as an individual and allowing risk to be seen as a moving rather than static entity, so that stage of disease, and any fluctuations in personal and environment factors are taken into consideration. Such an approach seems to mirror the objectives of the UKCC report, where it states that:

The assessment of risk is an essential part of the care and treatment of all patients. It is most important to stress that risk levels change. Therefore, […] the nature and level of risk should be subject to regular review (The recognition, prevention and therapeutic management of violence in mental health care (2002) London: United Kingdom Central Council for Nursing, Midwifery and Mental Health Visiting, p15, p22).

6.3. Training

There are currently no formal regulations governing training for the short-term management of disturbed/violent behaviour in the UK. There are more than 700 training providers in the UK. The David Bennett Inquiry (2004) recommended that a national approach to training should be set up in the next year. The National Institute for Mental Health in England (NIMHE) is currently mapping the various training packages on offer in the UK and, in conjunction with the NHS Security and Management Service (SMS), is drawing up a core training curriculum for the UK and setting up an accreditation scheme for trainers.

At present, very few of the training programmes are based on evidence of either the effectiveness of training or the benefits perceived by staff and/or service users. As Leadbetters and Perkin (2002) states:

The assumption that training is the key element in reducing risk and increasing safety is common […] Such simplistic populist assumptions support quick-fix organisational solutions […] and are challenged by conclusions from emergent research across the human services (Leadbetter and Perkins 2002, p20-21).

As training is expensive, it is necessary that services are able to measure its benefits. Without such an evidence base, there is a danger that training that is beneficial and possibly life-saving will not be sought or offered.

6.4. Working with service users

6.4.1. Service user perspectives

In recent years a great deal has been written within guidance material on the need to involve service users in their care. One of the guiding principles of the National Service Framework (NSF) on mental health is to involve service users and their carers in the planning and delivery of care (Mental Health National Service Framework 1999). This principle is echoed by the Department of Health, which argues that:

In order to create a genuinely patient-centred service several processes should be created to enable users to contribute to the design and delivery of care. The aim is to promote a non-judgemental, non-patronising, collaborative approach to care (Department of Health, Mental health policy implementation guide 2002, p14).

The UKCC has laid out a number of principles that they believe need to be met in order to fulfil such aims in relation to adult service users in psychiatric in-patient settings. It argues that:

  • The prevention and management of disturbed/violent behaviour should primarily be viewed as an occupational problem, requiring a cohesive, multi-faceted organisational approach. The safety and homeliness of clinical areas, the quality of life in clinical areas and the nature of staff interventions with patients and the assessment of the needs of patients and their clinical management are at least as important in this regard as training in and use of any specific intervention strategies. The importance of these factors needs to be recognised and emphasised in training and practice (The recognition, prevention and therapeutic management of violence in mental health care 2002).
  • Service users, their advocates, and their carers should be involved in reviews of policies, and their contribution to the planning and provision of training should be seen as essential. The inquiry into the death of David Bennett highlighted once more the need to consider race, culture, and ethnicity in all areas of policy, practice and training. The input by service users, advocates and carers noted above must be incorporated into these perspectives (The recognition, prevention and therapeutic management of violence in mental health care 2002).

6.4.2. Minority ethnic groups

The David Bennett Inquiry (2004) highlighted the importance of considering the needs of black and minority ethnic groups when managing disturbed/violent behaviour in the short-term. For the purpose of this guideline, the following definition of minority ethnic group has been adopted:

Minority ethnic group: A group which is numerically inferior to the rest of the population in a state, and in a non-dominant position, whose members possess ethnic, religious or linguistic characteristics which differ from those of the rest of the population and who, if only implicitly, maintain a sense of solidarity towards preserving their culture, traditions, religion or language. (F. Capotorti (1985) ‘Minorities’, in Bernhardt R et al. (editors) Encyclopedia of public international law. Amsterdam: Elsevier, vol.8, p.385.)

The importance of this area is widely recognised by health care professionals (Fernando 1998) and has recently been highlighted by a number of high profile inquiries. The most recent of which is the inquiry into the death of David ‘Rocky’ Bennett, an African Caribbean service user who died whilst being restrained on a secure unit.

The literature, around mental health and minority ethnic groups, highlights particular concerns relating to black and African Caribbean service users. For the purpose of this guideline the following definition of black, taken from They look after their own, don't they? (DH/Social Service Inspectorate 1998), has been adopted:

Black: refers to those members of the ethnic minority groups who are differentiated by their skin colour or physical appearance, and may therefore feel some solidarity with one another by reason of past or current experience, but who may have many different cultural traditions and values.

For this purpose of this guideline, the following definition of African Caribbean has been adopted:

Of or pertaining to both Africa and the Caribbean; used to designate the culture, way of life, etc or the characteristic style of music of those people of black African descent who are, or whose immediate forebears were, inhabitants of the Caribbean (West Indies) (Oxford English Dictionary Online).

It is maintained that black and particularly African Caribbean service users are over-represented within the mental health services in the UK, particularly in forensic settings. A variety of reasons have been advocated, including:

It is also suggested that recent shifts in Government policy have led to a more punitive approach within mental health services, particularly secure settings, and that young black African Caribbean men have been made to bear the burden of this altered approach (Fernando et al. 1998). Again it has been asserted that this burden reflects racial stereotyping that regards young African Caribbean men as ‘big, black and dangerous’ (Prins H, Big, black and dangerous? Report of the Committee of Inquiry into the death in Broadmoor hospital of Orville Blackwood and a Review of the deaths of two other Afro-Caribbean patients 1993). It is suggested that this stereotyping affects the treatment of African Caribbean service users within many mental health settings. (Littlewood and Lipsedge 1997).

As a result of the concerns relating to the treatment of African Caribbean service users, the review in this guideline has given particular attention to the short-term management of the disturbed/violent behaviour of African Caribbean service users in psychiatric in-patient settings. However, it has not done so to the exclusion of other ethnic groups.

6.4.3. Gender

As far as possible, gender needs must also be taken into consideration in the short-term management of disturbed/violent behaviour in psychiatric in-patient settings. For the purpose of this guideline the following definition of gender has been adopted:

Gender describes those characteristics of women and men that are socially determined, as opposed to ‘sex’, which is biologically determined. (Mainstreaming gender and women's mental health implementation guide 2003).

While general differences between men and women in terms of mental health have been recognised, (for example, women are more likely to self-harm and suffer from depression, and men more likely to experience earlier onset and more disabling courses of schizophrenia), a recent report by the Department of Health, The women and mental health strategy (2003) stresses that these differences should be used to inform our understanding of an individual, rather an obscure their individuality. A further report reinforced the message that women's mental health needs to be conducted in relation to an individual woman's experiences, beliefs and struggles, as well as her ethnic group, age and sexual preferences (Good practices in mental health 1996).

In terms of managing disturbed/violent behaviour in psychiatric in-patient settings, the main concern raised in The women and mental health strategy has been to identify gender specific needs, such as single-sex facilities, and to ensure that both male and female service users feel safe, listened to and involved in identifying and meeting gender related needs (Mainstreaming gender and women's mental health implementation guide 2003).

6.4.4. Other special concerns

This evidence review focuses specifically on disabilities, other than learning disabilities (excluded from this guideline), and aims to consider the effects of sensory impairment. It has been noted that service users with such sensory impairments are particularly vulnerable when managed using the interventions discussed in this guideline. One such example is the restraining of a deaf service user's hands, thereby preventing them from communicating.

Very little has been written on the needs of service users with a disability in relation to the short-term management of disturbed/violent behaviour in psychiatric in-patient settings.

6.5. Psychosocial intervention

6.5.1. De-escalation techniques

De-escalation (also referred to as ‘defusing’ or ‘talk-down’) involves the use of various psychosocial short-term techniques aimed at calming disruptive behaviour and preventing disturbed/violent behaviour from occurring. Every effort is made to avoid confrontation. This can include talking to the service user, often known as verbal de-escalation, moving service users to a less confrontational area, or making use of a specially designated space for de-escalation. Stevenson and Otto (1998) offer the following definition of verbal de-escalation:

What is verbal de-escalation? A nurse might describe it as “talking the patient down,” but it is actually a complex, interactive process in which a patient is redirected towards a calmer personal space.

There are competing theoretical approaches to de-escalation, including verbal de-escalation. Some approaches make use of communication theory (for example, Paterson and Leadbetter 1997), others of situational analysis (Rix 2001). All approaches emphasise the need to observe for signs and symptoms of anger and agitation, approaching the person in a calm controlled manner, giving choices and maintaining the service users dignity. Some approaches suggest mirroring the patient's mood. De-escalation techniques also emphasise the therapeutic use of the nurse's own personality and relationship with the person (use of self) as one method to interact therapeutically with the patient.

In all approaches to de-escalation, stress is laid on the need for training and self-awareness. For example, Rix (2001) comments:

Becoming competent at de-escalation is in itself a sophisticated activity requiring much more than just a theoretical understanding of aggression. It cannot be considered in purely academic terms. The practitioner must undertake a developmental process, resulting in highly evolved self-awareness enabling the skills of de-escalation to become instinctive.

However, a recent report notes that, despite the emphasis that is often placed on the importance of de-escalation, little research has been carried out into the effectiveness of any given approach, leaving nurses to contend with conflicting advice and theories:

Unfortunately, there has been little research conducted into the effectiveness of different approaches to de-escalation, or, for that matter, into the effectiveness of training in any given approach. As Paterson and Leadbetter (1999) note, there is no standard approach to de-escalation. At the same time, practitioners may be faced with contradictory advice provided in the context of differing theoretical explanations for the violent event (National Institute for Social Work Research Unit 2000, p24).

6.5.2. Observation

Although much of the research carried out on observation has been undertaken in relation to the management of suicide and self-harm, the UKCC report (Feb 2002), which focuses on the short-term management of disturbed/violent behaviour in psychiatric in-patient settings, argues that these principles form a good basis for the short-term management of disturbed/violent behaviour in psychiatric in-patient settings. The UKCC report (Feb 2002) recommends that the principles of observation found in Addressing acute concerns (1999) – a report that focuses on the management of suicide and self-harm – should be adopted nationwide.

Although the focus of the work on observation in Addressing acute concerns was on suicide and self-harm, there are obvious implications for the use of observation in recognising the possibility of violence occurring and for preventing interventions (The recognition, prevention and therapeutic management of violence in mental health care (2002) London: United Kingdom Central Council for Nursing, Midwifery and Mental Health Visiting, p24).

[…] observation (carried out as set out in Addressing acute concerns) should underpin all other strategies (The recognition, prevention and therapeutic management of violence in mental health care (2002) London: United Kingdom Central Council for Nursing, Midwifery and Mental Health Visiting, p24).

Addressing acute concerns defines observation as “‘regarding the patient attentively’ while minimising the extent to which they feel that they are under surveillance” (p2). The UKCC report (Feb 2002), regards observation as a ‘core nursing skill’ and ‘arguably a primary intervention in the recognition, prevention and therapeutic management of violence’ (The recognition, prevention and therapeutic management of violence in mental health care 2002) It suggests that observation must be a two-way relationship, established between a service user and a nurse, which is meaningful, grounded in trust, and therapeutic for the service user. This relationship is considered to be the basis on which risk assessment, violence management and a programme of supportive observation can then be undertaken (The recognition, prevention and therapeutic management of violence in mental health care 2002).

Addressing acute concerns outlines four levels of observation – general observation, intermittent observation, within eyesight, within arms length – which, with slight modification, have been adopted within this current guideline. Other reports and studies detail a variety of other terms and levels of observation. The UKCC report, The recognition, prevention and therapeutic management of violence in mental health care (2002) argues that there is a need for the terminology to be standardised, quoting the following passage from Addressing acute concerns:

Research on the nursing practice of observing patients who are at risk from self harm, or of causing harm to others, shows that there is no consistency in the definition of terms, principles or processes. In some trusts there is no written policy for observation. Trusts vary greatly in the indications for observation and in the personnel that are thought appropriate to perform it. Where policies and procedures do meet reasonable standards, they may not be implemented properly (Addressing acute concerns 1999, p15).

Whilst the UKCC report, The recognition, prevention and therapeutic management of violence in mental health care (2002) has stressed the value of observation, Addressing acute concerns suggests that both nurses and service users have found this a difficult intervention with many nurses considering it custodial and lacking in therapeutic value (Addressing acute concerns 1999).

6.6. Other interventions

6.6.1. Physical interventions

In the UK the physical intervention primarily used in the short-term management of disturbed/violent behaviour is manual holding, rather than the use of mechanical devices such as belts, body vests or handcuffs. These devices are rarely and only used in exceptional circumstances, usually within high security settings. Physical intervention is predominantly described in the literature as restraint. In this guideline, this terminology is avoided because of its association with particular techniques, associated with control and restraint (C&R) and its various modifications. C&R was originally developed in 1981 for prison staff and was taken up by the special hospitals in the mid 1980s. It is still widely used in the NHS, although modifications have been developed to make these techniques more appropriate to the therapeutic care of service users – for instance, C&R general services, which modifies uses of pain as a restraint technique (Wright 1999). Although still widely used, we believe that the association of the term ‘restraint’ with this approach is unhelpful, as a wide range of physical interventions are now currently employed, many very different from C&R or its modifications, such as MAPA, which makes use of therapeutic holding.

The use of pain compliance as a method of managing violent behaviour is controversial amongst health care professionals and service providers. Although practice currently continues in some services, the recommendations in this guideline severely restrict its use for rescue purposes only (see Section 8, para 1.8). For the purpose of this guideline, physical intervention is defined as:

A skilled hands-on method of physical restraint involving trained designated health care professionals to prevent individuals from harming themselves, endangering others or seriously compromising the therapeutic environment. Its purpose is to safely immobilise the individual concerned.

The current Code of Practice to the Mental Health Act 1983 states that physical intervention should be a last resort:

Physical restraint should be a last resort, only being used in an emergency where there appears to be a real possibility of significant harm if withheld. It must be of the minimum degree necessary to prevent harm and be reasonable in the circumstances. (18.10-18.11)

There appears to be a dearth of knowledge about current practice. The literature review undertaken for the UKCC report in 2002, found ‘no high quality studies that evaluated either the use of restraint or of seclusion in those with mental illness’ (The recognition, prevention and therapeutic management of violence in mental health care 2002) The rate of physical interventions per annum in the UK is currently unknown. At present the National Institute for Mental Health in England (NIMHE) is compiling a register of all the techniques used in the UK.

6.6.1.1. Staff injury

A significant issue relating to the use of physical interventions is the possibility of injury to staff or service users. A US study in a maximum security forensic hospital found costs incurred in relation to staff injury from violent incidents accounted for 2 per cent of the hospital budget; 45 per cent of injuries were sustained during physical interventions (Hillbrand et al. 1996).

6.6.1.2. Sudden death

Sudden death can occur when physical intervention is used, although this is a rare event. The David Bennett Inquiry drew attention to the need for a central agency to record physical intervention-related deaths in the UK. The national reporting and learning system is a non-mandatory system set up by The National Patients Safety Agency (NPSA) which records anonymised data on sudden death in in-patient settings. The confidential inquiry has also now extended its recording of homicides and suicides to cover all sudden and unexplained deaths involving mental health service users.

6.6.2. Seclusion

Seclusion is the formal placing of a service user in a specially designated room for the short-term management of disturbed/violent behaviour. While it is recognised that this intervention is unpopular with service users, it is sometimes the preferred course of action to prevent prolonged physical intervention where rapid tranquillisation is contra-indicated or when service users have indicated a preference for it in advance directives.

The RCPsych Council Report (41) argues that the definition of seclusion needs to be broad to allow for the seclusion room door being open, closed but unlocked or locked. Therefore, for the purpose of this guideline, the following definition of seclusion has been taken from the Code of Practice:

Seclusion is the supervised confinement of a patient in a room, which may be locked to protect others from significant harm. Its sole aim is to contain severely disturbed behaviour, which is likely to cause harm to others. Seclusion should be used as a last resort; for the shortest possible time. Seclusion should not be used as a punishment or threat; as part of a treatment programme; because of shortage of staff; where there is any risk of suicide or self-harm. Seclusion of an informal patient should be taken as an indicator of the need to consider formal detention.

Seclusion must be differentiated from asking a service user to go to a designated room for the purpose of calming down. The latter is a de-escalation technique and the seclusion room should not routinely be used for this purpose. Seclusion, if chosen, is not viewed as a therapeutic intervention. It simply allows for a period of calming in the service user and should always be managed in a designated room for seclusion, separating the service user from other service users and placing them in a positive milieu (Cashin 1996).

6.6.3. Rapid tranquillisation

6.6.3.1. Definitions

Rapid tranquillisation (also called urgent sedation): the use of medication to calm/lightly sedate the service user and reduce the risk to self and/or others. The aim is to achieve an optimal reduction in agitation and aggression, thereby allowing a thorough psychiatric evaluation to take place, whilst allowing comprehension and response to spoken messages throughout.

Calming: a reduction of anxiety/agitation.

Light sedation: a state of rest and reduction of psychological activity, but verbal contact is maintained.

Deep sedation: a reduction of consciousness and motor and sensory activity, where verbal contact is progressively lost.

Anaesthetised: a state of narcosis (unconsciousness), analgesia (lack of awareness of pain) and muscle relaxation. It is one stage beyond deep sedation. It implies loss of airway control and protective reflexes, and requires the constant attention of trained personnel to keep the patient safe. There is normally no verbal contact.

Sleep: a condition of body and mind such as that which normally recurs for several hours every night, in which the nervous system is inactive, the eyes closed, the postural muscles relaxed, and consciousness practically suspended.

Of all these terms, sleep is the one with the greatest terminological inexactitude. For the purposes of this guideline we have adopted this definition from the Oxford English Dictionary. However, because of its inexactitude, we have generally avoided using this term.

6.6.3.2. Rapid tranquillisation

Rapid tranquillisation – or urgent sedation (Broadstock 2001) as it is sometimes called – is used in situations requiring the rapid control of agitation, aggression or excitement. In the UK, deep sedation/sleep is not considered a desirable endpoint for rapid tranquillisation. A state of calm is preferred, with the service user remaining conscious where possible.

For the purposes of this guideline, rapid tranquillisation describes the use of medication to control severe mental and behavioural disturbance, including aggression associated with the mental illness of schizophrenia, mania and other psychiatric conditions. It is used when other less coercive techniques of calming a service user, such as verbal de-escalation or intensive nursing techniques, have failed. It usually involves the administration of medication over a time-limited period of 30-60 minutes, in order to produce a state of calm/light sedation. Other medication regimes would be administered over longer periods of time and not time limited.

Rapid tranquillisation differs from rapid neuroleptisation, which is the practice of giving a high dose antipsychotic at the beginning of ongoing treatment with the aim of rapidly stabilising symptoms. Rapid neuroleptisation has been found to be hazardous and no more effective than standard treatment (Royal College of Psychiatrists 1997).

The medications used for rapid tranquillisation should ideally have a low level of side effects and rapid onset of action. At present, there is no worldwide formal agreement on which drugs should be used as first line for rapid tranquillisation. As a consequence, there is a wide variation in the type of medications used in rapid tranquillisation throughout the world. This has been compounded by changes in the stated aims of rapid tranquillisation over recent years – that is to calm rather than put to sleep (Cunnane 1994; Pereira et al. 2003).

There is also little agreement about the doses to be used. Rapid tranquillisation is not a recognised clinical procedure in the British National Formulary (BNF). Although the use of high dose antipsychotics has been criticised by several inquiries (Royal College of Psychiatrists 1995), expert clinician opinion may from time to time support prescribing outside the dose limits set by the BNF or SPC (RCPsych draft report on antipsychotic drugs). The BNF has been formally consulted in the preparation of these guidelines and will carefully consider the findings to decide whether to incorporate any of the recommendations into its guidance at a future date, following the publication of this guideline.

This lack of standardisation also reflects the fact that very few randomised controlled trials have been conducted that examine the efficacy of medicines used for the purpose of rapid tranquillisation. Their use is often based purely on clinical experience. Overall there is a lack of high quality clinical trial evidence surrounding the drugs used for rapid tranquillisation and their safety, a point which has been noted in a number of recently conducted systematic reviews (Cure and Carpenter 2002; Carpenter 2002; Carpenter and Berk 2002). Clinical trials that examine the effectiveness and safety of drugs used for rapid tranquillisation encounter a number of ethical issues. Service users recruited into these clinical trials should ideally represent those with highly agitated states in circumstances similar to those encountered in normal clinical practice. Unfortunately such service users are normally unable to give consent, due to their highly agitated states.

6.6.3.3. Route of administration

It is generally accepted that oral formulations should be offered in the first instance. If these are refused or are inappropriate, medication should be administered parenterally. This involves administration by intramuscular (i/m) injection or, in exceptional circumstances, intravenously. The latter should only be done with extreme caution and with appropriate supervision and monitoring, as clarified by the recommendations in this guideline.

6.6.3.4. Drugs used for rapid tranquillisation

The classes of drugs commonly used in the UK for rapid tranquillisation are benzodiazepines and antipsychotics.

6.6.3.4.1. Benzodiazepines

Benzodiazepines are frequently used as first line treatments for rapid tranquillisation. Some, such as diazepam, have erratic and slow absorption intramuscularly and are associated with prolonged sedation following repeated doses. Lorazepam has a shorter elimination half-life than many other benzodiazepines, which limits the risk of excessive sedation due to the cumulative effects of the drug. For this reason it is often chosen as the first drug of choice in rapid tranquillisation. There is a risk of respiratory depression when benzodiazepines are given in high doses or when used in combination with other hypnosedatives, including alcohol and some illicit drugs (Broadstock 2001).

6.6.3.4.2. Antipsychotics

Antipsychotics are commonly used as second line treatments for rapid tranquillisation and, in some cases, as first line treatments if benzodiazepines are contraindicated or have proven ineffective in the past. Older antipsychotics (commonly called conventional antipsychotics) have a greater propensity to cause extrapyramidal side effects than the newer (commonly called atypical) antipsychotics.

6.6.3.4.3. Combination of drugs

Combinations of a benzodiazepine, an antipsychotic, and other drugs may be given either deliberately or inadvertently in rapid tranquillisation. It has become common practice to co-administer both a benzodiazepine and antipsychotic together. There is no evidence of a higher incidence of adverse effects with this combination and it is considered to have advantages, such as allowing a lower dose of the antipsychotic to be given when administered with a benzodiazepine (Beer et al. 2001). It has also been noted that there are other problems with combinations such as olanzapine and lorazepam, which will addressed in the recommendations.

In pharmaceutical practice it is stated that if combinations of intramuscular (i/m) injections are used they should not be mixed together in the same syringe.

Users may also inadvertently receive combinations of drugs through poor control of PRN prescribing. The practice of routinely prescribing a wide range of drugs for PRN use, without clear guidelines or preference, may lead to users inadvertently receiving combinations of drugs.

6.6.3.4.4. High doses

Sometimes it is necessary to knowingly exceed the BNF upper dose limits and knowingly use drugs outside of their marketing authorisation (off-label). In such circumstances, clinicians are referred to the recommendations of the Royal College of Psychiatrists' consensus statement of the use of high dose antipsychotic medication. For the purpose of rapid tranquillisation, care must be taken to ensure that high doses do not accidentally occur through the use of PRN medication given in combination with regular medication. If PRN medication is given, it is important to allow time for the drug to work before giving further doses by either oral or intramuscular means. In addition, clinicians must bear in mind that the plasma concentration of the antipsychotic is not only affected by the total dose, but also the route of administration. Clinicians should also be aware that absorption from intramuscular administration (i/m) can occur far more rapidly when a service user is agitated, excited or physically overactive (Keck 1991).

6.6.3.4.5. Dangers associated with antipsychotics

There are two main areas of concern with the use of antipsychotics for rapid tranquillisation – extrapyramidal effects and cardiac effects.

Extrapyramidal side effects are mostly associated with conventional antipsychotics. Side effects such as dystonia and occulogyric crisis are very unpleasant for the service user and may adversely affect their future preparedness to access either treatment or services. Fortunately the side effects can mostly be rapidly reversed by administration of antimuscarinic drugs such as procyclidine. The availability of atypical antipsychotic drugs provides an opportunity to avoid these side effects.

The second main issue of concern relevant to rapid tranquillisation is the rare occurrence of drug induced arrhythmias and sudden cardiac death. This happens because of the manner in which some antipsychotic drugs affect cardiac ventricular repolarisation in susceptible individuals. The main measure of ventricular repolarisation is the QT interval – the time from the onset of ventricular depolarisation to complete repolarisation. A number of cardiac, metabolic and other factors, such as physical exertion and stress, impact on the QT interval. Where the service user has a prolonged QT interval, they may be at increased risk of cardiac arrhythmias, particularly torsade de pointes. The cardiac QT interval usually measured as the QTc interval (QT corrected for heart rate) is a useful if somewhat imprecise indicator of the risk of cardiac events. This prolongation can be congenital or acquired however, service users who already have prolonged QT repolarisation are at risk of developing arrhythmia when given drugs which further lengthen the QT interval. Service users who have had Torsade de Pointes are at an increased risk, even where this was caused by a different drug. Service users with left ventricular dysfunction or hypertrophy are also at an increased risk as are service users with liver disease (Day et al. 1993). Diuretics also appear to increase risk. Women who have a longer QT interval on average than men appear to be at an increased risk of Torsade de Pointes (Rautaharju et al. 1992; Makkar et al. 1993).

An issue that further complicates the relationship between antipsychotics, ventricular tachycardia and sudden cardiac death is that service users are known to be a high-risk group for cardiovascular death (Hensen et al. 1997). However, it is known that QT prolongation and resulting arrhythmias are drug concentration related (Drici et al. 1998; Warner et al. 1996; Reilly 2000; Ray et al. 2001).It is also important to note that several case reports of sudden death involved agitated service users who were subject only to physical interventions. As discussed above, physical interventions have been linked to increased risk of arrhythmia, as has the use of illicit drugs, such as ecstasy (Drake and Broadhurst 1996) and cocaine (Pereira 1997).

6.6.3.5. Acute manic or mixed episodes in bipolar affective disorder

For service users with bipolar affective disorder the British Association of Psychopharmacology (BAP) guidelines should be taken into consideration.

6.6.3.6. PRN medication

Although only rapid tranquillisation is mentioned directly in the scope, PRN medication pro re nata (as needed) medication is also sometimes used in a similar way to rapid tranquillisation in psychiatric in-patient settings. A recent editorial suggests that very little has been written on the effectiveness of PRN medication as a short-term measure for managing disturbed/violent behaviour and that those studies that do consider this issue contain serious flaws (Ray and Meador 2002).

6.6.3.7. Service user views

Service user satisfaction with rapid tranquillisation was rarely, if ever, measured as a part of the few existing clinical trials.

6.7. Emergency departments

This guideline also considers the short-term management of disturbed/violent behaviour for adults with psychiatric illness who present in emergency departments for mental health assessment, immediately prior to admission to an adult psychiatric in-patient setting.

All the interventions and related topics (excluding environment, observation and seclusion) are relevant to emergency departments. However, emergency settings sometimes have special requirements in addition to those addressed in psychiatric in-patient settings. These requirements are considered in the specific recommendations in Section 8.

Copyright © 2006, Royal College of Nursing.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means electronic,mechanical, photocopying, recording or otherwise,without prior permission of the Publishers or a licence permitting restricted copying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1T 4LP. This publication may not be lent, resold, hired out or otherwise disposed of by ways of trade in any form of binding or cover other than that in which it is published,without the prior consent of the Publishers.

Cover of Violence
Violence: The Short-Term Management of Disturbed/Violent Behaviour in In-Patient Psychiatric Settings and Emergency Departments.
NICE Clinical Guidelines, No. 25.
National Collaborating Centre for Nursing and Supportive Care (UK).

NICE (National Institute for Health and Care Excellence)

PubMed Health Blog...

read all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...