International Review of Psychiatry, February 2011; 23(1): 61–69
Clinical and economic outcomes from the UK pilot psychiatric services for
personality-disordered offenders
ZOË FORTUNE1, BARBARA BARRETT1, DAVID ARMSTRONG2, JEREMY COID3,
MIKE CRAWFORD4, DAVID MUDD5, DIANA ROSE1, MIKE SLADE1, RUTH SPENCE1,
PETER TYRER4, & PAUL MORAN1
1
Health Services and Population Research Department (Institute of Psychiatry), King’s College London, London,
Department of Primary Care & Public Health Sciences, King’s College London, 3Department of Forensic Psychiatry,
Bart’s and the London, Queen Mary’s School of Medicine and Dentistry, St Bartholomew’s Hospital, London,
4
Division of Neuroscience and Mental Health, Claybrook Centre, Imperial College London, and 5School of Health
and Social Care, University of Teesside, Middlesbrough, UK
Int Rev Psychiatry Downloaded from informahealthcare.com by Kings College London on 04/19/11
For personal use only.
2
Abstract
Personality-disordered offenders are difficult individuals to manage, and knowledge about effective treatment is sparse.
In the UK, novel forensic psychiatric services were recently established for the treatment of offenders with personality
disorder. In this paper we report the clinical and economic findings from a 2-year follow-up of a cohort of service users
recruited from these services. Baseline information on developmental, clinical and offending histories was obtained from case
records. Case records were checked at 6 and 24 months for new episodes of self-harm, violence, alcohol and substance use,
and offending behaviour. Ratings of social functioning and therapeutic alliance were obtained from service users at baseline,
6 and 24 months. Fifty-six percent of service users were still engaged with the services at 24-month follow-up. Service users
involved in the greatest number of behavioural incidents had greater impairment in baseline social functioning and lower
IQ scores. There was no significant change in either therapeutic alliance or social functioning at 6 or 24 months. The
economic analysis showed that although the services were predominantly run by the Health Service, there were considerable
economic burdens shared by other service providers. Treatment costs at six-month follow-up were also significantly higher.
Implications are discussed.
Introduction
Personality disorders (PD) are common mental
disorders occurring in all cultures (Huang et al.,
2009). They are associated with a considerable
public health burden (Tyrer et al., 2010). Not only
does the diagnosis of PD predict violent behaviour
and self-harm (Coid et al., 2006; Haw, Hawton,
Houston, & Townsend, 2001), but robust associations also exist between PD and comorbid mental
and physical health problems (Hayward & Moran,
2008; Moran et al., 2007), substance abuse (Grant
et al., 2004; Moran, Coffey, Mann, Carlin, & Patton,
2006), unmet treatment needs (Hayward, Slade,
& Moran, 2006) and poor treatment response
for associated mental illness (Gorwood et al., 2010;
Newton-Howes, Tyrer, & Johnson, 2006).
Unsurprisingly, people with PD place considerable
burden on health and social care services and the
associated increased economic costs are substantial
(McCrone, Dhanasiri, Patel, Knapp, & LawtonSmith, 2008; Rendu et al., 2002).
Surveys of psychiatric morbidity have repeatedly
demonstrated that the prevalence of PD within the
Criminal Justice System is considerably higher than
the community (Fazel & Danesh, 2002; Singleton,
Meltzer, & Gatward, 1998). Those with antisocial
personality disorder are more likely to be violent
(Steadman et al., 1998) and commit further offences
whilst undergoing sentences (Coid, Hickey, Kahtan,
Zhang, & Yang, 2007). The associated economic
costs for offenders with personality disorder are high
(Barrett et al., 2009). Nevertheless, it is unclear
how many personality-disordered offenders actually
have problems that might benefit from psychiatric
treatment, and little is known about what might
constitute effective treatment for such individuals
(British Psychological Society & Royal College of
Psychiatrists, 2009).
Correspondence: Zoë Fortune, PO28, Health Services and Population Research Department, Institute of Psychiatry, De Crespigny Park, London, SE5 8AF.
Tel: 0207 848 5094. Fax: 0207 848 0333. E-mail: [email protected]
ISSN 0954–0261 print/ISSN 1369–1627 online ß 2011 Institute of Psychiatry
DOI: 10.3109/09540261.2010.545989
Int Rev Psychiatry Downloaded from informahealthcare.com by Kings College London on 04/19/11
For personal use only.
62
Z. Fortune et al.
Until recently there has been a deficiency in
health service provision for individuals with PD
(National Institute for Mental Health in England,
2003). In the UK this was recently addressed by the
development of new community-based services for
adults with PD (Crawford et al., 2008; Crawford
et al., 2009) and specialist forensic services for
offenders with PD, as part of the Dangerous and
Severe Personality Disorder (DSPD) programme
established by the Ministry of Justice and
Department of Health (Home Office, HPS, DoH,
2005). The DSPD programme provoked a great deal
of controversy, chiefly because of concerns relating to
the ethical ramifications of preventatively detaining
people with personality disorders (Moran, 2002;
Mullen, 1999). The programme led to the establishment of new high secure facilities at four establishments and three new non-high secure forensic
services. The aims of these services were to provide
treatments to reduce the risk of re-offending, address
mental health needs and improve social functioning
of personality-disordered offenders. The non-high
secure services were systematically evaluated by our
research team. In this paper, we report on the clinical
and economic outcomes from a two-year prospective
study of a cohort of service users recruited from these
services.
All three services only treated men with a primary
diagnosis of PD who were aged between 18–65 years
and whose IQ was greater than 70. All service users
were eligible to participate in the study.
Procedure
Ethical and R&D approval was obtained from the
local Research Ethics Committee. Having obtained
their written informed consent, each participant was
asked to complete the following three items:
1.
2.
Method
Sample and setting
To retain confidentiality, we refer to the services as
service 1, 2 and 3.
Service 1 comprised an inpatient medium secure
unit (MSU) and a residential service managed by
a local housing organization. The inpatient unit
consisted of two 10-bedded wards run along therapeutic community principles. The residential service
provided social care for eight residents, and assisted
in exploring local opportunities for education,
employment and other activities.
Service 2 comprised an inpatient MSU, and
a community team. The inpatient unit was a
16-bedded ward that ran a cognitive behavioural
treatment programme, incorporating both individual
and group treatments. The community team offered
an assessment and treatment programme aimed at
reducing risk of harm to others.
Service 3 comprised an inpatient MSU, a community team and a residential service, consisting of two
hostels. The inpatient unit consisted of a 15-bedded
ward and the hostels provided support for 10 residents. The service aimed to provide integrated care
across the three service components, using a treatment programme based on the Violence Reduction
Programme (Wong, Gordon, & Gu, 2007).
3.
Work and Social Adjustment Scale (WSAS;
(Mundt, Marks, & Shear, 2002). The WSAS is
a 5-item measure of work and social adjustment; each item can have a possible score of
0 to 8, with a total possible score of 40. Higher
scores indicate a higher level of impairment
in social functioning.
Working Alliance Inventory (WAI) (Horvath &
Greenberg, 1989). The WAI is a 12-item selfreport questionnaire, in which the respondent is
directed to rate their alliance with a nominated
member of staff on a 7-point Likert scale. It is
reliable and has good internal consistency.
Scores range from 0 to 168, with higher
scores indicating a better working alliance.
Nominated staff were also asked to complete
the WAI, rating their relationship with the
service user.
The Secure Facilities Service Use Schedule
(SF-SUS; (Barrett & Byford, 2007). This
schedule collects information on the service
user’s accommodation, including time spent in
a secure facility such as prison or secure NHS
unit, use of all health, social, voluntary sector
services, psychotropic medication and contact
with the police, lawyers and the courts. The
SF-SUS does not record routine care such
as nursing contacts on an inpatient ward.
Although all the service users in the study
were aligned to one of the MSU, residential or
community services, they continued to access
health, social, voluntary and criminal justice
services and the SF-SUS also collected information on all such contacts. At baseline, the
SF-SUS was used to collect service use data
for the 6 months preceding the interview.
In addition, the following background information
was obtained from case records: socio-demographic
information; highest education level attained; mean
full scale IQ (if recorded); psychiatric history including rating of personality disorder as formalized by
the International Personality Disorder Examination
(Loranger, Sartorius, & Janca, 1996); criminal history. Case records were also scrutinized for
Int Rev Psychiatry Downloaded from informahealthcare.com by Kings College London on 04/19/11
For personal use only.
Clinical and economic outcomes from the UK pilot psychiatric services
documentation of service contacts in order to supplement self-report SF-SUS data.
Interviews with service users and case record
searches were repeated at 6 and 24 months post
baseline interview. Participants’ case records were
searched for new incidents of behavioural disturbance occurring over the follow-up period. A positive
rating of behavioural disturbance was recorded if a
service user had been noted to engage in any of the
following: violence, self-harm, substance misuse,
alcohol misuse, and absconding behaviour. Case
record ratings were supplemented by information
supplied by staff. For each type of behaviour, reports
were combined across rating source and a positive
score recorded if indicated by either case records
or staff.
Data analysis
All analyses were conducted using SPSS version 15.
Participants were grouped according to whether they
were being treated by MSUs (n ¼ 30) or community/
residential services (n ¼ 24). Simple descriptive
statistics were used to describe the demographic,
criminal and psychiatric profile of the two groups.
Paired t-tests were used to assess for the presence of
statistically significant changes in social functioning
and working alliance. Pearson correlation coefficients
were used to assess similarities between staff
and service user scores on the working alliance.
Independent sample t-tests, Fisher’s exact tests and
Mann-Whitney U tests were used to assess the
significance of associations between baseline characteristics and whether service users had engaged
in new incidents of behavioural disturbance.
Costs
All costs were for the financial year 2005–2006.
The costs of time spent in the specialist PD services
were calculated using standard costing methodology
(Knapp & Beecham, 1990) and based on the
expenditure of each service and information from
managers on staff, overheads and capital costs.
Trust-specific costs for NHS hospital contacts were
sourced from NHS Reference Costs (Department of
Health, 2007) and community health and social
service costs were taken from national publications
(Curtis & Netten, 2007). The cost of medications
was calculated using the British National Formulary
(Royal Pharmaceutical Society of Great Britain,
2008). Contacts with criminal justice agencies were
costed using national publications and charging
information (HM Prison Service, 2007); Legal
Services Commission, 2004). The range of services
used by services users in each site was examined in
a descriptive analysis. Total costs were calculated
63
by the service-providing sector and we compared the
cost for the six months preceding baseline and the six
months follow-up using paired sample t-tests.
Results
Baseline characteristics of the sample
Recruitment to the study took place between
November 2005 and July 2006. During this period
89 service users were managed by the three services
(13 in service 1, 50 in service 2 and 26 in service 3),
of whom 54 were recruited to the study; an overall
response of 61% (12 from service 1, 20 from service
2 and 22 from service 3). A total of 35 eligible
participants did not participate: four refused and
27 did not respond to repeated attempts to contact
them, the majority of whom were from the community component of service 2. A further four participants were classified by staff as being too dangerous
to interview.
The mean age of entry into the services was 36.8
years (SD 8.3). Socio-demographic characteristics
of the sample are displayed in Table 1.
Some 91% of the sample had a history of violence
with 13% having convictions for homicide and 80%
convictions for other violent offences. Nearly two
thirds of participants had previously hurt someone so
badly that the victim had required hospital treatment.
Convictions were obtained for a variety of other
offences including sexual offences (31% of the entire
sample), acquisitive crime (74% of the entire sample)
and drug offences (28% of the entire sample). Index
offences (those for which the participant received
their current hospital term or prison sentence)
included manslaughter, attempted murder, rape,
buggery, assault and kidnapping. In some cases, the
violence was characterized by extreme cruelty; for
example, one participant had abducted and subsequently tortured a child by stubbing out cigarettes on
the victim’s face and pulling out the victim’s finger
nails.
As anticipated, service users in the MSUs had a
more severe criminal profile and a younger age of
first offending compared to those being managed in
the community. Mean age at first conviction was
18.6 years (SD 7.74, range 11–54) overall, 16.8 years
in the MSU group (SD 4.98, range 11–32) and 20.9
years (SD 7.74, 11–54) in the community group.
The mean number of prior convictions was 13.9
(range 0–75) overall, 15.0 (range 0–75) in the MSU
group and 12.4 (range 0–51) in the community
group. The mean number of prior offences was 28.9
(range 0–147) overall, 34.4 (range 0–147) in the
MSU group and 21.9 (range 2–50) in the community
group.
64
Z. Fortune et al.
Table 1. Characteristics of the recruited sample.
Int Rev Psychiatry Downloaded from informahealthcare.com by Kings College London on 04/19/11
For personal use only.
Variable
MSU
N ¼ 30
Admission source
High security
7
Medium security
5
PICU
1
Prison
17
Court
0
Home/community
0
Other
0
Ethnicity
White
21
Black
6
Other
3
Marital status
Single
22
Married
0
Divorced/separated
7
Widowed
1
Employment status prior to current sentence
Employed paid
5
Part-time paid
1
Unemployed/seeking work
24
Accommodation immediately prior to hospitalization/incarceration
Council housing
7
Private rented
4
Homeless
6
Absconded from psychiatric setting
1
Living with friends or family
9
Other
3
Highest education attainment
Did not complete secondary school
18
Finished secondary school with no qualifications
6
Finished secondary school with some qualifications
5
Other
1
IPDE diagnosis (N ¼ 44)
Schizoid
1
Paranoid
0
Dissocial
15
Emotionally unstable
4
Anxious
1
Dependent
1
Unspecified
1
Not given
2
Section of the 1983 Mental Health Act
Informal
0
37/41
7
47/49
15
Other
8
Of the entire sample 83% had a documented
history of child abuse, the most prevalent form of
which was physical abuse (69% of the entire sample),
followed by psychological abuse (52% of the entire
sample) and sexual abuse (39% of the entire sample).
Of the entire sample 80% had a previous history
of having been bullied and 59% had a history of
bullying others. Full scale IQ ratings were carried out
on 44 of the 54 participants. The mean IQ score of
these 44 individuals was 89.9 (SD 11.57) and full
scale IQ was slightly higher in the MSU group of
Community &
residential N ¼ 24
Overall
N ¼ 54 (%)
Overall
percentage
2
4
0
0
1
14
3
9
9
1
17
1
14
3
17
17
2
31
2
26
5
22
2
0
43
8
3
80
15
5
17
0
7
0
39
0
14
1
72
0
26
2
0
1
23
5
2
47
9
4
87
16
2
0
0
2
4
23
6
6
1
11
7
43
11
11
2
20
13
9
8
5
2
27
14
10
3
50
26
19
5
0
1
8
6
0
0
1
3
1
1
23
10
1
1
2
5
2
2
52
23
2
2
5
11
16
4
0
4
16
11
15
12
30
20
28
22
service users (92.9, SD 10.68, range 72–111) compared to those in the community (86.19 (SD 11.87,
range 67–110).
Of the entire sample 76% had a history of self
harm with 57% requiring hospital treatment at
some point for their self-inflicted injuries. A history
of substance misuse was also highly prevalent
across the entire sample (91%), together with a
history of alcohol misuse (74%), non-adherence
with medication (76%) and absconding behaviour (39%).
Clinical and economic outcomes from the UK pilot psychiatric services
30 inpatients (14 discharged)
9 residential service (3 discharged)
65
15 community (7 discharged)
5
1
1
Medium security
Police
Probation
1
5
Prison
1
1
Unknown
2
2
4
Successful discharge
Referring team due to
non-engagement
High security
1
Chose to leave
Int Rev Psychiatry Downloaded from informahealthcare.com by Kings College London on 04/19/11
For personal use only.
Figure 1. Discharge setting for the 24 service users discharged by two-year follow-up.
Table 2. Mean functional impairment scores at baseline compared with follow-up.
Mean WSAS score
MSU N ¼ 30
Community & residential N ¼ 24
Total N ¼ 54
Mean WSAS score at baseline (SD)
Mean WSAS score at 6 months (SD)
Paired t-tests between baseline and 6 months
(based on N ¼ 42)
Mean WSAS score at 24-month follow-up
Paired t-tests between baseline and 24 months
(based on N ¼ 25)
13.67 (9.08)
13.0 (9.05)
T ¼ 0.756
p ¼ 0.46
10.17 (9.97)
T ¼ 0.81
p ¼ 0.43
20.42 (12.12)
19.53 (10.97)
T ¼ 0.81
p ¼ 0.43
14.5 (8.3)
T ¼ 1.04
p ¼ 0.33
16.67 (10.96)
15.64 (10.23)
T ¼ 1.11
p ¼ 0.27
11.9 (9.41)
T ¼ 1.24
p ¼ 0.23
Characteristics at follow-up
At 6 months post baseline assessments, 11 of the
54 participants (20%) had been discharged from
the services for a variety of reasons and 42 of the
remaining 43 participants were successfully followed
up. By 24 months, 24 participants had been
discharged from the services. Therefore 30 participants remained with the services over the entire
24-month study period. It was not possible to follow
up five of these individuals; one community participant did not respond to repeated attempts to contact
him and a further four participants were considered
by staff to be too unstable to participate. Therefore,
25 participants were followed up at 24-month followup. There were no statistically significant differences
in the baseline characteristics of participants who
were followed up compared with those who were
not followed up. The average length of time that
participants had been with the services prior to
discharge was 18.9 months.
The discharge settings for the 24 service users
discharged by two-year follow-up are displayed in
Figure 1.
Functional impairment scores as measured by the
WSAS at baseline and at both follow-up points are
shown below in Table 2. There was no significant
change in functional impairment score at either
6- or 24-month follow-up, compared with baseline.
Similarly, there was no significant change in working
alliance score between baseline and 6-month or
between baseline and 24-month follow-up in either
the service users ratings of staff or the staff ratings
of service users.
A wide range of serious behavioural incidents
occurred over the follow-up period. At six months,
these included threats to destroy the unit, possession
of drug paraphernalia and a serious assault perpetrated by a service user who attempted to set fire to
his partner using petrol. At 24 months the list of
incidents included possession and supplying of drugs
and a sexual assault.
Six of the 24 participants were engaged in more
than six serious behaviour incidents. Notably, these
service users had a significantly lower full scale IQ
score of 80 (SD 14.73) compared with the remaining
service users of 97.67 (SD 10.34; t ¼ 2.50
p ¼ 0.03). They were also significantly more functionally impaired at baseline (WSAS score of the
group with six or more incidents ¼ 23.17 versus
12.61 for remainder of sample; t ¼ 2.52, p ¼ 0.02);
at 6-month follow-up (WSAS score ¼ 25.17 versus
9.28 for remainder of sample; t ¼ 3.78, p ¼ 0.001);
and at 24-month follow-up (WSAS score ¼ 19.50
versus 10.03 for the remainder of the sample;
t ¼ 2.34, p ¼ 0.03).
66
Z. Fortune et al.
Int Rev Psychiatry Downloaded from informahealthcare.com by Kings College London on 04/19/11
For personal use only.
The economic costs of treatment
Aggregate and six-month cost. Cost data was obtained
on 48 of the 54 recruited participants. The aggregate
costing analysis found the cost per service user per
year for the MSU services was between £192,978
and £199,696, a cost slightly higher than a nonspecialist MSU (National Health Service £155,597,
private sector provider £168,015), but below that of
the high secure PD services (£226,455) (personal
communication with St Andrew’s Healthcare). Many
service users had come from the criminal justice
services and this type of NHS provision costs
substantially more than a prison place (£21,976 to
£43,904) (HM Prison Service, 2007). The residential service costs were between £111,943 and
£162,752 per year. These costs are substantially
higher than existing estimates of the cost of specialist
hostels such as bail hostels (£27,916 sourced from
the Audit Commission), though it is important to
note that the residential services include the hostel
provision and supervision from the specialist teams at
the MSUs. The estimates of the cost per place
depend on the occupancy levels of the services and
all the above costs are based on the assumption that
all places were filled at all times. During the
evaluation period this was not the case, and therefore
the cost per service user was in fact considerably
higher for all services. The service use data for the
sample of service users from the three sites showed a
range of primary and secondary health care contacts.
These data are displayed in Table 3.
Service users had hospital appointments and
occasional admissions for a range of health problems.
There was good evidence that key workers and
mental health professionals had regular contact with
those in the residential and community services.
In most cases over half the service users were on
some form of psychotropic medication, and the
proportion on medication in the community was
consistently over 50% by follow-up.
From the service use data we are also able to see
where the service users being supervised in the
community were accommodated. For the most part,
service users in the community lived in their own
homes (usually rented from the local authority or a
housing association) or in non-specialist hostel
accommodation. However, we found some evidence
of service users using less stable forms of accommodation such as bed and breakfast, staying with friends
and on one occasion sleeping rough.
An analysis of individual economic data showed
that although the specialist services were predominantly run by the NHS, there were also considerable
economic burdens to other service providers, notably
social services, who were responsible for social
housing for service users in the community and the
criminal justice system, for prison costs and any
court costs, which could be substantial. Mean total
costs for the six months preceding baseline were
£51,352 (SD £37,345) compared with mean total
costs of £68,855 (SD £39,637) for the six months
post baseline; these costs were significantly higher
at follow-up (p50.001).
Discussion
Our study provides a detailed descriptive account
of the clinical population being managed by new
forensic personality disorder services in the United
Kingdom. By spring 2007, all of the pilot services
were operational and were accepting and treating a
relatively homogenous group of extremely challenging service users. The majority of these service users
were single white men who came from disadvantaged
backgrounds. The majority had a diagnosis of dissocial PD and over a fifth had a diagnosis of emotionally unstable PD. Unsurprisingly, violence and sexual
offending were highly prevalent among the sample;
in many cases, the violence had been characterized
by extreme cruelty towards the victim. However,
other findings were less predictable. The majority of
service users had experienced childhood maltreatment and poor education, highlighting a life
course of disadvantage (Maxfield & Widom, 1996).
Substance abuse and self-harm were highly prevalent
in the sample and these findings suggest that those
being treated by the services were themselves vulnerable and posed an ongoing risk to themselves as
well as to society.
During the 24-month follow-up period, mean
functional impairment scores dropped (improved)
by an average of five points although no weight can
be attached to these findings due to the lack of a
statistically significant finding. Service users involved
in the greatest number of incidents of problematic
behaviour had greater levels of impairment in social
functioning, highlighting the fact that impairment
in this domain can be an enduring feature of all
personality disorders (Skodol et al., 2005). The
finding also highlights the absolute necessity of
assessing functional impairment when formulating a
treatment plan for those with a personality disorder.
Another unanticipated finding was that many of
the men were of comparatively low intelligence
(the mean full scale IQ in the recruited sample:
90). Those engaging in new episodes of behavioural
disturbance were also more likely to have lower
IQ scores. In interpreting this finding, it should
be emphasized that IQ data were inconsistently
recorded across the services and it is entirely possible
that IQ tests were only selectively carried out on
participants who were perceived to be of low intelligence. Although selection bias might explain the
Clinical and economic outcomes from the UK pilot psychiatric services
67
Int Rev Psychiatry Downloaded from informahealthcare.com by Kings College London on 04/19/11
For personal use only.
Table 3. Service use (number of contacts) at six-month follow-up.
Accommodation
Own home (nights)
Hostel (nights)
MSU (nights)
Low secure unit (nights)
Prison (nights)
High secure hospital (nights)
Bed and breakfast (nights)
Homeless (nights)
Staying with friends (nights)
Health and community services
Inpatient stay (nights)
Outpatient appointment (attendances)
Accident and emergency (attendances)
General practitioner (contacts)
Practice nurse (contacts)
Key worker (contacts)
Psychiatric nurse (contacts)
Psychiatrist (contacts)
Psychologist (contacts)
Counsellor/therapist (contacts)
Drug and alcohol worker (contacts)
Dentist (contacts)
Occupational therapist (contacts)
Social worker (contacts)
Day centre (visits)
Criminal justice services
Probation (contacts)
Solicitor (contacts)
Police (contacts)
Police custody (sessions)
Court appearance (per case)
MSU (N ¼ 27)
Residential (N ¼ 7)
Mean
SD
Mean
SD
Mean
SD
0.0
0.0
181.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
11.3
0.0
0.0
0.0
0.0
0.0
0.0
8.7
172.9
0.0
0.0
2.4
0.0
0.0
0.0
0.0
23.1
22.5
0.0
0.0
6.4
0.0
0.0
0.0
0.0
92.9
46.0
0.0
0.0
0.0
0.0
9.8
4.4
21.9
96.5
71.5
0.0
0.0
0.0
0.0
36.6
16.3
57.0
0.3
1.7
0.7
5.6
5.4
27.4
7.0
12.9
20.9
10.9
0.1
1.7
1.8
0.3
0.0
1.7
2.6
1.8
8.0
11.9
39.4
22.1
13.4
26.2
35.6
0.4
3.0
5.2
0.9
0.0
0.7
2.3
0.6
2.3
4.2
22.9
13.9
5.4
17.1
11.1
0.1
1.7
6.6
3.7
18.9
1.3
5.6
1.0
3.1
7.5
17.2
17.6
10.1
18.7
20.5
0.4
2.2
11.4
9.8
48.6
5.7
0.9
0.4
3.3
0.5
29.6
10.5
1.1
4.4
0.0
0.4
0.4
0.1
5.9
12.1
12.5
2.3
0.6
4.8
1.6
30.8
16.1
1.2
13.9
0.0
1.6
0.7
0.3
10.3
32.3
0.2
4.2
0.1
0.0
0.1
0.5
6.6
0.5
0.0
0.3
1.9
0.9
2.1
0.3
0.3
4.9
1.1
2.4
0.5
0.5
3.7
1.8
2.0
0.6
0.3
6.5
3.9
3.9
0.9
0.6
finding of low IQ, it remains possible that the mean
IQ of service users within the services is indeed low.
If this is the case, some service users might not
be intellectually capable of participating fully in a
psychological treatment programme which requires
the performance of complex cognitive tasks (Katz
et al., 2005). For this reason, we suggest that IQ
testing should be routinely carried out on personality-disordered offenders who are being considered
for treatment within mental health settings. This
would provide not only an assessment of the service
user’s intellectual ability to use treatment, but it
would also highlight whether there are intellectual
difficulties requiring specific attention. Conversely,
it would also highlight whether some service users
have intellectual skills which could be put to better
use as part of their rehabilitation.
The fact that 44% of the sample had disengaged
with the services at 24-month follow-up is concerning. Qualitative data from this study (Fortune et al.,
2010) indicates that engaging in talking treatment
can be very stressful for personality-disordered
offenders and this may offer one explanation for the
Community (N ¼ 14)
high rate of disengagement. Previous research has
suggested that offenders who commence a treatment
programme, but do not complete it, are at greater
risk of recidivism compared to those who never
commence treatment (McMurran & Theodosi,
2007). Clearly the detailed and lengthy assessment
process was failing to detect those at risk of noncompletion and this suggests that further refinement
of this procedure is urgently required.
An analysis of individual economic data showed
that although the specialist services were predominantly run by the NHS, there were also considerable
economic burdens to other service providers, notably
social services, who were responsible for social
housing of service users in the community and the
criminal justice system, for prison and any court
costs, which could be substantial. On the whole,
costs for the six months post baseline were significantly higher than the costs for the six months
preceding baseline, reflecting the greater amount
of time at follow-up in a more resource-intensive
service. Unsurprisingly, given the intensive and
specialist nature of the PD programme, the unit
Int Rev Psychiatry Downloaded from informahealthcare.com by Kings College London on 04/19/11
For personal use only.
68
Z. Fortune et al.
costs of the MSU services were higher than the cost
of similar mainstream services and substantially
higher than the cost per night in prison. The
residential service and community costs were also
higher than other hostel services, though it should be
noted that these are specialist services that include
supervision from specialist staff. Costs were based on
the assumption that all places were filled and the cost
per place increased substantially when occupancy
levels fell below capacity. This finding highlights the
importance of filling places within specialist services
and filling places quickly when service users have
been discharged.
Strengths of the study include the fact that we
collected data from both case records and service
users and where necessary filled gaps in the data by
speaking to staff. Detailed quantitative data were
gathered in multiple domains (health, criminal and
economic) at two time points. In addition, the rate
of attrition at follow-up was comparatively low
(less than 30% were lost to follow-up across all
three services). Nevertheless, our findings need to
be considered in the light of some important methodological limitations. Firstly, we recruited nonrandom samples of service users from each of the
three services and although we recruited almost
complete samples from two services, response for
the third service was poor (40%), due to difficulty
recruiting community-based service users living over
a very wide geographical area. The fact that we
surveyed non-random samples of service users raises
the possibility of selection bias. In addition, the
recruited sample size per service was small and we
only recruited a total of 54 service users from all
three services. Given this, our failure to detect any
statistically significant differences in social functioning at follow-up might be attributed to type 2
statistical error. However, one might expect the
targeting of a comparatively small population with an
expensive treatment to produce a large, readily
detectable clinical effect and our study is not the
first to report disappointing clinical outcomes from
some forensic services (Davies, Clarke, Hollin, &
Duggan, 2007). Clearly this issue warrants further
examination. Secondly, it did not prove possible to
follow up service users who had left the services, thus
raising the possibility of selection bias in following up
a relatively ‘well behaved’ cohort of service users.
Finally, for practical reasons, we were unable to
apply a study design that incorporated a control
group and are therefore unable to comment on the
likely effectiveness/ineffectiveness of the services.
Acknowledgements
We are indebted to the service users and staff of the
three services for their time and support. We are
grateful to Nick Benefield (Department of Health)
who was a member of the Project Advisory Group
and provided helpful information about the policy
context for the study.
Declaration of interest: This paper presents
independent research commissioned by the
National Institute for Health Research (NIHR)
Service Delivery and Organization (SDO) programme. The NIHR SDO programme is funded by
the Department of Health. The views expressed
in this paper are those of the authors and not
necessarily those of the NHS, the NIHR or the
Department of Health. The authors alone are
responsible for the content and writing of the paper.
References
Barrett, B., & Byford, S. (2007). Collecting service use data
for economic evaluation in DSPD populations: Development
of the Secure Facilities Service Use Schedule. British Journal of
Psychiatry, 190, S75–S78.
Barrett, B., Byford, S., Seivewright, H., Cooper, S., Duggan, C.,
& Tyrer, P. (2009). The assessment of dangerous and severe
personality disorder: Service use, cost, and consequences.
Journal of Forensic Psychiatry & Psychology, 20, 120–131.
British Psychological Society & Royal College of Psychiatrists
(2009). Antisocial Personality Disorder: The NICE Guideline
on Treatment and Management. Leicester & London:
British Psychological Society & Royal College of
Psychiatrists.
Coid, J., Hickey, N., Kahtan, N., Zhang, T., & Yang, M. (2007).
Patients discharged from medium secure forensic psychiatry
services: Reconvictions and risk factors. British Journal of
Psychiatry, 190, 223–229.
Coid, J., Yang, M., Roberts, A., Ullrich, S., Moran, P.,
Bebbington, P., . . . , Singleton, N. (2006). Violence and psychiatric morbidity in a national household population – A report
from the British Household Survey. American Journal of
Epidemiology, 164, 1199–1208.
Crawford, M.J., Price, K., Gordon, F., Josson, B., Taylor, B.,
Bateman, A., . . . , Moran, P. (2009). Engagement and retention
in specialist services for people with personality disorder. Acta
Psychiatrica Scandinavica, 119, 304–311.
Crawford, M.J., Price, K., Rutter, D., Moran, P., Tyrer, P.,
Bateman, A., . . . , Weaver, T. (2008). Dedicated communitybased services for adults with personality disorder: Delphi
study. British Journal of Psychiatry, 193, 342–343.
Curtis, L., & Netten, A. (2007). Unit Costs of Health and
Social Care 2006. Canterbury: Personal Social Services
Research Unit.
Davies, S., Clarke, M., Hollin, C., & Duggan, C. (2007).
Long-term outcomes after discharge from medium secure
care: A cause for concern. British Journal of Psychiatry, 191,
70–74.
Department of Health. (2007). NHS Reference costs 2006.
London: Department of Health.
Fazel, S., & Danesh, J. (2002). Serious mental disorder in 23,000
prisoners: A systematic review of 62 surveys. Lancet, 359,
545–550.
Fortune, Z., Rose, D., Crawford, M., Slade, M., Spence, R.,
Mudd, D., . . . , Moran, P. (2010). An evaluation of new services
for personality-disordered offenders: Staff and service user
Int Rev Psychiatry Downloaded from informahealthcare.com by Kings College London on 04/19/11
For personal use only.
Clinical and economic outcomes from the UK pilot psychiatric services
perspectives. International Journal of Social Psychiatry, 56,
186–195.
Gorwood, P., Rouillon, F., Even, C., Falissard, B., Corruble, E.,
& Moran, P. (2010). Treatment response in major depression:
Effects of personality dysfunction and prior depression. British
Journal of Psychiatry, 196, 139–142.
Grant, B.F., Stinson, F.S., Dawson, D.A., Chou, S.P.,
Ruan, W.J., & Pickering, R.P. (2004). Co-occurrence of
12-month alcohol and drug use disorders and personality
disorders in the United States: Results from the National
Epidemiologic Survey on Alcohol and Related Conditions.
Archives of General Psychiatry, 61, 361–368.
Haw, C., Hawton, K., Houston, K., & Townsend, E. (2001).
Psychiatric and personality disorders in deliberate self-harm
patients. British Journal of Psychiatry, 178, 48–54.
Hayward, M., & Moran, P. (2008). Comorbidity of personality
disorders and mental illnesses. Psychiatry, 7, 105–109.
Hayward, M., Slade, M., & Moran, P.A. (2006). Personality
disorders and unmet needs among psychiatric inpatients.
Psychiatric Services, 57, 538–543.
HM Prison Service (2007). Prison Service Annual Report and
Accounts 2007. London: The Stationery Office.
Home Office, HPS, DoH (2005). Dangerous and Severe Personality
Disorder (DSPD) High Secure Services: Planning and Delivery
Guide. London: Home Office, HM Prison Service, Department
of Health.
Horvath, A.O., & Greenberg, L.S. (1989). Development and
validation of the Working Alliance Inventory. Journal of
Counselling Psychology, 36(2), 223–233.
Huang, Y., Kotov, R., de Girolamo, G., Preti, A.,
Angermeyer, M., Benjet, C., . . . , Kessler, R.C. (2009).
DSM-IV personality disorders in the WHO World Mental
Health Surveys. British Journal of Psychiatry, 195, 46–53.
Katz, E.C., King, S.D., Schwartz, R.P., Weintraub, E.,
Barksdale, W., Robinson, R., & Brown, B.S. (2005).
Cognitive ability as a factor in engagement in drug abuse
treatment. American Journal Drug Alcohol Abuse, 31, 359–369.
Knapp, M., & Beecham, J. (1990). Costing mental health services.
Psychological Medicine, 20, 893–908.
Legal Services Commission (2004). General Criminal Contract.
London: Legal Services Commission.
Loranger, A.W., Sartorius, N., & Janca, A. (1996). Assessment and
Diagnosis of Personality Disorders: The International Personality
Disorder Examination (IPDE). New York: Cambridge
University Press.
Maxfield, M.G., & Widom, C.S. (1996). The cycle of violence.
Revisited 6 years later. Archives of Pediatrics and Adolescent
Medicine, 150, 390–395.
McCrone, P., Dhanasiri, S., Patel, A., Knapp, M., & LawtonSmith, S. (2008). Paying the Price: The Cost of Mental Healthcare
in England to 2026. London: King’s Fund.
69
McMurran, M., & Theodosi, E. (2007). Is treatment noncompletion associated with increased reconviction over no
treatment? Psychology Crime & Law, 13, 333–343.
Moran, P. (2002). Dangerous severe personality disorder – Bad
tidings from the UK. International Journal of Social Psychiatry,
48, 6–10.
Moran, P., Coffey, C., Mann, A., Carlin, J.B., & Patton, G.C.
(2006). Personality and substance use disorders in young
adults. British Journal of Psychiatry, 188, 374–379.
Moran, P., Stewart, R., Brugha, T., Bebbington, P., Bhugra, D.,
Jenkins, R., & Coid, J.W. (2007). Personality disorder and
cardiovascular disease: Results from a national household
survey. Journal of Clinical Psychiatry, 68, 69–74.
Mullen, P.E. (1999). Dangerous people with severe personality
disorder. British proposals for managing them are glaringly
wrong-and unethical. British Medical Journal, 319, 1146–1147.
Mundt, J.C., Marks, I.M., & Shear, M.K. (2002). The Work and
Social Adjustment Scale: A simple measure of impairment in
functioning. British Journal of Psychiatry, 180, 461–464.
National Institute for Mental Health in England (2003).
Personality Disorder: No Longer a Diagnosis of Exclusion. Policy
Implementation Guidance for the Development of Services for People
with Personality Disorder. London: Department of Health.
Newton-Howes, G., Tyrer, P., & Johnson, T. (2006). Personality
disorder and the outcome of depression: Meta-analysis of
published studies. British Journal of Psychiatry, 188, 13–20.
Rendu, A., Moran, P., Patel, A., et al. (2002). Economic impact of
personality disorders in UK primary care attenders. British
Journal of Psychiatry, 181, 62–66.
Royal Pharmaceutical Society of Great Britain (2008). British
National Formulary 55. London: BMJ Group and RPS
Publishing.
Singleton, N., Meltzer, H., & Gatward, R. (1998). Psychiatric
Morbidity Among Prisoners in England and Wales. London:
Stationary Office.
Skodol, A.E., Pagano, M.E., Bender, D.S., Shea, M.T.,
Gunderson, J.G., Yen, S., . . . , McGlashan, T.H. (2005).
Stability of functional impairment in patients with schizotypal,
borderline, avoidant, or obsessive–compulsive personality disorder over two years. Psychological Medicine, 35, 443–451.
Steadman, H.J., Mulvey, E.P., Monahan, J., Clark Robbins, P.,
Applebaum, P.S., Grisso, T., . . . , Silver, E. (1998). Violence by
people discharged from acute psychiatric inpatient facilities and
by others in the same neighborhoods. Archives of General
Psychiatry, 55, 393–401.
Tyrer, P., Mulder, R., Crawford, M., Newton-Howes, G.,
Simonsen, E., Ndetei, D., . . . , Barrett, B. (2010). Personality
disorder: A new global perspective. World Psychiatry, 9, 56–60.
Wong, S.C., Gordon, A., & Gu, D. (2007). Assessment and
treatment of violence-prone forensic clients: An integrated
approach. British Journal of Psychiatry Supplement, 49, 66–74.