From Pathological Dependence to Healthy Independence: An emergent grounded theory of facilitatingindependent living

Liz Jamieson, Ph.D; Pamela J. Taylor, F Med Sc; Barry Gibson,


People with mental disorder are admitted to high security
hospitals because of perceived risk of serious harm to others.
Outcome studies generally focus on adverse events, especially reoffending,
reflecting public and government anxieties. There is
no theoretical model to provide a better basis for measurement.
There have been no studies examining discharge from the
perspectives of those involved in the process. This paper begins
to fill this gap by generating a grounded theory of the main
concerns of those involved in decisions to discharge from such
hospitals. Data were collected by semi-structured interviews
with staff of various clinical and non-clinical disciplines, some
with a primary duty of care to the patient, while mindful of public
safety, and some with a primary duty to the public, while mindful
of patients’ rights. The data were analysed using a grounded
theory approach. Their main concern was ‘pathological
dependence’ and that was resolved through the process of
‘facilitating independent living’. Clinicians and non-clinicians
alike managed this by ‘paving the way’ and ‘testing out’. The
former begins on hospital admission, intensifies during residency,
and lessens after discharge. Testing out overlaps, but happens to
a greater extent outside high security. Factors within the patient
and/or within the external environment could be enhancers or
barriers to movement along a dependence-independence
continuum. A barrier appearing after some progress along the
continuum and ending independence gained was called a
‘terminator’. Bad outcomes were continuing or resumed
dependency, with ‘terminators’, such as death, re-offending or readmission,
modelled as explanations rather than outcomes per se.
Good outcomes were attainment and maintenance of community
living with unconstrained choice of professional and/or social
supports. Although this work was done in relation to high
security hospital patients, it is likely that the findings will be
relevant to decision making about departure from other closed
clinical settings.

KEYWORDS: pathological dependence, independent living,
grounded theory, mentally disordered offenders, high security
(special) hospitals


Most countries have special secure healthcare facilities for
people with a major mental disorder thought to pose a serious
threat of harm to others, generally after at least one serious
criminal conviction. It is difficult, however, to compare outcome
studies between different countries because laws, policies, social
structures and service availability may each vary widely.
Facilities may be entirely within the health services, entirely
within prisons, or a mixture of the two. Not all countries provide
every level of security, and there may be international differences
in definitions of ‘high’, ‘medium’ and ‘low’ security. There is,
though, common ground in being held in such a secure institution
– in constraints to freedom and autonomy within and outside the
unit and long enforced proximity to others with grave health and
behavioural problems. In England and Wales, people with a
major mental disorder, detainable under mental health
legislation and thought to pose a high risk of serious and
imminent harm to the public, may be admitted to a high security,
or ‘special’ hospital. Median length of stay there is over six years
(Butwell, Jamieson, Leese & Taylor, 2000).

Perhaps the most common ground to date between studies
internationally and over time is in choice of outcome measures.
Studies in both the United Kingdom and North America, for
example, have focused almost exclusively on re-offending
(Jamieson & Taylor, 2004; Steadman & Keveles, 1972; Steadman
& Cocozza, 1974; Thornberry & Jacoby, 1979; Pruesse & Quinsey,
1977). There is less common ground between nations, however,
in definition of offences and base rates of crime, both important to
making sense of this type of outcome (SWANZDSAJCS, 2006).
Russo (1994), who studied such discharges in Barcelona, Norris
(1984) and Steels, Roney, Larkin, Jones, Croudage & Duggan
(1998), who studied them in England, also examined mortality
and simple social indicators, such as return to families.

The tendency to focus on adverse events reflects social and
political concerns with re-offending. No health care worker wants
to be associated with repetition of a serious offence. These
concerns have led to an increase in the use of risk prediction
tools, although, in prospective studies, even actuarial measures of
risk prediction have been shown to perform at about chance levels
(Buchanan & Leese, 2001). Attempts to ‘allay public anxiety
through legal measures have been finely balanced against
professional medical opinion’ (Symonds, 1998), and remain
controversial (Joint Committee on the Draft Mental Health Bill,

Government responses to a single, rare tragedy often seem to
shape the manner in which discharge from high security hospital
units is considered. There has been very little attempt, however,
to understand the perspectives of those who are active in deciding
on discharge, and neither the process nor its evaluation is
theoretically driven. It was with this in mind that one of us (LJ)
started to consider generating a theory centred on the main
concerns of staff involved in the discharge process. A grounded
theory approach (Glaser & Strauss, 1967) was chosen. Concerns
exist for both the people detained and the people effecting the
detention or release from detention. We accept that each might
have a different perspective, and we acknowledge the lack of
secure hospital service user input as a possible limitation on the
theory generated. The reality is, however, that for this group of
hospital patients the departure and discharge decisions are made
by the sort of people interviewed rather than the patients/service
users themselves. On this basis it was considered important to
focus on the decision makers.


Ethics Committee approval

This research was approved by the West London Mental
Health Trust’s Ethics Committee and formed part of the doctoral
thesis of the first author. Each interviewee was provided with an
information sheet which explained the purpose of the study,
together with a consent form to sign, on agreement that his/her
data could be used. Signed consent was obtained for every direct
quotation that has been included and selected quotations were all
validated by the respondents themselves. Four interviewees
wished to clarify or amend certain quotations and this was done
in consultation with them. Two quotations which the
respondents subsequently considered were inaccurate were

Sample selection

There was no pre-determination of sample size. Its nature
was defined by the requirement that participants should have
some kind of role with respect to the outcome and rehabilitation
of serious offender patients. The candidates for the sample were
initially selected on the basis of their capacity and potential for
providing data on different aspects of the research question.
Thus, as the task was to develop a substantive theory according
to the main concerns of professionals about discharge from high
security, there was a requirement that the candidates for the
sample had knowledge of the hospitals, the full range of other
relevant services, their residents and, in general terms, possible
outcomes after discharge. Variation in level of security (low,
medium and high) in which they worked was sought deliberately,
as Glaser & Strauss (1967) suggested that categorical
development is slower in a single location. Theoretical sampling
helped to maximise the differences in the data and saturate the
categories. For instance, participants who would enable variety
in perspectives by virtue of training or disciplinary background,
or their position in the before and after discharge spectrum (the
position of recommending, promoting, or determining discharge
or of receiving or observing the discharged patient) were selected.

The final sample was made up of forensic psychiatrists (7)
with a range of experience and from a variety of secure settings, a
nurse manager, a nurse employed to facilitate transfer of
patients, a psychologist, social workers (2), psychotherapists (2),
education staff (2), Home Office Mental Health Unit civil
servants (3) and a specialist mental health lawyer.

Characteristics of the participants

The first interviewee, who was chosen arbitrarily, was a
consultant forensic psychiatrist. The next was chosen from the
same professional group to see whether there was withindiscipline
variation. Participants were interviewed thereafter
according to constant comparative analysis. They were from
various disciplines and settings. The point at which saturation of
categories occurred and there was a high degree of theoretical
integration between categories was after 20 interviews. Fifteen
had been with men and 5 with women. Their length of
experience ranged from a few months working directly with
special hospital patients, to over 20 years. Thus, the sample
included clinical and non-clinical respondents, those whose
clientele was explicitly made up of patients while mindful of
public safety, and those whose clientele was the public, while
mindful of individual patient rights, those with long experience
and those with relatively little, and people who had had minimal
or maximal contact with special hospital patients. Details of the
participants are shown in Table 1. Names have been changed for
the purposes of anonymity. Gender-appropriate, but false, names
have been given to interviewees.

Table 1 Summary of characteristics of participants
[please see PDF version for table]

Data collection

Data were collected over ten months from semi-structured
interviews. Each participant was interviewed once. Outcome of
discharge has typically been measured over long periods of five,
ten or more years. The research question was: ‘Please will you
describe your experiences and knowledge of discharges over a
period of about 10 years’ follow up?’ The interviews were open
ended and participants were encouraged to speak openly and
freely. Our theoretical interest was in their principal concern
with discharge. Shorthand notes were made by the researcher
during the interview, and typed up in full immediately, or up to
24 hours later in all but two cases (which were completed within
48 hours).

Data collection ceased when there was a high degree of
theoretical integration of the concepts and the data collected
provided no new insights to advance the theory. Saturation was
tested by continuing to look through the remainder of the data
set, returning to what seemed the most divergent examples
within the sample, looking for negative cases which did not fit
with the theory.

Data analysis

The data were coded line-by-line using a process called open
coding, in which each datum was inspected to generate categories
and their characteristics from particular indicators. Key words
and phrases that captured the essence of the data were used for
the categories, and noted on the margins of the transcripts.
These categories were not mutually exclusive, so one category
could be illustrated by many indicators or incidents in the data.
Data were examined for similarities, differences, and
consistencies (Glaser, 1978; Glaser & Strauss, 1967). Each
category was considered in terms of its dimensions or
characteristics. Questions were asked such as: ‘what category or
property of a category does this incident indicate?’ Each new
instance (indicator) was compared to other instances and to
categories as they emerged, using the process of constant
comparative analysis. Thus dimensions of categories came from
the data rather than being logically deduced or forced from
previous theory and these emergent distinctions made the
concepts rich and sensitive in explanatory power.

During the process of selective coding, important categories
were collapsed into a list of more general categories, or, in
Glaser’s terms, substantive codes. All subsequent transcripts
were coded using these categories, and text excerpts were
gathered under each category to show the range of variation
within each substantive code. The decisive criterion for the core
category was that it could encompass and explain the area of
interest, namely, the main concern regarding discharge of special
hospital patients. Inter-relationships of all categories to the core
category were considered.

Records of data collection and analysis were kept in the form
of memos, informal notes on the conceptualisations that emerged
from each coding session and which served as the building blocks
of the theory. Near the end of the analysis, memos were sorted
on the conceptual level, which put fractured data back together
and was the key to formulating the theory.

A literature review was not conducted until after the
substantive theory was formulated. PsychINFO (supplied by
OVID) and Medline were then searched for “Grounded Theory
studies”, “independence/ independent living”, “rehabilitation” and
“community /community care”. The literature was then
integrated into the theory.


The core concern

In this study, the main concern centred on movement
between pathological dependence and healthy independence.
Participants consistently raised themes of ‘dependence’ and
‘independence’, with comments about the ideas often entangled,
for example: ‘Dependence in our society influences everybody –
financial, work, relationships. They [the patients] are
disadvantaged in all three areas.’ (Susan1, 3.12.02, Clinical
Psychologist). Pathological dependence was influenced by
support. The greater the amount of personal support someone
could secure by his or her own efforts, the greater the degree of
independence that would be attributed to him or her, with
acknowledgement of the paradox in ‘healthy independence’,
incorporating mutuality and choice about dependencies and
reciprocities in relationships: ‘Well, none of us lives
independently. We should be encouraging people to use networks
– social services, support groups. They need to actively engage’
(Mike, 07.02.03, Occupational Therapist (OT)) and ‘Everyone is
supported by others. Having other people you can relate to and
managing an independent environment…you would have a
community mental health team. Support. But they would still be
living in the community’ (Paul, 10.02.03, Social Worker (SW)).
Community, however, is a nebulous concept, often evoking
scepticism and masking conflicts (Leff, 2001). Entanglements
appear in the literature too. A community may exist within an
institution (Wing, 1990), even a secure institution. Nevertheless,
there is usually a gap between a ‘made’ therapeutic community,
which even in high security may achieve a measure of democracy
and flattened hierarchies, and the community inhabited by the
wider public, which often demands separation and protection
from the very people health care professionals are hoping to
integrate within it (Symonds, 1998). There is therefore often a
conflict between the need for greater therapeutic independence
versus public reflections of the need for protection.

Pathological independence sometimes occurred by default.
‘Default independence’ occurs, for example, when a patient’s
detention order is absolutely discharged by a Mental Health
Review Tribunal. ‘They have been discharged because they are
untreatable, not because they are safe. When patients are
conditionally discharged, they have conditions – after absolute
discharge there is no statutory requirement for supervision’
(Helen, 11.10.02, Non-Clinical). ‘Absolute discharge’ most usually
happens for those detained under the legal category of
‘psychopathic disorder’ or ‘mental impairment’. For both, there is
a requirement under present mental health legislation for
England and Wales that “such treatment is likely to alleviate or
prevent a deterioration of [the patients] condition”, and if it is
determined that this is not the case, the patient must be
absolutely discharged from the treatment/hospital order. The
concept of treatability was further clarified in Cannons Park.
Absolute discharge generally means that the patient returns
directly to the community without supervision. Services can and
often do cease to offer anything to such people, who, in turn,
rarely choose to attend services consistently. One feature of
pathological independence is being unable to make appropriate
use of services and/ or not being able to access them at all, then
frequently returning to pathological dependence. ‘PDs (patients
under the legal category of psychopathic disorder) get “stuck” in
security or discharged via Tribunal to the community and fare
less well’ (Tom, 18.10.02).

The hospital was generally construed as a place of
dependency, providing safety or asylum as well as of constraint,
and the community, for various reasons as less safe, supportive or
protective, therefore demanding different qualities in
dependency. For example: ‘Special hospitals also provide drugs,
therapy, absence from drugs’ (William, 03.01.03, CFP). Years of
dependence on a protective environment and supervising and
treating staff were seen as threatening to leave the patient
unable to resolve his or her potential for pathological dependence:
‘Patients are scared of freedom and their potential. They need to
acclimatise, being seen as different, looking different, being
tattooed, looking like they are on medication, not fitting in, not
having support networks, food changes, routine, cars drive faster
than 10 years ago’ (Susan, 12.12.02, Clinical Psychologist).

The participants sometimes spoke of dependence simply in
terms of the opposite of independence. They identified three
main markers of independence: financial, functional and
emotional, including the ability to form reciprocal relationships.
Treatment for pathologies of dependence often involved
structuring arrangements around degrees of independence, with
full independence in one’s own home, in a satisfying partnership
and in paid employment achieved by few. In this respect,
facilitating independent living was a normalisation process,
enabling people to be as much in ‘the community like anybody
else… ’ (Clive, 29.11.02, non-clinical).

The core concern thus manifested itself both in binary code
(dependence v. independence; pathological v. healthy) and as a
continuum, with pathologies of dependence at one end and
healthy independence at the other. Evidence for the continuum is
illustrated in the next section which shows how staff resolve this
core concern and the quotations provided demonstrate how
participants spoke about ‘phased progression’ from dependence to
independence, ‘transition through to the community’ and ‘return
(back up the continuum) to hospital’ (dependence). Thus the
resultant theory is about position and/or movement between
these two poles through time.

Core category: Facilitating independent living

The core category in this study was about the resolution of
pathological dependence. This was achieved through facilitating
independent living. Attainment of the highest level of
independence for patients was the principal goal for health
professionals: ….’ (a good outcome)… independent living,
because that is what we aim for. A bad outcome is not achieving
independent living’. (Julian, 08.08.02, Forensic Psychiatrist).

Facilitating independent living refers to the basic social
process of helping patients become more independent and move to
the community, and appears to occur in two principal phases:
Phase I, a pre-discharge phase and Phase II, a post-discharge
phase. The process is about overcoming pathologies of dependence
and independence. For those who are facilitators, independent
living appears as a continuum with full pathological dependence
at one end, with the patient incarcerated and full healthy
independence at the other, with the ex-patient free in the
community, and professional help exclusively at his/ her request.

Facilitating independent living is directed at the goal of
increasing independence. It can, however, include temporary
increase in dependence. Kaliski (1997), for example, suggests
that many patients benefit from lifelong attachment to a forensic
unit, which allows for many episodes of independent living
interspersed with returns to the unit for stabilisation if there is
any indication of deterioration, but before the occurrence of any
disastrous event triggered by the illness. Transition into the
community could be characterised for some patients as a multiple
exit/re-entry/re-exit cycle. Each return is treated as an
opportunity to examine and modify the previous clinical plan, and
enable the patient to resume considerable self-sufficiency whilst
perhaps never attaining complete separation/independence from
their forensic unit.

Strategies for facilitating independent living

Early stages in the discharge process of a patient with
serious mental disorder involve clinical and social efforts directed
at enabling disengagement from exacerbating or undesirable
behaviours (e.g. dependence on drugs or alcohol) on the one hand,
and engagement with clinical and institutional requirements on
the other. This is paradoxical, in that such high dependency on
clinicians and services would be construed as pathological in the
absence of serious illness, but, given illness, it is people who
cannot achieve that dependency who are regarded as having a
pathology of dependency (i.e. when they fail, in Mechanic’s (1995)
terms, to give up their usual roles to take on the ‘proper role’ of a
sick person, namely taking on the task of getting well). If there is
recovery from the illness, but failure at this point to give up such
dependency, then once again dependency per se becomes

Facilitating independent living is achieved through paving
the way and testing out strategies, whereby the management of
the ‘risk’ of deterioration in the patient which might lead to an
undesirable outcome is shifted from clinician to patient as part of
the normalisation process. Health independence is achieved
when the patient makes all major decisions and assumes the
consequences. ‘Well in some ways we are a nanny institution but
we teach patients to take responsibility…..’ (Robert, 21.01.03,
Education). Patient competency, however, varies over time
‘…their ability to make decisions for themselves is variable and
variable over time. Independent living is where a person is given
autonomy to make their own decisions regardless of whether or
not they are competent. This varies’ (Jim, 17.02.03, SW).

‘Paving the way’ involved getting people to prepare for postdischarge
independence, for example, teaching the person
practical skills such as how to budget or cook. Paving the way
begins within high security and become less of a feature after
discharge. This preparatory process relies on various educational
techniques, inclusive of acquiring skills that may help prepare
the patient to gain paid employment and/or to ‘have a
constructive use of leisure time’ (Robert, 21.01.03, Education).
On pre-discharge wards, working with currencies and shopping
takes place – a kind of ‘social education rather than literacy.
More liberal approaches to escorted leaves’ (William, 03.01.03,
CFP). Sometimes this strategy just involves encouragement and
pushing the patient a little, ‘patients are frightened. Sometimes
you have to be cruel to be kind and force the issue’ (Mike,
07.02.03, OT).

Other teaching involves getting people to see themselves as
subjects and objects in social situations. ‘We also do role-play
asking what they would do if someone came up to them in a pub
and started calling them names. We teach them coping skills…..’
(Sarah, 07.01.03, Nurse Manager). Psychotherapy is considered
to facilitate independence because it improves selfunderstanding
‘patients who had self-awareness and
understanding of their own vulnerabilities, knowing what they
need’ (Angela, 08.01.03, Non-Clinical). They need to ‘become
aware of dangerousness of himself to himself’ (Michael, 03.10.02,
Psychotherapist). Paving the way also involves engaging and
informing families who ‘hate not being given information. Give
them a diagnosis and you empower them……’ (Jim, 17.02.03,
SW). The process suggests that the greater the cumulative effect
of these strategies, the better the outcome.

The second phase of facilitating independent living involves
‘testing out’. Testing out involves a patient being given
opportunities within and outside the institution to show to himor
herself and to staff levels of competency with new or reacquired
skills. It involves a tentative switch from pathological
dependence to pathological independence by monitoring the
effects of exposure to life outside institutional care. A property,
therefore, of testing out is its experimental nature. Testing out
takes different forms at different stages of the rehabilitation
process. Initial tests may include observing a patient cook and
clean without prompting from the staff; a next stage test might
include escorting the patient to shops outside the hospital.
Testing out can be extended when the patient is in another
hospital of lower security or in the community. It will be
‘….probably six months on trial leave before a formal transfer
would take place’ (Paul, 11.10.02, CFP).

Paving the way and testing out overlap to some extent, and
the ‘preferred’ route to independent community living as a phased
progression through ever lower levels of security and active
clinical input support might be seen as an indicator of that:
‘Medium security is preparation for discharge to the community.
There are increasing periods of leaves, escorted and unescorted.
They are assessed, targets are met. ‘About 70-80% will go through
some secure setting as an intermediate stage. About 20-30% will
go through all levels of security, i.e. high to medium to low
secure/rehab’ (Clive, 29.11.02, Non-clinical). These people were
said to be ‘those with clear mental illness who have responded
well to treatment…where it was felt necessary to re-assess risk in
a carefully controlled manner by moving patients to less
restrictive settings in a stepwise manner’ (Geoff, 28.10.02, CFP).
The process of facilitating independent living continues to
assist the patient to relinquish pathological dependence, but also
prevent pathological independence, whereby the patient rejects
further treatment or support rather than accommodating a
reduction of it. Instability in mastery of the underlying core
problem (pathologies of (in)dependence) means that re-admission
to a higher level of security (back to pathological dependence) can
happen soon after the initial transfer to lower security. This can
vary by disorder. ‘Break down because of change of environment,
new team. Could either be people with psychopathic disorder –
issues around attachment/ trust/confidentiality or long-term
chronic schizophrenics – stress of moving’ (Julian, 08.08.02, CFP).

It can also vary with the nature of transition. Sometimes
patients would have ‘to start over with psychologists and prove
themselves all over again with a new clinical team’ (Angela,
08.01.03, legal representative). Good communication between old
and new teams is important to smooth progression along the

Facilitating independent living: Enhancers and

The facilitating process is affected by ‘enhancers’ or
‘barriers’. Many factors can be either. Enhancers include
available support systems or confidence of the clinical team.
Barriers include permanent inhibitors which cannot be changed,
for example past violence or crime and personal characteristics,
and temporary inhibitors which can, like medication compliance
or service provision. These are comparable to the fixed
dispositional and historical factors and mutable contextual
factors described by the MacArthur risk study group (Steadman,
Monahan, Applebaum, Grisso, Mulvey, Roth, 1994). While
enhancers tend to allow stepwise progression, barriers may
prevent, slow or abruptly cut off the process. A barrier appearing
after some progress along the continuum and ending
independence gained was called a ‘terminator’.

Youth was an example of a ‘dispositional factor’ seen by some
as a potential barrier to facilitating independent living, first in
securing departure from special hospital. ‘I believe the younger
remain for longer… [in special hospital]’ (Paul, 10.02.03, SW).
Older, not chronically institutionalised patients were considered
to pose low risk of repetition of harm, and so be more safely
tested out and gain movement towards independence. Over time
patients mature and the personality disorder could be outgrown
or mental illness burnt out. ‘Special hospitals …… allow the
personal maturation process.’ Researcher: So are older people
more likely to go to the community then? Yes’ (William, 03.01.03,
CFP). The opposite view was, however, also expressed. ‘The
elderly or infirm, those who are older’ or ‘people with severe
enduring mental illness. I would say a small percentage never
reach the community, about 20-30%’ (Sam, 10.09.02, CFP).

The barrier to facilitating independent living may lie as
much with others as with the patient: ‘There is a resistance to
take on PDs or people who have proved difficult/ assaultive in the
past. They believe that therapy is impotent to do anything about
patients with PD. They see the risk as chronic and it is difficult
to measure and assess clinical risk….people with PD are anxiety
provoking’ (Paul, 11.10.02, CFP). This pessimism about outcome,
and fear about the potential of the people with personality
disorder to disrupt services is not uncommon (Coid & Cordess,
1992). Others, however, point to a substantial evidence base for
explanatory pathways into personality disorder and effectiveness
of some treatments (Taylor, Newrith & Meux, 2006).

The nature of the legal constraint is another contextual
factor which may be construed both as an enhancer and a barrier
to independent living. Under the Mental Health Act 1983, in a
higher Court, a restriction order may be added to a hospital
order, if the judge considers it necessary for the protection of the
public. This has the effect of restricting powers of discharge, but
also ensures that clinicians continue to offer appropriate
treatment, support and supervision. Transitional arrangements
within this framework tend to be particularly highly structured.
It was perceived, though, that unrestricted patients would move
through the system more quickly ‘…by definition they are less
dangerous so I imagine a greater number would go directly to the
community than the restricted’ (Helen, 11.10.02, Non-Clinical).
Dell (1980), however, found that patients under restriction orders
and others with criminal convictions were, in fact, the most
readily placed.

Enhancers to facilitating independent living

Personal characteristics, such as compliance,
trustworthiness, motivation and insight, with ability to form,
build and maintain relationships were cited by the participants
as factors in the patient which would enhance the process of
facilitating independence. Achievement of awareness of anxieties
about living in the community and of personal vulnerabilities was
also an indicator of such useful insight: ‘Realisation that mental
disorder is for life. Insight is vital….’ (Helen, 11.10.02, Non-

Clinical team confidence in the patient and reciprocal patient
confidence were seen as the other enhancers: ‘Outstanding thing
is a good relationship between the patient and just one member of
the clinical team. It must be based on mutual trust. So it
becomes a conversation between two people, one with the
knowledge of psychiatry, the other with knowledge of themselves’
(Angela, 08.01.03, Non-Clinical). Responsibility for beginning the
discharge process is borne by the team but, as the patient
progresses along the continuum of independence, it transfers to
the patient.

Strengths in the patient’s own support systems provided
another layer of enhancers: ‘Yes, we do look for family support.
Stable structures. Someone to watch over them. Encouragement.
This would help us decide on absolute discharge’ (Helen, 11.10.02,
Non-Clinical). ‘Family support seems to be vital. For example,
one patient was discharged to a staffed hostel but his parents are
still willing to look out for him’ (Sam, 10.09.02, CFP).
Conversely, having no family or friends was seen as a potential
barrier to facilitating independence: ‘The patients with no family
are the most reluctant to leave’ (Sarah, 07.01.03, Nurse

Barriers to facilitating independent living

The offence leading to the admission to high security
hospital, an immutable factor, was perceived as contributing to
the route of discharge and speed of progression. ‘Some will go out
via medium security/ regional security, others will go to
supervised hostels. Researcher: What factors determine this?
Answer: The index offence’ (Sarah, 07.01.03, Nurse Manager).
Lack of a clear link between the illness and the index offence, or a
perception that the mental disorder which had been present at
the time of the offence was untreatable, were both factors which
were regarded as barriers to the process. Relationship to the
victim was also seen as important, and probably wisely so
(Johnston and Taylor, 2003): ‘very unusual for anybody to be
returned to family, index offence is often against people [in the
family]’ (Simon, 19.02.03, CFP). Sometimes a high community or
media profile, largely secondary to the offence, was thought to
pose a further barrier. Pressure from third parties or victims,
public hostility, or political sensitivity would have to be managed
as well as the actual safety of the patient. ‘There are other issues
like DSPD [Dangerous Severe Personality Disorder] and sex
offenders – those who respond to treatment – but whether they
are safe in the community is a different matter. That group
needs close support. So stigmatised’ (Susan, 12.12.02, Clinical

Some barriers were considered to lie within the patient.
These included reluctance to take medication, lack of empathy or
insight, resentment of supervision, inability to cope with money
and/or to build and maintain relationships. These barriers were
all, however, regarded as mutable, and therefore temporary
inhibitors. ‘They worry that they cannot cope with life out there
and money’ (Sam, 10.09.02, CFP). Consequences of long-term
institutionalisation were also regarded as remediable for many,
although not all: ‘Patients are scared of freedom and their
potential. They need to acclimatise…..’’ (Susan, 12.12.02, Clinical
Psychologist). ‘Money is an issue – they don’t get paid in an RSU
[Regional secure unit]. Daytime activities are another.
Integration or not is another….family closeness. Access to the
community’ (Deborah, 17.02.03, Community Psychiatric Nurse
(CPN)). By contrast, co-morbidity, here meaning drug or alcohol
abuse co-occurring with the main mental disorder, was commonly
construed as a problem for life. It was often thought to play a
part in terminating independent living. Problems with alcohol
are strongly related to re-conviction and re-admission (Norris,
1984), and patients with such co-morbidity are ‘difficult to place’
because of not falling within responsibilities of one single agency
(Johnstone, Owens, Gold, Crow & Macmillan, 1984).

Women patients were said to encounter more barriers to the
process of facilitating independence than were men: ‘I think
females have more problems with consistency of staff; they are
more complex and harder to manage’ (Deborah, 17.02.03, CPN).
Also: ‘…often there has been abuse.. . very little thought about
what they do need’ (Simon, 19.02.03, CFP).

Another demographic, ethnic group or race was seen as a
potential barrier because of a likely mismatch between race of
patient and of authority figures: ‘…There is a cultural difference
– managers are white, Tribunals have white people. There is a
definite disadvantage to being black’ (William, 03.01.03, CFP).
This may have similarities with suggestions of such problems in
the Criminal Justice System (e.g. Hood, 1992; Fitzgerald, 1993).
The NHS generally is not regarded as immune to such problems.
Bhui, Stansfeld, Hull, Priebe, Mole, & Feder (2003) found that
black people are over-represented amongst in-patients and four
times more likely to experience a compulsory admission than
white people, although adjusting for diagnosis reduces such
difference (Harrison, 2002). African Caribbeans with psychosis
are, however, no more likely to be in an English high security
hospital than in general psychiatric services (Walsh, Leese,
Taylor, Burns, Creed, Higgit, & Murray, 2002).

External factors were sometimes seen as barriers in
themselves, including delays in Home Office approval of plans for
restricted patients. Further, some placements require
identification of new funding, with delays in locating it. Lack of
suitable beds was also raised: ‘….. There may not necessarily be a
purpose in sending them to medium security but it is to do with
bed availability’ (William, 03.01.03, CFP). A further set of
barriers were considered to lie in staff training. Decisions about
transferring or discharging patients are influenced by attitudes
and philosophies among receiving clinicians, as well as by their
knowledge of the patients (Peay, 1989). ‘Many staff are not
trained forensically’ (Sarah, 07.01.03, Nurse Manager). For other
staff, an “unlearning” of previous experience would be needed
(Rawlings, 2001). ‘I think the route you come in can seriously
affect your thinking. Some staff come onto the admission ward
having worked in the community. People don’t always apply
security. I think there is a division between mental health and
forensic’ (Mike, 07.02.03, OT). Again, the theme came through
that patients with personality disorder might experience special
barriers: ‘Inexperience with the community team. They have
never been trained how to deal with people with personality
disorder. They will be successful with MI. Community
psychiatric services do not want to know people with PD so the
team does not get experience. Community Psychiatric Nurses
need more training’ (Jim, 17.02.03, SW).


A ‘good outcome’ of the discharge process was having
‘successful accommodation, employment, a lack of drug use,
limited alcohol intake, maximised his/her potential, secure
relationships of his/her choice, a constructive and contented life,
the ability to deal with problems, and being in the most
independent placement possible, compliant and not
institutionalised’. ‘Return to safe responsible ‘independent’ living
that they are happy with’ (Helen, 11.10.02, Non-Clinical).
Additionally, there must be a lack of re-offending and rehospitalisation
and being free of, or at least untroubled by,

‘Bad outcome’ was characterised by lifelong pathological
dependence and institutionalisation, with no self-sufficiency for
the patient, long re-admissions to hospital, re-offending, suicide,
relapse, or being stuck in an inappropriate placement – ‘Staying
with no prospect of moving on or moving on to MSU with no
rationale for moving there’ (William, 03.01.03, CFP). A bad
outcome also referred to premature discharge, a form of
pathological independence. There was also recognition of high
rates of physical ill health ‘There is high morbidity due to
smoking and lack of exercise. Suicide rate in schizophrenics is
about 11%’ (Tom, 18.10.02, CFP)

Such good and bad outcomes constitute extreme poles; for
most patients, it was more a case of ‘As good as it gets….’ where
this would refer to the very best progress that a patient could
make. ‘To become as independent as possible recognising that
some won’t make it to the end’ (Tom, 18.10.02, CFP). In this
context, ‘the end’ or the real aim of discharge could be taken to
mean healthy dependence. To this end there was recognition of
the value of teaching patients to be as dependent as they feel they
need be to meet real needs. The dependence only becomes
pathological if independence is the healthy, appropriate
adaptation and vice versa.

A patient may be appear to be living independently, not
offend, not be re-admitted, and comply with all requirements but
have no quality of life, no job, self-esteem or relationship. ‘It
doesn’t matter what the job is but they need an occupation – it
could be a porter, whatever, but it needs boring regularity
coupled with friends. If there was family support on top, that
would be marvellous’ (Michael, 03.10.02, Psychotherapist).
Occasional relapses would be acceptable: ‘Living in the
community; ‘blips’ get noticed perhaps with ‘blipping’ admissions;
low level offending (not violence)’ (Simon, 19.02.03, CFP).


Information derived from interviews about discharge of
patients from high security hospitals, undertaken with a mix of
clinical and non-clinical staff involved in decisions to discharge,
revealed a core concern about pathologies of dependence and
independence. This was accompanied by recognition that a
measure of dependence could be healthy if freely chosen;
examples of pathologies of independence include enforced
detachment from services or total rejection of attachment. The
emergent theory was of a process of facilitating healthy
independent living, with two principal phases, which overlap.
These phases are ‘paving the way’, mainly occurring in the predischarge
period, and ‘testing out’ occurring more in the post
discharge phase.

Facilitating independent living is an active process by
clinicians and other professionals creating movement from full
pathological dependence at one end of a continuum, – here,
compulsorily detention in hospital – towards healthy
independence at the other end – here, when the patient becomes
a person who is able to live safely in circumstances of his or her
choosing, whether alone or with a partner, family or friends, and
with or without professional help. In this model, progress and
deterioration are movements up and down a dependenceindependence
continuum, personal or social network strengths
are enhancers of progress towards independence and adverse
events, such as relapse of illness or re-offending, barriers to
independence, in themselves and in the nature of the responses to
them – perhaps return to hospital.


The theory generated provides a positive and fresh
framework for understanding and measuring outcome on leaving
a high security hospital, which fits better with a philosophy of
clinical practice and rehabilitation than old methods of focussing
almost exclusively on re-offending, or, from time to time, on other
negative events. The majority of people leaving high security
hospitals are not now and never have been convicted of a further
offence; e.g. in the UK, Buchanan (1998), Jamieson and Taylor
(2004), Tennent and Way (1984) and Tong and Mackay (1959);
and, in the USA, Steadman and Keveles (1972) and Steadman
and Cocozza (1974). If used alone and without context, this is an
unsatisfactory marker of outcome because the offences these
patients commit tend to have a low base rate, some patients may
be violent but escape involvement with the criminal justice
system, and both the definition of offences and attitudes to them,
including the likelihood of prosecution vary geographically and
over time. A model which is less susceptible to social and cultural
vicissitudes and which more explicitly puts gains for the patient
as well as others at the centre of the rehabilitation process is
likely to enhance co-operation between staff and patients and, in
turn, the chance of good outcomes.

This perspective fits well with the ethical code of most people
providing services for such patients, and may also fit better with
community safety than a risk focussed approach. Munro and
Rumgay (2000), for example, analysed a consecutive series of
independent inquiries after homicide by people who had been in
contact with UK mental health services, published 1988-1997,
and found that even with the advantage of hindsight, only one
quarter of the inquiry teams had judged the homicide as having
been predictable, but in two-thirds it was considered that it may
have been preventable with treatment. In other words, focus on
the patients’ health and social needs rather than their offending
would probably have been more effective in prevention of these

Our theory of facilitating independent living also has
important implications for the broad social processes of deinstitutionalisation
and the move towards community care,
designed to help patients achieve independent living. One of the
guiding principles for the treatment of mentally disordered
offenders in the UK Department of Health & Home Office’s (1992)
Review of Health and Social Services for Mentally Disordered
Offenders was that they should be cared for “as far as possible in
the community, rather than in institutional settings”; “under
conditions of security no greater than is justified by the degree of
danger they present to themselves or others” and “in such a way
as to maximise rehabilitation and their chances of sustaining an
independent life”. Indeed, the whole concept of ‘community care’
is about providing care for mentally disordered people outside
hospital in community residential settings, allowing them the
freedom to make their own decisions. The process aims to ensure
that some of the protective functions of the institution are fully
provided in the community and the negative aspects of
institutional care are not perpetuated (World Health
Organisation, 2001). Thus, de-institutionalisation is associated
with “de-hospitalisation” but not synonymous with it. A package
of resources, including health workers and rehabilitation services,
general hospital psychiatric beds, other specialist accommodation
and home care, crisis support, protected housing, and sheltered
employment allows service users more choice in selecting the
optimal cluster of resources for their needs. Although terms such
as ‘protected’ and ‘sheltered’, however, carry some connotations of
dependency, elements of choice on the part of the service user and
openly negotiated agreement between clinician and user on the
choice of service use implies a substantial measure of healthy

The core problem that this theory articulates is that a
complex dynamic exists between pathological dependence and
pathological independence. We all are dependent on such things
as food, money and relationships in our lives, to various extents,
but the level of dependence and independence should interact
positively with each other to allow an individual to develop or
maximise his/her potential and to have some control over the
balance of dependence/ independence. It is a fundamental
characteristic of a social being that he or she is not wholly selfsufficient,
practically or emotionally. As human beings, we
depend on relationships with others to survive, and interaction
with others is generally reciprocal.

There is, nevertheless, a general pattern among healthy
social animals that they start life as highly dependent on one or
both parents or caregivers, and progress towards increasing
capacity for independence, including autonomy in choice and
actions. Some kind of disruption to this process may be referred
to as a pathology of dependence. For people with developmental
disorders, such as learning disability or personality disorder, such
progression may be slowed or arrested. Among mechanisms for
understanding the latter, is disorder of attachment (Bowlby,
1969; Barber M., Short J., Clarke-Moore J., Lougher M., Huckle
P., & Amos T., 2006).

Direction and stability of pathologies of dependence may be
affected, such that an individual might be rendered overdependent
at one extreme, autistically separate and without
reciprocity at the other, or swinging between the two. Illnesses,
disorders which constitute a break in health, may also affect
dependence. The nature of care and treatment for such disorders
may compound any pathological dependence directly related to
the disorder. Where disorder is profound and chronic, and longterm
care and treatment has been provided in an institution, this
is particularly likely.

The aim that a child should grow up to become confidently
independent is synonymous with the aim that he should grow up
to be mentally healthy (Bowlby, 1956). There are parallels in the
aims of clinicians, and their sometimes parent-like roles in easing
a patient along towards independence. While not expressed by
any of the interviewees, a difficulty here is the risk of reinforcing
a sort of pathological dependency, colluding with a form of
infantilization of the patient. Most patients are well over 30
years old by the time of consideration of discharge from high
security. Many of them experienced pathological attachments
within their family; most would have felt or been pathologically
dependent, whether or not that dependence was met, when their
mental disorder left them less able to survive independently in
the community. Both schizophrenia, which most commonly has
its onset late in adolescence, and personality disorder, which has
generally affected the individual for the whole of his or her life,
interfere with the normal process of gaining independence from
parents/parental surrogates. Work with families, with the
consent of the patient, is likely to be important.

Forms of dependence can be both natural, as in a motherchild
relationship, and pathological from a societal point of view,
as in institutionalisation. A further distinction exists between
forced dependence, as in incarceration, and free-choice
dependence, when deprivation of a particular object, person, or
substance or place would cause distress. This can be seen in
patients who are distressed when discharge is considered. One
way of looking at violence born out of paranoia is that it is a
disruption of a former pathological dependence created by a belief
system about a perceived persecutor. Treated patients who have
been freed from such pathology may experience emptiness
without the beliefs and mourn their loss.

A flawed notion of community at the heart of the process of
de-institutionalisation in some respects contributes to the
pathology of independence. Communities are complex tapestries
of social and political forces, not always welcoming to those who
are committed to its care. A more critical awareness of what
community actually means in community care is needed.


This theory is substantive because the focus has been on a
specific area of inquiry that concerns inter-disciplinary practice.
The aim of facilitating independence for people who have been
detained in high security hospitals appears to have much in
common with the overarching aim for psychiatric patients or
other people in institutional care, in which case, the implications
are profound for the theory and its generalizability. In order to
develop or refine the substantive theory further, resident patient
views should be obtained. In order to build formal theory
pertaining to the conceptual area of pathology of dependence, the
phenomenon should be examined under several types of
situations, such as people with mental illness living in the
community and within hospital, both before and after treatment.
This would have the aim of exploring further how the discharge
process affects the pathology of dependence, as entry into the
patient role (or sick role) is often a last resort, following extensive
delays and repeated attempts at self-care (Mechanic, 1995).
Further work could examine Bowlby’s (1956) distinction between
actual dependence and feeling dependent. It could be that for
patients with little experience of survival in the community, the
feeling of dependence is hard to overcome and thus a patient may
appear to an outsider to be living independently and to be in a
free-choice state, but in reality have feel very dependent.

Since this theory was developed, a paper has been published
by Draine, Wolff, Jacoby, Hartwell, & Duclos, 2005, who
developed a model of prisoner re-entry into the community
through interdisciplinary team effort and refined through a focus
group process that included advocates, community members and
other informants from mental health and criminal justice
systems in five states. It illustrates dynamics related to both
individuals with mental illness leaving prison and their
interaction with the community setting. It also has distinct
parallels with the theory presented in this paper in terms of
Enhancers and Barriers as the authors model how resources and
needs at both community and individual levels can interact to
support or hinder the community integration of individuals
leaving prison.


The principal researcher (LJ) who conducted all the
interviews and the analyses is not a clinician, and had few
preconceived ideas about discharge and outcome for high security
hospital patients, however, she did regularly attend clinical
seminars, and the principal supervisor (and one of the co-authors)
is a clinician of long experience at all levels of secure service
provision and management. Every attempt was made to
approach this material with an open mind, and to reintroduce
prior knowledge and research evidence as a final stage of the
analytic process, but bias cannot be ruled out.

The theory that was developed is not necessarily the only one
that might plausibly have been derived from the data.
Sometimes it is possible to account for behaviour by more than
one concept. However, the theory derived in this study was the
interpretation of the data and all decision-making processes have
been made explicit by giving examples of direct quotations.
Every attempt was made to ensure that theoretical saturation
had been achieved but new information might have arisen had
the interviews continued and this is true of any grounded theory
study. A grounded theory is never right or wrong but is always
modifiable in the light of new information. For example, this
study only examined outcome from the perspective of those
making discharge decisions. Addition of a service user perspective
might result in key modifications.


It was possible, by adopting a grounded theory approach, to
develop a substantive theory of facilitating independent living
that explains the main concerns with discharge from special
hospital, according to mental health staff. After 20 interviews,
the theory was saturated. The theory developed was that
discharge for special hospital patients is a process of facilitating
independent living. Independent living is portrayed as a
dependence-independence continuum. The process is facilitated
through strategies of ‘paving the way’ and ‘testing out’, to each of
which both patients and others, especially professional staff,
contribute. The process is affected by ‘influencing factors’, which
may be ‘enhancers’ or ‘barriers’, and these in turn may be in the
patient or in their external environment.


Liz Jamieson, formerly Research Fellow, Broadmoor Hospital,
West London Mental Health Trust, Crowthorne, Berkshire,
England, RG45 7EG.

Correspondence to:
Pamela J. Taylor, Professor of Forensic Psychiatry, Wales College
of Medicine, Cardiff University, Cardiff CF14 4XN.
Telephone: +44 (0) 29 2074 3090

Barry Gibson, Lecturer in Medical Sociology, Department Oral
Health and Development, School of Clinical Dentistry, Claremont
Crescent, Sheffield, S10 2TA.
Telephone: +44 (0) 114 271 7889


Liz Jamieson would like to acknowledge the guidance and advice
received from Dr. Barney Glaser in the early stages of coding.
This theory was developed as part of her doctoral thesis.
Professor Pamela Taylor supervised the PhD and contributed to
the later refinement of the theory. She provided significant
academic input and wrote sections of this paper. Dr Barry Gibson
initially provided feedback on the early coding and methodology
and later contributed to the further development of the theory,
drawing on his expertise in this area. He also contributed to the
writing of this paper, shaping it for publication.

The authors would also like to acknowledge the contribution of
the participants and the valuable comments received by the peer


Barber M., Short J., Clarke-Moore J., Lougher M., Huckle P., &
Amos T. (2006). A secure attachment model of care:
meeting the needs of women with mental health problems
and antisocial behaviour. Criminal Behaviour & Mental
Health 16: 3-10.

Bhui, K., Stansfeld, S., Hull, S., Priebe, S., Mole, F., & Feder, G.
(2003). Ethnic variations in pathways to and use of
specialist mental health services in the UK. British
Journal of Psychiatry, 182, 105-116.

Bowlby, J. (1956). The growth of independence in the young child.
Royal Society of Health Journal, 76, 587-591.

Bowlby, J. (1969). Attachment and loss: Attachment (Vol. 1). New
York: Basic.

Buchanan, A. (1998). Criminal conviction after discharge from
special (high security) hospital. Incidence in the first 10
years. British Journal of Psychiatry, 172, 472-6.

Buchanan, A., & Leese, M. (2001). Detention of people with
dangerous severe personality disorders: A systematic
review. The Lancet, 358, 1955-1959.

Butwell, M., Jamieson, E., Leese, M., & Taylor, P. (2000). Trends
in special (high security) hospitals. 2: Residency and
discharge episodes, 1986-1995. British Journal of
Psychiatry, 176, 260-265.

Coid, J., & Cordess, C. (1992). Compulsory admission of
dangerous psychopaths. British Medical Journal, 304,

Cannons Park R v. Cannons Park Mental Health Review
Tribunal, ex parte A [1993] Court of Appeal [1994] Times
Law Report 2.3.1994 .

Dell, S. (1980). The transfer of special hospital patients to NHS
hospitals. British Journal of Psychiatry, 136, 222-34 &
Special Hospital Research Report, No.16. London.

Department of Health & Home Office (1992). Review of Health &
Social Services for Mentally Disordered Offenders &
Others Requiring Similar Services (Chairman: Dr. John
Reed). Final Summary Report. Cm 2088. London: HMSO.

Draine, J., Wolff, N., Jacoby, J.E., Hartwell, S. & Duclos, C.
(2005) Understanding Community Re-Entry of Former
Prisoners with Mental Illness: A Conceptual Model to
Guide New Research.

Behavioral Sciences and the Law, 23, 689–707.

Fitzgerald, M. (1993). Ethnic Minorities and the Criminal Justice
System. The Royal Commission on Criminal Justice
Research. London: HMSO.

Glaser, B.G., & Strauss, A.L. (1967). The discovery of Grounded
Theory: strategies for qualitative research. Chicago:

Glaser, B.G. (1978). Theoretical sensitivity. Mill Valley, CA:
Sociology Press.

Harrison G. (2002) Ethnic minorities and the Mental Health Act.
British Journal of Psychiatry 180:198-199.

Hood, R. (1992). Race and Sentencing. Oxford: Oxford University

Jamieson, E., & Taylor, P.J. (2004). A re-conviction study of
special (high security) hospital patients. British Journal
of Criminology, 44, 783-802.

Johnstone, E.C., Owens, D.G.C., Gold, A., Crow, T.J., &
MacMillan, J.F. (1984). Schizophrenic patients discharged
from hospital – a follow-up study. British Journal of
Psychiatry, 145, 586-590.

Johnston, I., & Taylor, P.J. (2003). Mental disorder and serious
violence: the victims. The Journal of Clinical Psychiatry,
64, 819-824.

Joint Committee on the Draft Mental Health Bill (2005). Report.
Accessed on March 23, 2005. Available from

Kaliski, S.Z. (1997). Risk Management during the Transition
from Hospital to Community Care. International Review
of Psychiatry, 9, 249-256.

Leff, J. (2001). Editorial. Why is Care in the Community
perceived as a failure? British Journal of Psychiatry, 179,

Mechanic, D. (1995). Sociological dimensions of illness behaviour.
Social Science and Medicine, 41(9), 1207-1216.

Munro, E., & Rumgay, J. (2000). Role of risk assessment in
reducing homicides by people with mental illness. British
Journal of Psychiatry, 176, 116-120.

Norris, M. (1984). Integration of special hospital patients into the
community. Aldershot: Gower.

Peay, J. (1989). Tribunals on Trial. Oxford: Clarendon Press.
Pruesse, M., & Quinsey, V. (1977). The dangerousness of patients
released from maximum security. Journal of Psychiatry
and the Law, 5, 293-299.

Rawlings, B. (2001). Life and therapy in a prison therapeutic
community: The experience of researcher, staff and
inmates. Therapeutic Communities: the International
Journal for Therapeutic and Supportive Organisations,
22(4), 319-334.

Russo, H. (1994). Follow-up of 91 mentally ill criminals
discharged from the maximum security hospital in
Barcelona P.G. International Journal of Law and
Psychiatry, 17(3), 279-301.

Steadman, H.J., & Keveles, G. (1972). The community
adjustment and criminal activity of the Baxstrom
patients: 1966-1970. American Journal of Psychiatry,
129, 304-310.

Steadman, H.J., & Cocozza, J.J. (1974). Careers of the criminally
insane. Lexington, Massachusetts: Lexington Books.

Steadman, H., Monahan, J., Applebaum, P., Grisso, T., Mulvey,
E., Roth, L., Robbins, P., & Klassen, D. (1994). Designing
a new generation of risk assessment research. In:
Monahan J. Steadman H., eds. Violence and Mental
Disorder: Developments in Risk Assessment. Chicago, Ill:
University of Chicago Press: 297-318.

Steels, M., Roney, G., Larkin, E., Jones, P., Croudace, T., &
Duggan, C. (1998). Discharged from hospital: A
comparison of the fates of psychopaths and the mentally
ill. Criminal Behaviour and Mental Health, 8: 39-55.

SWANZDSAJCS international research collaboration on
treatment of offender patients (2006) Offenders with
mental disorder on five continents: common ground and
differences in demographics, culture, laws and services.
Conference presentation, International Association of
Forensic Mental Health Services.

Symonds, B. (1998). The philosophical and sociological context of
mental health care legislation. Journal of Advanced
Nursing, 27(5), 946-954.

Taylor P.J., Newrith, C., & Meux, C. (2006). Treating Personality
disorder: towards a future of more mutually satisfactory
and rewarding outcomes. In Newrith, C., Meux C. &
Taylor, P.J., eds. Personality disorder and serious
offending: Hospital treatment models. London, UK:
Hodder Arnold.

Tennent, G., & Way, C. (1984).The English Special Hospital – a
12-17 year follow-up study. A comparison of violent &
non-violent re-offenders & non-offenders. Medicine,
Science & the Law, 24(2), 81-91.

Thornberry, T.P., & Jacoby, J.E. (1979). The criminally insane.
Chicago: The University of Chicago Press.

Tong, J.E., & Mackay, G.W. (1959). A statistical follow-up of
mental defectives of dangerous or violent propensities.
British Journal of Delinquency, 9, 276-84.

Walsh, E., Leese, M., Taylor, P.J., Burns, T., Creed, F., Higgit, A.,
& Murray, R. (2002). Psychosis in high security and
general psychiatric services. Report from the UK700 and
Special Hospitals’ Treatment Resistant Schizophrenia
groups. British Journal of Psychiatry, 180, 351-357.

Wing J.K (1990). The Functions of Asylum. British Journal of
Psychiatry, 157, 822-827.

The World Health Report 2001 – Mental Health: New
Understanding, New Hope. World Health Organisation.
Chapter 3. Solving Mental Health Problems.