Copyright © Des Johnson 1997

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 the prior written permission of the copyright holders. Permission will usually be given for reproduction of graphs and tables if appropriately referenced. This publication is intended for use by researchers for backup or support of their own data and may not be utilised or presented in the original format.
Send mail to Des Johnson

A copy of this report is downloadable in Portable Document format   Click here to obtain a copy (or right mouse click and save target as)

You will need the appropriate version of the Adobe Acrobat Reader program to view the file. The program is freely available if you do not already have it.

Seclusion has been the focus of much political, professional, clinical, ethical and moral debate and argument in recent years. A complex and often emotive process, its use has been advocated by those who consider it as either therapeutic or necessary in the control of the violent and the disturbed, and opposed by those who consider it a relic of the past, potentially punitive, and lacking in therapeutic benefits.

Its use within Special Hospitals has been particularly criticised in light of several damning public inquiries focusing upon the deaths in seclusion of (mainly) Afro-Caribbean patients in the 1980’s, and in light of the closed and insular cultures traditionally associated with such institutions.

This study arose in part from the continuing efforts made in recent years by the Special Hospitals to improve standards in the practice of seclusion. Its aims were to identify factors considered in the literature to be important in the decision to use seclusion and to identify to what degree these were considered important in the decision to continue or discontinue such regimes within a U.K. Special Hospital.

The study was able to identify particular themes within the literature that were associated with the decision to use seclusion. The study suggests that these themes appear to hold varying degrees of importance in the decision to continue or discontinue seclusion, and has allowed for the rating and weighting of particular factors in this decision making process.

The main findings of the study were consistent with views supported in the literature, identifying the most important factors as being the threat or presence of violence and other fear inducing behaviours. However there were some factors, considered in the literature as being of potential importance, that were held to be of little importance by the respondents in the study.

These were largely external factors such as the age, experience, attitudes and levels of staff; environmental issues, racial considerations, and cultural and organisational concerns. Potential reasons for these factors not supporting the evidence available in the literature are discussed, from which potential areas for future research are forwarded.

ACKNOWLEDGEMENTS

For Supervision:
Dr Tom Mason, Ph.D., BSc (Hons), RMN, RNMH, RGN.
Senior Research Fellow, University of Wales.

For Guidance and Support:
Dr Phil Woods, Ph.D., Dip. Health Care Research, RMN, EN(M), ENB830.
Lecturer, University of Manchester

CONTENTS

2.2  Culture

2.4  Current approach to Seclusion

3  Literature Review

3.1  Introduction 

3.2  Defining Seclusion

3.3  Theoretical basis of seclusion 

3.3.1  Seclusion as a Therapeutic Intervention

3.3.3  Seclusion as Punishment

3.4  Factors associated with the use of seclusion

3.4.1  Staff factors

3.4.2  Diagnostic factors

3.4.3  Demographic factors

3.4.4  Environmental factors

3.5  Indications for the use of seclusion

3.5.1  Violence and the threat of violence

3.5.2  Agitation, frustrations and coping skills

3.5.3  Cognitive processes and functioning

3.5.4  Mood and affective functioning

3.5.6  Compliance and conformity

3.6  Decision Making 

3.7  Risk Assessment 

4  Aims of the Study

5  Method 

5.2  Subjects (Ss)

5.3  Data collection

5.4  Procedure

5.4.1  Literature Review

5.4.2  Piloting & identification of potential questionnaire items

5.4.3  Questionnaire Design

5.5  Reliability and Validity 

6  Results

6.1  Response & background information

6.3  Ranking of themes

7  Discussion

7.2  Relationship to previous findings

7.3  Limitations to the study

7.4  Implications for future research

7.5  Summary

8  References

Appendix 1: Questionnaire

Appendix 2: Summary analysis of questionnaire items

Appendix 3: Summary analysis of items relating to each theme

Appendix 4: Cross-Tabulation of item-combined themes

LIST OF TABLES

Table 1 Qualified psychiatric and qualified forensic psychiatric experience of respondents

Table 2 Ranking of individual themes

Table 3 Comparison of ranking of individual and item-combined themes by arithmetic mean

Table 4 Comparison of score frequencies by Gender for item-combined themes

LIST OF FIGURES

Figure 1 Gender frequencies (%) of nursing staff and respondents

Figure 2 Age of respondents

Figure 3 Frequencies (%) of questionnaire items

The confinement of an individual in a locked room from which they have no means of egress is widely regarded as one of the most restrictive practices used in modern psychiatry. Commonly known as ‘seclusion’, it is regarded by some as not only "anti-therapeutic" (Topping-Morris, 1993, p7) but also a "treatment relic of the past" (Pilette, 1978, p19) and an "embarrassing reality" (Soloff, 1979, p302). Others, however, have offered less negative views of this practice and forwarded arguments supporting its validity as both a treatment modality (Grigson, 1984; Gutheil & Tardiff, 1984; Soloff, 1987; Cotton, 1989) and as an important intervention in the control of the violent individual, the reduction of stimuli and the control of agitation (Gutheil, 1978).

Hammill (1987, p38) has questioned whether seclusion continues to be used because it benefits patients or because "it is still easier to isolate an out-of-control patient behind a locked door than deal with the underlying problem." Yet regardless of one’s own opinion; whether one considers it a necessary tool in the management of the potentially violent individual or simply an easy option when faced with potentially difficult patients, reality is that the practice of seclusion remains widely used. This, despite the advent of modern psychiatric drugs and continuing developments in psychological, nursing and medical theory and practice.

The often emotive (Leopoldt, 1985) and controversial nature of this practice has given rise to increasing discussion, literature and research within recent years and encouraged both its advocates and opponents to put pen to paper in their attempts at circulating, validating, and at times rationalising their opinion. Indeed Alty & Mason (1994) highlight how the controversial and emotive nature of the subject has on occasion given rise to personal opinions and views permeating so-called ‘empirical’ studies. Yet seclusion has become not only the subject of academic debate, but has also found itself facing increasing political, social, ethical, moral and even legal scrutiny (Mason, 1995). This is particularly so in the U.S. where the use of emergency restrictive interventions have been challenged in law (Gutheil, 1980), and in the U.K. where concerns have reached both British Courts as well as the European Court of Human Rights (Mason, 1995).

Much of this recent focus on seclusion in the U.K. has arisen from inquiries into Special Hospitals. Providing treatment under conditions of high security for those regarded as having dangerous, violent or criminal propensities, these hospitals have traditionally been regarded as closed, and archaic, and have faced criticism from the public and media alike (Kerr, 1997). Historically, reluctant to open their doors to scrutiny and inspection, the deaths in seclusion of several patients within these hospitals throughout the 1980’s and 1990’s gave rise to criticisms of seclusion practice in the inquiries and debate that subsequently followed. This increasing political, public and professional profile of seclusion helped encourage the Special Hospitals Service Authority (SHSA) to examine seclusion practice within the three Special Hospitals in England, and as the then managing authority direct resources into researching its use.

A further inquiry into complaints about Ashworth Hospital (Blom-Cooper, 1992), one of the three Special Hospitals in England, intensified the focus of attention further. Examining amongst other issues the further death of a patient in seclusion, this Inquiry went so far as to recommend a phasing out of the use of seclusion and its eventual abolition within the Special Hospital system. A position not wholly supported by the SHSA, however, who whilst applauding the principle, failed to give assurances of meeting this goal in light of current knowledge and practice (SHSA, 1993). Indeed seclusion remains a recognised and accepted practice within the U.K. Special Hospital system today.

In 1992, coinciding with the research into seclusion, the SHSA formed a steering group to review the practice within the hospitals. Their aims were to define its use, highlight alternative management strategies and to ensure practice met the standards laid down in the Mental Health Act (1983) Code of Practice (HMSO, 1990, revised 1993). This gave rise to changes in the policies and procedures governing its use within each of the hospitals; a process that has continued to develop to date.

This study has arisen, in part, as a result of this continuously developing process. Following the report of this steering group (SHSA, 1993) a new seclusion policy was formulated and piloted within Ashworth Hospital. Following from its pilot study, this policy was re-examined and is in the final stages of validation at the time of writing. Part of this final examination focused specifically upon the review process and how safeguards could be placed within the policy to ensure periods of seclusion were minimised and that patients were released from seclusion at the earliest opportunity.

To help achieve this, 1996 saw the author co-opted onto the steering group and commissioned to formulate a ‘review checklist’ to help focus the clinicians’ thoughts onto specific aspects of patient characteristics, behaviour or functioning that may be considered important in deciding whether to continue or terminate a seclusion regime. In preparing work on this ‘review checklist’ the author found that there was remarkably little study in this specific area within the literature. It was this paucity of information available that provided the motivation for this study. It was recognised that in order for this ‘review checklist’ to provide clinicians with an accurate aid, one would have to identify factors recognised in the literature as being of potential relevance or importance in the decision to continue or terminate seclusion. More specifically, to identify to what degree these factors, among others, were considered of potential importance in this decision to continue or terminate seclusion by the clinicians at Ashworth Hospital in light of the residing culture of, and practice within, Special Hospitals.

2.1  History

The U.K. Special Hospitals have a long and controversial history. Evolving from the criminal lunatic asylums of the 19^th Century they have their roots soundly rooted in the cultures and philosophies of the prison service. Indeed, today a large number of the nursing staff within these institutions remain members of the ‘Prison Officers Association’. This is perhaps not surprising given the relatedness of the users of both services and the role traditionally held upon the hospitals by the Home Office with its powers of discharge of some detained patients.

Historically, functioning as separate institutions, the Special Hospital system was formally brought together with the 1959 Mental Health Act (HMSO, 1959) and subsequently by the 1977 National Health Service Act. (HMSO, 1977). Whilst remaining separate from the National Health Service (NHS) the hospitals now became a collective service in their own right. The further commissioning of the SHSA in 1989 brought the hospitals more in line with NHS structure, administration and practice and set the agenda for the 1990’s.

At this time there are three Special Hospitals providing high security psychiatric services for England and Wales; Broadmoor, Rampton and Ashworth. Ashworth was formed in 1990 with the amalgamation of two existing Special Hospitals (Park Lane and Moss Side Hospitals) on adjacent sites in North Liverpool. A ‘state’ hospital, Carstairs, provides a similar function in Scotland. It is Ashworth Hospital that provides the setting for this study.

There is no escaping the fact that the Special Hospitals have a difficult task to perform. At times they have to provide reassurance to a concerned public and media that public safety is being maintained by the containment and control of the ‘dangerous and deranged’. However, at other times they are the recipients of public and media criticism due to the perceived persecution and unjust treatment of the ‘unfortunate and vulnerable victims’ of a harsh and uncaring system.

Although the example given highlights views at differing ends of what is a constantly changing continuum, it can be argued that historically the Special Hospitals have done little to actively help their cause. Major inquiries into conditions of service and care, allegations of abuse and neglect, and deaths in seclusion (mainly Afro-Caribbean patients) throughout the 1980’s and 1990’s have tarnished further the reputation of these large institutions and given rise to calls for their closure (Blom-Cooper, 1997). The traditional reluctance of these institutions to have aspects of their practice open to scrutiny and inspection, together with their long-standing relationship with custodial care and penal philosophies gave rise to insular practice and a deep-rooted culture based on control and containment.

2.2  Culture

Mason (1995) describes the historical developments of the Special Hospitals culture in detail. He argues that control is the theme that underpins the culture of these institutions. This may take various forms and can be seen in the control of the patient by psychiatric practice in the prescribing of drugs, the incarceration of the patient behind walls, doors, locks and bars, and the physical control of the patients by the staff. He proffers that this control is central because it is anticipated by a society that expects to be protected from what it considers to be the dangerousness of the mentally ill. The authority for this control is therefore implied in the right to detain the mentally ill and has been maintained through the use of power within the Special Hospitals. Mason further argues (p43) that with this power came elitism; the view that the Special Hospitals were indeed special. As they dealt with the most violent of offenders, and the most disturbed of the mentally ill, an internal culture developed that was built upon the principles that this power and control must be maintained. Not only was it to be maintained, but it was the role of the nursing staff to achieve this; achieved historically through restrictive practices such as seclusion.

This gave the nursing staff within these hospitals a great deal of power. This power moved the profession above the usual place of nurses in the health care hierarchy. As the main body of professionals facing the disturbed behaviour, and dealing with the violence and aggression they were able to command a great deal of say in the decision to seclude or to terminate seclusion. Even today it is nursing staff who largely implement and terminate seclusion regimes within Special Hospitals.

Mason (1995, p43) likens the construct of this Special Hospital culture to the findings of Morrison (1990) in her study of general psychiatric units in U.S.A. In a study on attitudes of new starters to U.S. psychiatric units Morrison observed the socialisation of these new starters into the existing culture, with the threat of social sanctions against those who failed to adopt the accepted behaviours and attitudes of the existing group. This can be seen in the Special Hospital culture where there was undoubtedly pressure upon staff to observe the cultural norms (Blom-Cooper, 1992). This closed culture intensified the ability of the nursing staff to control and exert power over patients. Mason (1995) highlights how digressions from acceptable behaviour by patients would give rise to recognised forms of intervention by staff; interventions that may have been determined more by the culture of the institution than by the needs of the patient or the requirements of the situation. This point can be readily understood in examining the folklore surrounding these institutions. Common folklore in at least one of the Special Hospitals has it that historically it was common practice for the nurse who initiated seclusion to decide when it should be terminated. Should this nurse then be away from the hospital through, for example, either sickness or annual leave, the patient would remain in seclusion until this nurse was back on duty; in some cases perhaps weeks later.

This oppressive culture within the Special Hospitals has been recognised and, indeed, criticised. The 1992 inquiry into complaints about Ashworth Hospital (Blom-Cooper, 1992) described a brutal and damaging regime (Dale, Rae & Tarbuck, 1995) and commented that "The culture of denigration and of devaluing patients needs to be addressed immediately…"(Blom-Cooper, 1992, Vol 1, p148). This came only a few years after an inquest jury recorded a verdict of ‘accidental death aggravated by lack of care’ on the death of an Afro-Caribbean patient in seclusion in Broadmoor Hospital (Francis, 1985).

The description of Special Hospital culture highlighted above paints a bleak picture of traditional life within a Special Hospital. The SHSA, however, responded to the criticisms of the culture in Ashworth Hospital and made active steps to elicit change through the setting-up of a post-inquiry task force. This process of eliciting change continues today through changes in management structures, contracting of management consultants, and the appointment of ward managers to focus change at grass roots level. The move closer towards the NHS has also provided the nursing staff with alternatives to their traditionally insular practice, views and opinions and has offered them the opportunity of joining alternative staff associations other than the Prison Officers Association; another small step away from the old culture.

There appears to have been a degree of success in these attempts at changing the attitudes and opinions of many of the staff within the Specials. A 1995 Health Advisory Service Report (HAS, 1995) into the services provided by Ashworth Hospital praised the hospital for undergoing change and commented that not only were areas of exemplary practice evident but also that the major dehumanising practices had been eliminated (Dale et al, 1995). It was recognised that the "structures that have been well established bastions of the hospital culture have been radically revised" (HAS, 1995, p105) and that "overall, a great deal of progress had been made in addressing some of the unhelpful aspects of the previous hospital culture" (p106).

2.4  Current approach to Seclusion

Seclusion continues to be an accepted intervention in the Special Hospital system despite the 1992 inquiry into complaints about Ashworth Hospital (Blom-Cooper, 1992) recommending its phasing out and eventual abolition. Whilst recognising that it had been overused in the past the SHSA failed to give assurances of meeting this goal in light of current knowledge and practice (SHSA, 1993). Within this policy statement on the use of seclusion in the Special Hospitals the SHSA recognised it as a medical emergency, to be used for the shortest possible period, only in response to disturbed behaviour, and in accordance with the Mental Health Act (1983) Code of Practice (HMSO, 1993).

These principles have been adopted by Ashworth following from the disbanding of the SHSA and the founding of independent Hospital Authorities at Broadmoor, Rampton and Ashworth itself. At the time of this study Ashworth Hospital Authority is currently finalising its updated seclusion policy to include these principles. Indeed it is from this process that this study has evolved.

3.1  Introduction

The literature on seclusion appears to be on the increase; a response perhaps to its increasing political, public and professional scrutiny and profile. Indeed the written word on the subject is now rich and varied and covers all aspects of seclusion practice. Yet seclusion is a complex issue. Medical, legal, ethical, historical, educational and social perspectives can cloud, pervade and influence basic arguments, premises and opinion and give rise to personal judgements, opinion, attitudes and concerns that can bias empirical data and reasoned debate. This is not surprising when one considers that seclusion is one of the most controversial practices in modern psychiatry; with psychiatry itself being one of the most controversial areas of medical practice.

As psychiatric philosophies and principles can be influenced by the current opinion and attitude towards the mentally ill within society; so too, the practice of seclusion within psychiatric practice is subject to the changing opinion of not only those within psychiatry itself, but also by public and political thought. This has been recognised by Soloff (1984) in highlighting how society’s approach towards the mentally ill can hinge upon the political, public and social philosophies and structures of the day. He argued that these philosophies can determine such issues as punishment or treatment, permissiveness or restrictiveness of care, psychological versus physical therapies and even exclusion or confinement.

In tracing the modern use of seclusion Mason (1995) highlights how the changing attitudes and approaches towards the mentally ill in the 18^th, 19^th and 20^th centuries led to a turning away from physical restraints and the confinement of the mentally ill. This being a further move away from the philosophies underpinning the asylums of the 17^th century that Foucault (1967) argued had arisen more out of a need for social control than for therapeutic intervention. It appears, therefore, that seclusion is more than a straight forward medical intervention; its use being determined not only by the current medical theory, but also by external pressures exerted by the social, cultural and political perspectives of the time. It is a practice that elicits high emotion in many, resignation in some and condemnation from others.

Several authors have reviewed the literature on seclusion in recent years including Fisher (1994) and Alty & Mason (1994). Fisher (1994) grouped his review into categories concerned with indications and contra-indications; demographic, clinical and environmental factors influencing rates of seclusion; effects upon patients and staff; and finally implementation and training. Alty & Mason (1994), however, found from their review that the literature could be placed into five broad categories. These were facts, statistics and research findings; theory and interpretations; methods and procedures; opinions, beliefs and points of view; and finally anecdotes, clinical impressions and narrations. It can be seen from both these reviews that there is a vast diversity of information available relating to seclusion.

This literature review will focus upon several of the themes highlighted above. First the theoretical basis for seclusion and factors associated with its use will be discussed to provide an overview of the current opinion and thought concerning its use in modern practice. This will be related to seclusion use within Special Hospitals where appropriate. Following this there will be discussion of those factors identified as being of particular relevance to the initiation of seclusion; factors that may prove influential in the decision to continue or terminate these regimes.

3.2  Defining Seclusion

Before one can examine, or research, aspects of a subject one has to be able to define what it is one is examining or researching. In his review of the definitions of seclusion, Mason (1992) found that whilst specific definitions varied amongst authors, among professions, and between countries, there were particular elements that held a degree of common ground in the definitions offered. From this review he identified seven components fundamental to a universal definition; place, social isolation, egress, compulsion, time, rationale and establishment. He further argued that in examining these components it can be seen that the underlying theme of control is central to the majority of definitions; "the forced confinement of patients held fast, against their own wishes" (Mason, 1995, p24).

One can readily relate these components of a definition to aspects of common seclusion practice. Common practice requires the seclusion of a patient in a designated area (place). This may be a local procedure or even a legal requirement. For example in U.K. the Mental Health Act (1983) Code of Practice (HMSO, 1993) requires that seclusion rooms be specifically designated for the purpose and continues by providing guidelines on aspects such as furnishings, lighting, heating and ventilation. This removal of the individual to a designated area usually isolates him from the social company of others, with seclusion rooms often located in areas of wards out of immediate access by the main body of patients (social isolation). One of the fundamental aspects of seclusion, and one of the most clearly obvious aspects of control, is the inability of the individual to leave the locked room of his own volition (egress). Indeed the act of secluding an individual often requires his removal from one area to another against his will, therefore giving rise to a forcing of one will over another (compulsion). Mason (1992) found only one definition recognised the element of time in the seclusion process. He argued that this was an important factor in recognising the temporary nature of the regime, and had the potential for emphasising the application of other interventions should seclusion itself not prove effective. Similarly, a rationale was not included by all authors, although the reasons for seclusion were included into the definitions used in two of the three U.K. Special Hospitals; namely for the protection of self or others. Finally he argued that the nature of the establishment may influence their definition of seclusion, although the author felt that the nature of the establishment should not necessarily be reflected in the definition itself. It can be seen from this that there is no one absolute statement that accurately defines the practice of seclusion. Whilst common components highlight the practical aspects of the process itself, the ethical, legal, and social issues can impact upon the need to rationalise and validate the practice.

This leads to the question as to how to define the practice for the purpose of this study. Given that the setting for the study is a U.K. Special Hospital it would seem sensible to utilise the definition used within the seclusion policy of that hospital. This would be readily and universally recognised by clinicians within the organisation and eliminate confusion or misunderstanding about the practice being examined. The definition to be used, therefore, is the one outlined by Ashworth Hospital Authority. This is taken from the Mental Health Act (1983) Code of Practice (HMSO, 1993) and holds seclusion to be "the supervised confinement of a patient alone in a room which may be locked for the protection of others from significant harm" (p79).

3.3  Theoretical basis of Seclusion

Argument over the clinical validity of seclusion has been debated in the literature for years. Many have advocated its value as a treatment modality (Royal college of Psychiatrists, 1982; Grigson, 1984; Gutheil & Tardiff, 1984; Soloff, 1987; Cotton, 1989) or as a valuable intervention in the control of the violent individual, the reduction of stimuli and the control of agitation (Gutheil, 1978). However others have also condemned the practice as being a violation of not only a patient’s civil liberties but also their human rights (Pilette, 1978), as being potentially psychologically damaging to the patient (Orr & Morgan, 1995), and even potentially punitive (SHSA, 1992). Mason (1995, p17) argued that even the advocates of seclusion remain aware of its potential abuse, stating that "It is interesting to note that those who could be deemed proponents of seclusion remain cautious about its implementation, maintenance and duration; and readily admit it is prone to abuse unless very closely monitored." Others still have tried merely to report on its use, identify and provide empirical data on both clinical and non-clinical aspects of its use, and to offer opinion and narration regarding the reasons for its use and potential abuse.

Mason (1993a) has suggested that the theoretical foundations of seclusion use are centred around three main themes; therapy, containment and punishment, although he continues by highlighting how these "continue to be enmeshed in a complex relationship…" (p101). Richardson (1995) appeared to mirror this view of three themes; stating that actions resulting from misdemeanours could be punitive, protective or therapeutic. And whilst not always explicitly stated, the review of the literature does appear to confirm the existence of these underlying themes; themes each with potential relevance to this study. Yet even with these three distinct themes, there is often overlap that can cause conflict and confusion when attempting to examine each independently.

3.3.1  Seclusion as a Therapeutic Intervention

In 1978 Gutheil forwarded a theoretical base for the use of seclusion based upon the three principles of preventing injury, isolating the individual from interpersonal contact and the reduction of sensory overload. He concluded that if not abused, seclusion was safe and effective. His premise was that through containment the patient would be safe from the effects of injuring others and that others would be safe from the patient’s actions. Through social isolation the patient would gain respite from the potentially intense pressures of social and therapeutic relationships and finally through reduction in stimuli the patient would gain respite from often overwhelming sensory inputs; stimuli that can be exaggerated in many mental illnesses.

Others too have also advocated its value as a treatment in a variety of psychiatric settings. Cotton (1989) argued that it may prove beneficial in the psychiatric treatment of children displaying maladaptive behaviours stating that if used as part of a behavioural programme it could help the development of more adaptive behaviours and the re-channelling of maladaptive impulses. Grigson (1984) argued that maturational needs could be addressed by eliciting patient participation in treatment programmes that included seclusion; achieved by focusing upon the "interferences in maturation that cause these patients to maintain the same impulsive and destructive behavioural patterns for most of their lives" (p137"). Even the Royal College of Psychiatrists as late as 1982 pronounced that it could be used by clinical teams as part of treatment programmes, although they failed to elaborate on specific therapeutic benefits.

Fitzgerald & Long (1973) argued that the efficacy of seclusion in the treatment of the severely manic and depressed, stating that it could "provide a feasible, humane and therapeutic method of treating uncontrolled, destructive, panic-stricken, regressive and other severely ill patients" (p59). They further argued that seclusion could help in the development of "trusting therapeutic relationships between the patient and members of the treatment team" (p60) that might not have been possible within the main ward milieu.

Yet whilst the authors highlighted above all advocate to one degree or another the clinical validity of seclusion as a treatment technique, Orr & Morgan (1995) have argued that "the concept of seclusion as a treatment has found little acceptance over the decade since the publication of the MHA 1983" (p108). A point recognised earlier by McCoy & Garritson (1983, p14) when stating that "the effectiveness of seclusion remains based on conceptualisations, not reliable research findings".

3.3.2  Seclusion as Containment

The containment principle as a valid argument for the use of seclusion has received much recognition and support in the literature. A core principle of Gutheil’s (1978) theoretical base for seclusion it continues to hold support largely due to the lack of effective alternatives in the control of the violent or disruptive patient. Walsh & Randell, (1995) have argued that advocates of seclusion as a treatment modality have failed to provide evidence that seclusion is more than merely a containing intervention or indeed that patients learn more adaptive coping strategies as a result of being secluded. Indeed Mattson & Sacks (1978, p1212) conclude that "It is clear that seclusion is not a ‘cure’, but simply a more efficient way of enabling the staff to provide continued surveillance".

Others, however, fail to accept the need for seclusion even in the containment of the violent or disruptive patient. Topping-Morris (1991, p39) stated that "Sadly, seclusion remains within the repertoire of management skills adopted by custodians and controllers" and that "thankfully the typical psychiatric nurse’s repertoire of skills extends to the adoption of therapeutic interventions as an alternative to seclusion". Walsh & Randell (1995, p28), however, appear to take a more realistic view of restrictive interventions such as seclusion and restraint when commenting that "those who have suggested elimination of these interventions have given little attention to the severely violent patient and have failed to suggest practical alternative interventions." A point highlighted by Soloff (1987, p1354) when stating that "…psychiatrists must overcome their professional disdain for physical controls, recognise and accept the limitations of psycho-dynamic understanding and pharmacological management, and develop a pragmatic balance of treatment approaches to the violent patient".

Mason (1993b) was able to focus the issue when stating that when faced with a patient in the throes of violent combat there are just four practical alternatives available; namely "seclude them, restrain them, medicate them, or pass the problem to someone else" (p59). Recognising in this that "although the containment principle appears a desperate surrender to the lack of alternative effective therapies it is grounded at a practical level in that quite clearly, something has to be done when there is an outburst of violence" (p97).

Special Hospitals unfortunately have little opportunity of ‘passing the problem to someone else’; more often than not being on the receiving end of this problem. This generally allows for the use of restraint, medication or seclusion. And whilst it has been argued by Soloff (1987) that seclusion, by maintaining patient dignity, providing a degree of freedom of movement and allowing for a reduction in stimuli can prove the least restrictive and intrusive of these alternatives, within the Special Hospitals it is now considered an intervention to be used only as a last resort. Not only is it only to be used as a last resort, but even then only in accordance with the principles highlighted in the Mental Health Act (1983) Code Of Practice (HMSO, 1993); namely that its sole use is to "contain severely disturbed behaviour which is likely to cause harm to others" (p80).

3.3.3  Seclusion as Punishment

In stating that "Madness will be punished in the asylum, even if it is innocent outside of it" Foucault (1967, p269) aspires to the notion that transgressions from not only societal, but even institutional norms will result in punitive measures against the detained mentally ill. Indeed few would argue that at times punitive interventions with the mentally ill have been rationalised as therapeutic either on a conscious or unconscious level. Alty & Mason (1994, p85) highlight this when stating that "When the object of a society is the control of humans, their behaviours and their thought processes, via psychiatry, it is perhaps not surprising that some of the controllers become aberrant". This potential abuse of power or control holds true for the use of seclusion and clouds even further the debate over whether seclusion is a therapeutic, protective or punitive intervention. This point was highlighted by Fennell (1996, p226) when stating "the problems of unravelling the punitive and the protective elements in seclusion remain as intractable as ever".

Yet punishment is an emotive word. It conjures up feelings of abuse, neglect, tyranny and persecution, all of which have been used within the literature to describe seclusion by those opposing its use. However, when associated with seclusion the term punishment appears mostly mentioned in personal narrations and recounts of personal experiences rather than as a justifiable indication for its use. Mason (1993b) highlighted how authors who do make reference to the use of seclusion as punishment tend not to dwell on the subject and provide no more than personal opinion; further highlighting how other studies had argued that seclusion is rarely used punitively.

Alty & Mason (1994) stated that seclusion can be seen as a sanction to be used against rule breakers and that the fundamental role of punishment is to control behaviour. From this it could be argued that the concept of punishment relates closely with that of discipline, and with it the concepts of conformity and control. Whilst few would sanction or condone (in the context of social retribution) the punishment of the mentally ill, it has been argued that there are occasions when punishment is used as a means of behaviour modification; used within behavioural treatment programmes as a form of operant conditioning. Indeed Tardiff (1984) has argued that punishment need not be painful and that ‘seclusionary time-out’ can be therapeutic if used for brief periods when used in behavioural programmes. He stresses, however, that this form of punishment should not be confused with social retribution.

Crichton (1997) has highlighted how there is legal justification for disciplining detained patients. He cites the House of Lords ruling in Poutney v Griffiths (1976) that ‘implicit in the power to detain is the power to control and discipline’. However no clear guidelines were given on how this discipline or control should be exercised. Others, however have indicated that seclusion has been, and continues to be used as a means of social control; to ensure compliance and conformity with ward or hospital rules and guidelines. Gostin (1986) however, recognises that it is difficult to differentiate between acceptable discipline and unacceptable punishment, and continues by stating that any discipline should be done in a therapeutic spirit and not one of revenge or punishment. However, Gentilin (1987) highlights how often the motive for disciplinary action can be uncertain, stating "there are occasions when people, in order to rationalise passive-aggressive impulses, will couch disciplinary measures in terms of the patient’s own best interest" (p14). This element of control has been related to the traditional culture of the Special Hospitals highlighted earlier in Chapter 2.2 describing how control and conformity played a large part in the maintenance of the power hierarchy of such institutions.

As with all aspects of seclusion, this area of discipline and control is complex. Crichton (1997) has argued that "the control of rule breaking goes beyond simply the need to keep order and maintain a safe environment, it is also part of the treatment objective" (p37); thus implying a therapeutic element to the control of the patient. He further argues that discipline is a more helpful description of the control mental health professionals have over patients, and states that the aim of discipline should be to encourage self-control. This issue was also highlighted by Tardiff (1989, p37) when he discussed the issue of a gradual re-integration back into ward community from seclusion. He stated that "any evidence of loss of control or lack of co-operation should result in movement back to more restrictive steps in the procedure"; highlighting, along with the concept of self-control, the importance of conformity in the seclusion process.

3.4  Factors influencing the use of Seclusion

Aside from the theoretical issues highlighted above there are a number of other factors that may influence the initiation, duration and termination of seclusion regimes. These have been widely debated in the literature and as with the theoretical and clinical aspects of seclusion highlighted above, can give rise to the same contradictory and conflicting opinion and argument. Each has specific relevance to the use of seclusion within the Special Hospitals.

3.4.1  Staff factors

There have been a number of studies looking at staff and the effects of levels, experience and attitudes upon seclusion use and abuse. Mason (1993b, p98) suggested that the duration of seclusion regimes were closely related to the "affronting of staff dignity at the time of the crisis leading to the use of seclusion." A particular issue within Special Hospitals where traditional culture has the assaulting or threatening of staff as a major transgression of acceptable behaviour. Further, Walsh & Randell (1995, p30) stated that "length of time in seclusion or restraint often appears to be dependant upon common practice rather than upon clinical criteria…"; again, an aspect of the traditional Special Hospital culture highlighted in Chapter 2.2.

A number of authors have suggested that staffing ratio’s and experience may effect seclusion rates. Mattson & Sacks (1978) argued that staffing levels, staff availability, general anxiety on the ward and staff training can effect the decision to seclude. A point supported by Morrison & Le Roux (1987) who concluded that the level of experience of the Nurse in Charge of the ward related strongly with the tendency to use seclusion. The argument was supported further by Crichton (1997, p42) when stating that seclusion rates can be effected by the "demographic characteristics of the staff" including "the setting in which they work, their experience and training." And whilst Gerlock & Solomons (1983, p53) concluded that staff attitudes towards seclusion could greatly influence "when seclusion is used, who is secluded, and for how long" due to the varying tolerance levels of staff towards disturbed behaviour, not all studies have supported this view. Schwab & Lahmeyer (1979) were able to find no correlation between either the age or experience of the nursing staff and the frequency with which seclusion was used.

A number of studies have highlighted how staff attitudes can contribute to the duration of seclusion and have questioned their motives for its use. Gair (1984) even discussed the possibility of seclusion use as a result of staff’s sadistic tendencies; although Alty & Mason (1994) found no support for this in their review of the literature on seclusion.

3.4.2  Diagnostic factors

Studies within the literature have identified that certain diagnostic categories of patient are more likely to be secluded than others. It has been reported in several studies that patients with diagnoses of psychoses had increased seclusion rates in comparison to those with diagnoses of non-psychoses (Binder, 1979; Soloff & Turner, 1981).

Karson & Bigolow (1987) looked at violence in a group of 140 volunteer psychiatric in-patients. They reported that of the 97 diagnosed as schizophrenic, 41 displayed violence; whereas of the 43 with other diagnoses only 4 became violent. From this they concluded that violence by the mentally ill occurs more in those diagnosed with schizophrenia. This finding was supported by Betemps, Somoza & Buncher (1993) in a study of seclusion and restraint over a one year period in over 80 U.S. hospitals. They found that patients with a diagnosis of schizophrenia accounted for over 65% of all incidents of seclusion and restraint. This was followed by 17.8% for those diagnosed as suffering from affective psychosis. In a further study of seclusion on a 25-bed in-patient university unit, Oldham, Russakoff & Prusnofsky (1983) found that those most likely to be secluded were those with a "manic-type illness" (p646).

However, not all studies have supported this view that patients suffering psychotic illnesses are more likely to be secluded. Phillips & Nasr (1983) recorded opposite results suggesting that psychotic patients are not necessarily more prone to violence towards others than non-psychotic patients. Fisher (1994) also found from his review of the literature finding "character disorders, manic symptoms, abnormal EEG’s and mental retardation" as factors associated with higher seclusion and restraint rates.

3.4.3  Demographic factors

Fisher (1994) in his review of the literature on seclusion has indicated that there is no clear evidence that demographic variables have influenced seclusion rates one way or another. He states that the only consistent theme is that age is positively related to the use of seclusion with the younger the patient the more likely they are to be secluded. Mason (1995) further highlights this issue, commenting that the likelihood of seclusion diminishes steadily with age. Fisher (1994) continues by stating that race and gender do not appear to be significant in the decision to seclude, although this view does not appear to be universal. Crichton (1997, p42) indicated that race and gender may prove influential in the decision to seclude when stating "other factors which may influence staff response include…. the patient’s gender and racial origin…". Indeed Noble & Rodger (1989, p389) noted that "Afro-Caribbean’s tended to be younger, more verbally aggressive, more seriously violent and more psychotic. They are often big and physically strong. These are all factors which contribute to staff apprehension and to the use of restrictive measures".

This issue of coloured patients being perceived as big, strong and violent has been noted in official inquiries and holds specific relevance to the use of seclusion within Special Hospitals. The 1993 Inquiry into the death in Broadmoor Hospital of an Afro-Caribbean patient in seclusion also reviewed the deaths in seclusion of two other Afro-Caribbean patients within Broadmoor during the 1980’s. The inquiry team titled their report "Big, Black and Dangerous?" implying through this the cultural bias’ and ignorance that can give rise to false perceptions by staff; perceptions by staff that can influence decision making within the psychiatric setting. This 1993 inquiry (Prins, 1993) held that racism within forensic psychiatric institutions arose more from the cultural ignorance of the mainly white-european staff than from overt and deliberate action. The 1992 Inquiry into complaints about Ashworth Hospital, however, found more deliberate racism; commenting that "The culture at Ashworth nurtures covert, and fosters overt, racism" (Bloom-Cooper, 1992, Vol 1, p148). Therefore whilst recognition has been made of culture change within Special Hospitals in recent years (Dale et al, 1995; HAS, 1995), one must remain aware of the potential for continued racist approaches towards minority patients within these institutions, be they conscious or unconsciously motivated, and the effects that this may have upon the use of seclusion.

This can be seen through the work of Mason (1995) who studied the use of seclusion throughout the Special Hospitals. He found that whilst the number of coloured patients secluded was proportionate to the Hospitals’ population, they tended to spend longer in seclusion. He attributed this to either malevolent racism or to the misperceptions and unconscious beliefs highlighted earlier in the view of some patients as "Big, Black and Dangerous" (Prins, 1993).

3.4.4  Environmental factors

There have been a number of studies which have looked at the physical environment and the potential effects upon the use of seclusion. These have ranged from those looking at the physical layout and issues such as crowding (Palmstierna, Huitfeldt & Wistedt, 1991), to those looking at the nature of the institutions themselves. Crichton (1997) highlighted the environment as a relevant factor in determining the staff’s response to disturbed behaviour.

Larkin, Murtach & Jones (1988) whilst working for the Mental Health Act Commission looked at incidents within a Special Hospital and concluded that "incidents occurred more frequently in the Special Hospital. These incidents were also more serious in nature and resulted in greater injury" (p226). A point Coldwell & Naismith (1989, p116) state should be expected "given that all special hospital patients have been considered to display dangerous, violent or criminal propensities, and that the majority have histories of violent behaviour". They argue from this that "a higher rate of violent behaviour might be expected than is found in less secure treatment settings".

There exists an assumption, perhaps as a result of the violent disposition of many of the Special Hospital patients, that seclusion is used more within Specials than within less secure settings. Mason (1992, p261), however, argues that this "supposition remains extant due to absence of research and published data" and that this has been "largely based on hearsay and speculation" (Mason, 1993a, p95). In fact the 1995 Health Advisory Report into services at Ashworth Hospital commented that in the years since the 1992 inquiry the use of seclusion had significantly decreased. The SHSA (1993) itself believed that seclusion had been over used in the past and subsequently committed itself to the promotion of "alternative approaches to the care and treatment of disturbed behaviour and to limit the use of seclusion to exceptional circumstances" within the Special Hospitals.

3.5 Indications for the use of Seclusion

The reasons for the use of seclusion have been well documented in the literature; literature which has identified particular aspects of patient behaviour, characteristics or areas of functioning that may contribute or hold influence in the decision to initiate seclusion. Each of these has the potential to influence the decision to initiate seclusion, and as such may hold particular relevance or importance in the decision to continue or terminate such regimes. For the purpose of this study these themes have been categorised as violence and the threat of violence; agitation, frustration and self-control; cognitive processes and functioning; mood and affective functioning; comprehension and understanding; and conformity and compliance. Each of these broad themes will be discussed separately highlighting their potential influence upon the use of seclusion. Following from this there will be a brief discussion about decision making and risk assessment in relation to the use of seclusion, and its use within Special Hospitals where appropriate.

3.5.1 Violence and threats of violence

Many studies and writings on seclusion have identified the primary reason for the use of seclusion as being the response to violence (Mattson & Sacks, 1978; Soloff & Turner, 1981; Baradell, 1985; Tardiff, 1992; Betempts et al, 1993; Mason, 1995). And whilst the use of seclusion in response to violence appears to be relatively accepted by the proponents of seclusion, others have highlighted how the threat of violence is a valid and important criteria for its use. Crichton (1997, p52) has indicated that the use of seclusion appears to relate "directly to the perception of personal threat." Mason (1995, p82) summarised this point when stating that the main assumption from the studies focusing upon the reasons for seclusion use was that "nursing staff use seclusion because they perceive themselves, or others to be under threat. Usually in the case of actual or threatened violence".

Mason (1995) examined in depth the use of seclusion within the Special Hospitals. He identified that it was used in 24.2% of cases as a result of violence towards staff, in 22.1% of cases as a result of violence against patients, in 17.9% of cases as a result of threats of violence towards staff, and in 10.5% of cases due to threats of violence towards patients. The other uses for seclusion were identified as self-harm, threats of self- harm, and violence and threats of violence towards property. From this it can be seen that the use of seclusion in response to actual violence was 46.3% as opposed to just 28.4% for threats of violence; suggesting that actual violence was the primary criteria for using seclusion. However, of interest and particular relevance to the use of seclusion within Special Hospitals is Mason’s further finding concerning the length of time spent in seclusion. This finding indicated that those secluded for violence against staff spent on average four times longer in seclusion than those secluded for violence against patients; averaging 80.8 hrs for those violent against staff compared to 19.4 hrs for those secluded due to violence against patients. It has been argued that threats of violence hold an important relationship with violence itself. Noble & Rodger (1989, p389) in their study of violence by psychiatric in-patients revealed that the behavioural characteristics of the violent group included verbal aggression and threatening behaviour. Further, in their study within a Special Hospital Larkin et al (1988, p227) concluded that "we acknowledge that there is a relationship between verbal threats and physical assault." This view supported by Kay, Wolkenfeld & Murrill (1988) who found a high association between verbal and physical aggression.

Tardiff (1992, p496) advocated that "a history of violence or other impulsive behaviour is often predictive of future violence", a view supported by Mossman (1995, p227) when stating that "violence during adolescence and having a juvenile record also makes adult violence more likely". Mossman further argues that this is in line with general findings that "past illegal behaviour or violence is perhaps the best predictor of future violence." This view, however, has been questioned by others. Bowden (1997, p36) stated that "cliché’s such as ‘the best predictor of future violence is past violence’ have found kudos more by virtue of repetition than because they possess any internal validity." Further, Taylor & Hodgins (1994) have argued that considering past violence as the best predictor of future violence may be a risk in itself; observing how noting it may promote fear whilst noting its absence may promote a false sense of security.

3.5.2 Agitation, frustration & self-control

Many studies have advocated the use of seclusion for the control of agitation (Wells, 1972; Gutheil, 1978; Plutchik, Karasu, Conte, Siegal & Jerret, 1978; Outlaw & Lowery, 1992). Agitation as a concept, however, can cover many aspects of mental state and physical behaviour. Cahill, Stuart, Laraia & Arana (1991) highlight some of its components to include motor agitation, irritability, loud and pressured speech, demanding behaviour, irritability, anxiety, euphoria, anger, lability of affect, memory impairment and disorientation. Agitation has, however, been recognised as being a potential precursor to violence (Davis, 1991) and can certainly be fear-inducing for staff.

Others have highlighted the use of seclusion to enable the restoration of self-control; to bring internal controls back into equilibrium (Mason, 1993b). Tardiff (1989, p37) discussed the re-integration of secluded patients back into the ward milieu, stating that "any evidence of loss of control or lack of co-operation should result in movement back to more restrictive steps in the procedure". This indicated the self-control concept of patient behaviour to be an important aspect of the decision to continue or terminate seclusion. McCoy and Garritson (1983) have argued that the key indicators of self-control are the ability to tolerate frustration, the ability to control impulses and a willingness to agree to contractual behaviours. The potential importance of low tolerance in increasing the likelihood of violence, particularly in the non-paranoid schizophrenic, was recognised by Tardiff (1992). Further, Cahill et al (1991) have argued that patients with borderline traits can prove dangerous as a result of a low tolerance for frustration and poor impulse control.

3.5.3 Cognitive processes and functioning

Much of the established work conducted on the prediction of violence and dangerousness has focused upon the long-term nature of this concept. There is, however, an increasing body of literature that has now concerned itself with the prediction of imminent violent behaviour and the characteristics of those likely to be violent in the immediacy. This is relevant to the use of seclusion and in particular to the assessment of whether a patient is likely to become violent if released from a seclusion regime. This concept of risk assessment is discussed in Chapter 3.6.

A number of studies and reports have indicated that active symptoms of severe mental illness increase the risk of violence (Swanson, Holzer, Ganju & Juno, 1990; Link, Andrews & Cullen, 1992; Grounds, 1995). Swanson et al (1990) and Link et al (1992) both found that those experiencing symptoms such as delusions and hallucinations were involved in violence. In a study of psychiatric in-patients Noble & Rodger (1989, p389) found that "the violent group showed a significantly higher proportion of patients with delusions and hallucinations." Tardiff (1992, p493) has also argued that schizophrenic patients were "overly represented in groups of patients who are violent toward other people just before and/or during psychiatric hospitalisation"; further stating that "the presence of psychosis should make one take threats of violence very seriously and makes the assessment of violence potential essential" (p496). Davis (1991) also highlighted how many of the studies have suggested that schizophrenics are violent mainly during acute phases of their illness and when actively experiencing delusions and hallucinations.

However, like in many of the studies concerning the mentally ill, difference of opinion is common. In their 1993 study of the relationship between delusions and ‘acting out’ behaviours, Wessely, Buchanan, Reed, Cutting, Everitt, Garety & Taylor found that "violent behaviour in response to delusions was uncommon" (p69).

3.5.4 Mood and Affective Functioning

Some authors have reported the use of seclusion as being indicated for hypomanic (Wells, 1972), manic and depressed patients (Fitzgerald & Long, 1973; Binder & McNiel, 1988; Janofsky, Spears & Neubauer, 1988; Davis, 1991; Fisher, 1994; McNiel & Binder, 1995). Tardiff (1992, p494) stated that "manics have been found to have sudden severe violent episodes." One could argue that this aspect of mood disorder relates closely to Gutheil’s (1978) theory of the use of seclusion to reduce sensory input. Indeed few would argue that manic and hypomanic patients can become hostile and violent when faced with sensory overload. McCoy and Garritson (1983) argue that removal from what can be "frustrating social interactions" (p10) can help the patient regain the ability to tolerate incoming stimuli. Further, Davis (1991, p588) highlights how "patients in an acute manic state are often physically agitated"; a state recognised as being a potential precursor to violence.

Swanson (1994) in a study of violence in the community found that the prevalence of affective disorder was three times higher in the violent subgroup than the non-violent subgroup, indicating to some degree a relationship between mood disturbance and violent behaviour. Betemps et al (1993) in their study of seclusion in over 80 U.S. hospitals also found that it was used in patients suffering from affective disorders in 17.8% of the total recorded seclusion episodes. Further, Cahill et al (1991, p242) also recognise mood disorders as a condition "frequently associated with violence", stating that "euphoria, irritability and lability in affect may indicate that a patient is having difficulty maintaining control" (p243).

3.5.5 Comprehension and understanding

Tardiff (1992) has argued that psychotic disorganisation can contribute to violence in long-stay patients, with Cahill et al (1991) highlighting how disorientation can give rise to agitation; indicating how understanding and awareness can be of potential importance in influencing patient behaviour. Studies have focused upon patient’s perceptions of the seclusion process and their feelings associated with its use, yet have often failed to attribute relevance of current perceptions and comprehension as relevant or potentially important factors in determining future short-term behaviour. Hammill (1987) has argued that patients should be informed of the criteria for their release from seclusion, yet many studies and writings have failed to focus upon the patient’s awareness and understanding of this criteria; a potentially important aspect of the assessment process when deciding to continue or terminate seclusion.

McCoy and Garritson (1983) have stated that one of the criteria for self-control is the ability to enter into contractual behaviours. This ability will be dependent upon, and influenced, by the patient’s ability to comprehend his current situation. Further, his willingness to enter into such agreements may depend upon his perceptions of the legitimacy of the seclusion regime and the behaviours explained to him as necessary for its termination. How the patient views his current situation may prove of importance in determining his future behaviour. Further still, the seclusion process may have impacted negatively upon self-esteem and status among peers and given rise to loss of face. Indeed Davis (1991) suggests that the patient’s self-concept may prove important in the potential for violence.

3.5.6 Compliance and conformity

This is perhaps the most controversial reason given for the use of seclusion. It implies its use as a means of discipline and social control rather than for the containment of disruptive behaviour or the treatment of clinical symptomatology. Chapter 2.2 highlighted how the traditional culture of the Special Hospitals was based upon power and control and how compliance and conformity were important aspects of the regime within such institutions, Further, Chapter 3.3.3. discussed how the use of seclusion as a means of ensuring discipline through conformity and compliance has been likened to punishment.

The concept of compliance and conformity is an important aspect of the seclusion process. The secluded patient is often considered as ‘out of control’; lacking internal behavioural controls that can determine appropriate social behaviour within the ward environment. McCoy and Garritson (1983) highlighted that the key indicators of self-control were the ability to tolerate frustration, the ability to control impulses and a willingness to agree to contractual behaviours; all important aspects of compliance and conformity. From this it could be argued that these concepts are a valid and useful means of assessing a patient’s level of self control. As such they may play a valuable part in helping the clinician determine when a patient has regained internal controls and is thus less likely to display the behaviours that gave rise to the seclusion regime.

3.6 Decision Making

The actual decision to initiate, or indeed continue or terminate a seclusion regime, is a complex one. Indeed, decisions are influenced by many factors, some consciously known, some unconsciously motivated. An expanse of literature on decision making has given rise to theory generation and attempts at analysing its components in a number of areas and situations; including that of decision making within organisations. However, whilst not focusing upon some of the more complex theories of this process, there are some clear principles that can be seen to have potential influence in the decisions concerning seclusion; especially in Special Hospitals.

Whilst it could be argued that the patient’s best interests would be served by an assessment and subsequent decision based purely on clinical grounds, in reality there are a multitude of concerns and bias’ that can influence the decision making process when dealing with seclusion. These can include cultural expectations, norms and pressures; personal opinion, morals and ethics; concerns for policies, procedures and organisational requirements, sanctions or retributions.

Alty & Mason (1994) highlight how the cultural and organisational expectations and frameworks can affect decision making and can influence individuals to make decisions within the organisational role that they may not ordinarily make in other aspects of their life. This concept that the culture of an institution can influence decision making has particular relevance in the study of seclusion within Special Hospitals. The historical nature of Special Hospital culture has been highlighted in detail in Chapter 2.2. Here it was shown that within this culture there was great pressure for individuals to conform to the expected staff norms and behaviours. With this there were clearly recognised consequences to transgressions in patient behaviour with staff encouraged through the culture of these organisations to adhere to these practices often regardless of clinical reasoning, judgement or validity. And whilst recent findings have suggested that the culture within the Special Hospitals is changing, the degree to which this has developed in the few years since the criticisms highlighted in the 1992 Ashworth Inquiry report remains to be seen.

3.7 Risk Assessment

As with the concept of decision making, the concept of risk assessment can prove as equally complex and confusing. Currently high profile within both the literature and professional circles this concept has been taken on board by many organisations concerned with the prediction of future dangerousness of its patient population. However, whilst much of the work in this area has focused upon the long-term future dangerousness of psychiatric patients, a number of the principles of risk assessment are relevant to the use of seclusion, and in particular in the decision to continue or terminate such regimes. After all, given the uses of seclusion highlighted earlier, it could be considered that all decisions concerning its use, and potential abuse are in fact risk assessments.

Carson (1993) has highlighted that the assessment of risk involves consideration of three different components. These are consideration of, firstly, the outcome; secondly, the likelihood of the outcome; and thirdly, an expected timeframe. This has particular relevance to the decision to continue or terminate seclusion in that, firstly, there should be a consideration of what behaviours we do not want the patient to display in the immediacy following termination; namely violent or seriously disturbed behaviours. Secondly, there needs to be consideration of how likely this will be given what we know about the patient, their previous behaviours, the nature of their illness and the environment in which they will be returning. Finally, there needs to be some consideration of the time frame in which this process is occurring; the need to remain aware of the time scale in which expected behaviours are required before termination can be achieved.

However, this process is not as simple as it first appears. McNiel & Binder (1996, p846) highlighted how "evaluation of potential violence is inherently problematic, ie: some people evaluated as low risk will become violent and some people assessed as high risk will not become violent." Grounds (1995, p46) also highlighted the difficulties associated with risk assessment; stating that "it has to be recognised at the outset, first that there are limits to our knowledge about risk, secondly that there are limits to our ability to assess it, and thirdly that there are limits imposed upon us due to the structure and ethos of our services." Indeed, Grounds (1995) proffers that we worry too much about some patients and not enough about others.

These arguments appear to reflect the literature on long-term prediction of dangerousness which recognises that clinicians are poor at accurately predicting the long-term behaviours of patients. Holloway (1997, p36) has argued that "risk assessment is now a routine, if poorly understood element of clinical practice" Indeed even within the literature there remains conflicting opinion as to the ability of clinicians to accurately predict future violence. Otto (1994) has highlighted how prior to the ‘first generation’ work on risk assessment mental health professionals considered that they had the ability to predict violent behaviour with some accuracy, although this first generation of work clearly showed that this was not the case. Otto highlights this in citing Ennis & Litwack (1974) who, in summarising the ability of clinicians to accurately predict future violent behaviour stated that their predictions were less accurate than a toss of a coin. This continuing uncertainty about clinicians’ ability to accurately predict future behaviour will undoubtedly impact upon the decision to not only initiate, but also to continue or terminate seclusion.

4  AIMS OF THE STUDY

This study arose out of the need to formulate a checklist for clinicians within Ashworth Hospital to help focus their attention to those factors considered important when reviewing the need to continue or discontinue seclusion. It was felt, however, that there were a number of steps that would need to be taken before such a checklist was to be of value. This study aims to address the first steps in this process.

It was decided that the first step would be to identify, through the literature, what factors were recognised as being of potential relevance or importance in the decision to initiate seclusion. From this, one could identify the degree to which these factors, among others, were considered important in the decision to continue or discontinue seclusion by clinicians within the hospital; this being of specific importance in light of its culture and practices.

The aims of the study, therefore, are:

1. Through a review of the literature identify factors recognised as being associated with the decision to initiate seclusion.
   

2. To identify to what degree nurses within Ashworth Hospital recognise these factors, amongst others, as being important in the decision to continue or terminate seclusion.
   

3. To identify the weight of importance attached to each factor identified by nurses within Ashworth Hospital as being important in the decision to continue or discontinue seclusion.

5.1 Introduction

A descriptive methodology was used for this study which was based on a survey research strategy. Widely used in the social sciences, this method has been described as highly appropriate for use in health care research (Reid, 1993). Robson (1993) highlighted how survey strategies were well suited to descriptive studies where the aim was to establish or identify how many people in a given population possess a particular attitude or opinion. He also professed that survey data could be used to "explore aspects of a situation, or to seek explanation, and provide data for testing hypotheses" (p49).

5.2 Subjects (Ss)

The location for the research was Ashworth Hospital, Maghull, Liverpool, U.K.; one of three ‘Special Hospitals’ providing high security psychiatric care in England and Wales. An overview of the Special Hospital system was given earlier in Chapter 2. The study sample (n=160) was selected randomly from the total qualified psychiatric nursing population at the hospital (N=484^*). This represented a 33% random sample of the total qualified nursing population. Qualified nurses were chosen as respondents because of their traditional involvement in the seclusion process within Special Hospitals. This involvement has been discussed in detail in Chapter 2.2. All respondents remained anonymous throughout. There was no discrimination made between first and second level nursing qualifications.

5.3 Data Collection

Data was collected by way of a questionnaire (attached as Appendix 1), sent by means of anonymous postal survey. This method was chosen after careful consideration of the its potential advantages and disadvantages.

General advantages to the postal method were recognised as low cost, and being the least time consuming. However, the most important factor in choosing this collection method was that it ensured the maintenance of anonymity for respondents. It was felt that this anonymity was particularly important given the emotive and controversial nature of the subject being researched and in light of the potential reluctance of Special Hospital nurses to participate in research studies (see Chapter 2.3). This anonymity would be even more important in light of the managerial position held by the researcher and the subsequent effects that a lack of this may have upon respondents and their willingness to provide honest and accurate data.

Potential disadvantages of this approach were recognised as overall low response, or superficial responses in those who do respond; well recognised features of postal surveys (Robson, 1993). Therefore in an attempt to improve response rates, a covering letter was attached to each questionnaire explaining the reasons for the research and asking for their support, with a follow-up letter sent approximately two weeks after the questionnaire to each person.

5.4 Procedure

The procedure involved 4 distinct stages. First there was the literature review to identify the broad themes to be studied. The second stage involved the breakdown of the themes into smaller factors or items. These factors were then piloted to validate their relationship with not only the broad themes identified in the literature review, but also their potential importance in the decision to continue or terminate seclusion. The third stage involved the questionnaire design and subsequent distribution to the sample group, with the fourth and final stage concerned with the analysis of the data collected.

5.4.1 Literature review

A review of the literature highlighted seven broad themes that were considered of potential importance in the decision to continue or terminate seclusion. These themes were discussed in detail in Chapters 3.4 and 3.5 and have been identified as (i) violence and threat of violence; (ii) agitation, frustration and coping skills; (iii) cognitive processes and functioning; (iv) mood and affective functioning; (v) comprehension, understanding and perceptions; (vi) compliance and conformity; and (vii) external factors such as staffing, demographic, environmental and diagnostic variables.

5.4.2 Piloting & Identification of potential Questionnaire Items

The next step was for the researcher was to further explore these seven themes and to break them down into smaller component factors or items. These items had to relate to specific external factors, behaviours, characteristics, or aspects of patient functioning that may be of potential importance in the decision to continue or terminate seclusion. They also had to remain directly associated with the broad themes identified through the literature.

It was at this stage that a pilot study was undertaken. The pilot group (n=10) consisted of colleagues with varying experience in forensic psychiatric nursing. Their remit was twofold. Firstly it was to offer opinion as to whether the relationship between these newly identified factors or items and the broad themes remained valid. Secondly, they were to offer opinion as to the validity of each factor or item as having potential importance in the decision to continue or terminate seclusion. The particular factors or items of each theme validated by the pilot group were…

Violence & the threat of violence.
(i) physical violence, (ii) threats to display physical violence, (iii) threats to destroy property, (iv) verbal abuse and hostility, (v) provocative behaviour, (vi) aggressive posturing, (vii) generalised hostility (not directed at any specific person), (viii) focused hostility (directed at specific person or group of persons), (ix) threats of retaliatory action, (x) history of retaliatory action and acting on threats, (xi) history of behaviour escalating from verbal aggression to physical violence.

Agitation, frustration & coping skills.
(i) agitation, (ii) restlessness, (iii) tension, (iv) irritability, (v) tolerance to the approaches of staff, (vi) tolerance to the approaches of peers, (vii) ability to wait for gratification of needs.

Cognitive processes & functioning.
(i) organisation of thoughts, (ii) relevance of speech content, (iii) presence of delusions, (iv) acting on delusions, (v) presence of hallucinations, (vi) acting on hallucinations, (vii) suspiciousness, (viii) orientation to time, place, person.

Mood & affective functioning.
(i) current mood and level of arousal, (ii) incongruity of affect (out of character), (iii) range of affect (out of character), (iv) lability of affect (out of character).

Comprehension, understanding & feelings.
(i) the patient’s understanding of the behaviours expected of him to have seclusion terminated, (ii) perception of current management (ie: over restrictive, punitive etc), (iii) remorse for actions, (iv) appreciation of seriousness of actions, (v) perception of behaviour of others in incident giving rise to seclusion, (vi) whether the patient takes responsibility or blames others for the incident giving rise to seclusion.

Compliance and conformity.
(i) compliance with staff requests, (ii) acceptance of prescribed medication.

External variables.
(i) age of patient, (ii) ethnic origin of patient, (iii) physical prowess of patient, (iv) gender of patient, (v) ward atmosphere at time of review, (vi) staffing levels at time of review, (vii) duration of seclusion regime reflects severity of incident, (viii) concerns over criticisms from colleagues, (ix) concerns over criticisms from managers, advocates, solicitors etc, (x) concerns over the effect termination of seclusion may have upon ward atmosphere.

5.4.3 Questionnaire Design

Items were placed on the questionnaire at random, with each question relating to a specific external factor, behaviour, characteristic, or aspect of patient functioning. The respondents were asked to rate the potential importance of each when considering the decision to continue or terminate a patient’s seclusion regime.

It was recognised that the traditional format of a Likert scale is to have a range of potential answers ranging from strong positive to strong negative responses. However, given the already lengthy nature of the questionnaire (48 items) it was felt that to provide potentially strongly correlated negative question items and response categories would simply lengthen the process and generally give rise to less accurate information. Robson (1993) highlighted how the length of the questionnaire can affect the quality of the responses and this was born in mind at this stage.

The questionnaire, therefore, was designed as a five-point Likert scale with potential responses ranging from "1 = not important" to "5 = very important". Rather than a ‘strongly negative – neutral – strongly positive scale’, it was a ‘neutral – strongly positive scale’. For the purpose of statistical analysis a response of 3, or greater than 3, was considered to be of potential importance. Further to completing the questionnaire itself, respondents were requested (giving an ordinal rating of 1-7) to rank the broad themes identified in the literature in order of considered importance in the decision to continue or terminate seclusion.

It was recognised, however, that the final questionnaire remained lengthy and that this may impact negatively upon not only the number of staff willing to take the time to respond, but also in the quality of the responses

5.5 Reliability and Validity

An instrument can be said to be reliable if "its measures accurately reflect the true measures of the attribute under investigation" (Polit & Hungler, 1993); in essence this is the consistency with which the instrument performs (Gibbon, 1995). Given the descriptive nature of the research, reliability could not readily be established. The responses given may change over a period of time, or may even have been influenced by the frame of mind of the respondent at the time of completing the questionnaire. Polit et al (1993) defined validity as "the degree to which an instrument measures what it is supposed to be measuring". Again, given the descriptive nature of the questionnaire it has not been possible to readily establish several aspects of validity. Face and content validity were established by the literature and through the piloting of the themes underpinning the questionnaire items to colleagues.

5.6 Data Analysis

The data was analysed by descriptive, often called ‘summary’, tests. Descriptive, or summary, statistics are methods of representing important aspects of a set of data by a single number. They concern themselves with the distribution of the data, and in particular the level and the spread of this distribution.

6.1 Response and background information

Of the 160 questionnaires sent to the sample group a total of 54.38% (N=87) were returned. However, of these 87 responses a total of 8.3% (N=12) failed to complete the second part of the questionnaire (where respondents were requested to rank the identified themes in order of importance). General background information was collected on age, gender, experience (in years) as qualified psychiatric nurses and experience (in years) as qualified forensic psychiatric nurses.

Figure 1 highlights how the gender of respondents was fairly representative of the overall gender mix of qualified nursing staff within the hospital. Of a hospital wide mix of 76.6% (N=371) male to 23.4% (N=113) female staff, respondents were represented by 80.5% (N=70) male and 19.5% (N=17) female. No information was available regarding the breakdown of qualified staff into age bands prior to the undertaking of the study. However, Figure 2 highlights how 87.3% (N=76) of respondents were over 30 years of age and how of these, the breakdown into age bands of 30-34 years, 35-39 years and 40+ years had a fairly equal distribution of 31.03% (N=27), 28.73% (N=25) and 27.63% (N=24) respectively.

As with the ages of nursing staff within the hospital, the researcher was unable to ascertain the level of experience (in years) of the total population of qualified nursing staff within the hospital prior to the study. The breakdown of experience as qualified psychiatric nurses and as forensic psychiatric nurses can be seen in Table 1. It can be seen from this that by far the largest groups of respondents were those with between 10-14 years experience as both qualified psychiatric nurses (47.15%, N=41) and qualified forensic psychiatric nurses (48.35%, N=42). Also of note is that 76.8% (N=67) of the respondents had 10 years or more experience as qualified psychiatric nurses and that 66.7% (N=58) had 10 years or more experience as qualified forensic psychiatric nurses.

Table 1: Qualified psychiatric and qualified forensic psychiatric experience of respondents (yrs)

6.2 Scoring of Questionnaire Items

All 87 respondents completed the first part of the questionnaire. This required them to score 48 items on a scale of ‘1-5’ depending upon the potential importance of each item when considering the decision to continue or discontinue seclusion. A summary analysis of each item is given in Appendix 2. For the purpose of analysis a score of ‘3 or greater’ was considered to be of significant or particular importance. Scores of ‘1’ or ‘2’ were to be considered as being of less importance with respect to continuance or discontinuance criteria for seclusion.

Analysis of the scoring of the questionnaire items showed that 70.8% (N=34) of the items obtained a frequency of over 70% at a score of ‘3 or greater’, with 58.3% (N=28) of the items obtaining a frequency of over 80% at a score of ‘4 or greater’. Only 4.1% (N=2) of the items obtained a frequency of over 70% at a score of ‘5’; these being ‘displays of physical violence’ and ‘threats of physical violence’ with frequencies of 78.2% and 77% respectively.

The frequencies at the different scores of ‘3 or greater’, ‘4 or greater’ and ‘5’ can be seen in Figure 3. It can be seen from this that the frequencies for scores of ‘3 or greater’ and ‘5’ display a tendency to drop noticeably at certain points, whilst the frequencies for scores of ‘4 or greater’ show a more gradual decrease. With the frequencies for the scores of ‘3 or greater’ this is around the 60% mark with 79.1% (N=38) of the items achieving frequencies above this mark. With the frequencies for scores of ‘5’ this fall occurs after just 4.1% (N=2) of the items and sees the frequencies dropping from 77% to under 50%.

There were two points of note in observing the arithmetic means of each item. The first was that all of the items (n=11) belonging to the theme of ‘aggression, hostility and violence’ were within the 17 highest rated items, giving an aggregated arithmetic mean of 4.04. The second point of note was that all of the items (n=10) belonging to the theme of ‘external variables’ were within the 18 lowest rated items, including all 7 of the 7 lowest rated items, and giving an aggregated arithmetic mean of 1.94. The frequencies of individual items belonging to the other five themes appear to be relatively evenly dispersed.

The range of the responses can also be seen in Appendix 2. From this it can be seen that scores for the majority of the items differed greatly. It can be seen that 75% (N=36) of items achieved the maximum range of ‘4’, with 18.75% (N=9) of items achieving a range of ‘3’, and just 6.25% (N=3) of items achieving a range of ‘2’.

Of further interest when observing the range of response scores is that the two of the three items with the range of ‘2’ were ‘threats of physical violence’ and ‘displays physical violence’. Both these items had the highest and second highest arithmetic means and frequency of importance at all three levels of importance; ‘3 or greater’, ‘4 or greater’ and ‘5’. The third item with a range score of ‘2’ was ‘gender of patient’ which was the item with the lowest arithmetic mean and frequency of importance at all three levels.

Observation of individual item scores and frequencies of importance at differing levels highlighted specific interesting points. One such item was how the patient perceives the behaviour of others involved in the incident that gave rise to seclusion. This item had a frequency of 97.7% at score ‘3 or greater’, a frequency of 82.7% at score ‘4 or greater’, yet a frequency of just 5.6% at score ‘5’, indicating that whilst a large proportion of the respondents considered this factor as being fairly important, very few of them considered it to be very important. A further item of interest was concerned with the duration of seclusion not being long enough in light of the severity of the incident. This item scored a frequency of 24.1% at score ‘3 or greater’ indicating that almost a quarter of the respondents considered a time element to be of some importance in the seclusion process.

Yet further items of note were tolerance towards staff and tolerance towards peers. Tolerance towards staff scored frequencies of 95.4%, 90.8% and 43.7% at the three respective score levels of ‘3 or greater’, ‘4 or greater’ and ‘5’. However, tolerance towards peers recorded frequencies of 85.1%, 51.7% and 5.7% at the same respective score levels. This indicates that whilst the scores are fairly consistent at the ‘3 or greater’ score level, there are inconsistencies at the higher score levels.

A yet further interesting result was that one respondent scored the ethnic origin of the patient at level ‘3 or greater’, with a further respondent rating the item at a score level of ‘5’. This would indicate that whilst one respondent considered this factor as being fairly important in the decision to continue or discontinue seclusion, a further respondent believed it to be very important in this process.

6.3 Ranking of themes

The second part of the questionnaire required the respondents to rank the seven identified themes in order of importance ranging from ‘1’ as most important to ‘7’ as least important. Of the 87 respondents 91.7% (N=75) completed this part. An analysis of the arithmetic mean for each theme can be seen in Table 2. It can also be noted from Table 2 that apart for the theme of ‘aggression, hostility and violence’, the range of ratings for each of the other themes was large.

Of further note was the comparison of the arithmetic means of the individual themes with the aggregated arithmetic means of the items belonging to each theme. (A summary analysis of these ‘item-combined’ themes can be seen in Appendix 3).

Table 2: Ranking of individual themes

The results of the comparison of means can be seen in Table 3. It can be seen from this that there is a high degree of consistency between the rankings of the individual themes and the overall rankings of the ‘item-combined’ themes.

Aside from the themes of ‘cognitive processes and functioning’ and ‘compliance and conformity’, which were ranked 3^rd and 4^th respectively in the individual themes and then 4^th and 3^rd respectively in the ‘item-combined’ themes, all other themes showed the same ranking.

Table 3: Comparison of ranking of individual and item-combined themes by arithmetic mean 

Cross-tabulation analysis was performed on the item-combined themes by age, gender, qualified psychiatric experience and qualified forensic psychiatric experience. The results of this at the score levels of ‘3 or greater’, ‘4 or greater’ and ‘5’ can be seen in Appendix 4.

This cross-tabulation analysis highlighted several points of note. The frequency scores of male and female respondents appeared consistent for each item-combined theme at each of the three analysed score levels of ‘3 or greater’, ‘4 or greater’ and ‘5’. This can be seen in Table 4.

Further, this analysis showed that at the score level of ‘3 or greater’ there was little spread in the frequencies (%) of each item-combined theme by any of the cross-tabulated factors of age, gender, qualified psychiatric experience or qualified forensic psychiatric experience. The only exception for this was for external variables by respondents in the age range of 18-24 years. However given that only 2.3% (N=2) of the respondents were in this age range the frequency scores could not be held to be significant.