See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/328698498

Beyond the Walls: An Evaluation of a Pre-Release Planning (PReP) Programme

for Sentenced Mentally Disordered Offenders

Article  in  Frontiers in Psychiatry · November 2018

DOI: 10.3389/fpsyt.2018.00549

CITATIONS READS

0 88

14 authors, including:

Damian Smith Sarah Hennessy

Trinity College Dublin Ireland's Health Services
6 PUBLICATIONS   67 CITATIONS    2 PUBLICATIONS   5 CITATIONS   

SEE PROFILE SEE PROFILE

Pauline Gill Cornelia Carey

National Forensic Mental Health Service Central Mental Hospital Dundrum Dublin … Royal College of Surgeons in Ireland
4 PUBLICATIONS   11 CITATIONS    11 PUBLICATIONS   0 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

schizophrenia, neurocognition and violence View project

A cross platform risk assessment mobile application for precise data collection View project

All content following this page was uploaded by Damian Smith on 02 November 2018.

The user has requested enhancement of the downloaded file.

 ORIGINAL RESEARCH
published: 02 November 2018
doi: 10.3389/fpsyt.2018.00549

Beyond the Walls: An Evaluation of a

Pre-Release Planning (PReP)
Programme for Sentenced Mentally
Disordered Offenders
Damian Smith 1,2 , Susan Harnett 1 , Aisling Flanagan 1 , Sarah Hennessy 1,2 , Pauline Gill 1 ,
Niamh Quigley 1 , Cornelia Carey 1 , Michael McGhee 1 , Aoife McManus 1 , Mary Kennedy 3 ,
Enda Kelly 3 , Jean Carey 3 , Ann Concannon 1 , Harry G. Kennedy 1,2 and Damian Mohan 1,2*
1
National Forensic Mental Health Service, Central Mental Hospital, Dublin, Ireland, 2 Department of Psychiatry, Trinity College,
Dublin, Ireland, 3 Irish Prison Service, Dublin, Ireland

Edited by:
Norbert Konrad, Background: Prison mental health services have tended to focus on improving the
Charité Universitätsmedizin Berlin, quality of care provided to mentally disordered offenders at the initial point of contact
Germany
with the prison system and within the prison environment itself. When these individuals
Reviewed by:
Manuela Dudeck, reach the end of their sentence and return to the community, there is an increased risk of
Universität Ulm, Germany morbidity, mortality, homelessness and re-imprisonment. New models of care have been
Birgit Angela Völlm,
developed to minimize these risks.
University of Rostock, Germany
Stelios Panagiotis Kympouropoulos, Objectives: The objective of this project was to establish a Pre-Release Planning
University General Hospital Attikon,
Greece
(PReP) Programme with social work expertise, to enhance interagency collaboration and
*Correspondence:
improve continuity of care for mentally disordered offenders upon their release. We aimed
Damian Mohan to evaluate the first 2 years of the programme by measuring its success at improving
[email protected]
the level of mental health support and the security and quality of accommodation
Specialty section:
achieved by participants upon release in comparison to that reported at time of
This article was submitted to imprisonment. Additionally, we aimed to explore the impact of these outcomes on rates
Forensic Psychiatry,
of re-imprisonment.
a section of the journal
Frontiers in Psychiatry Methods: A process of participatory action research was used to develop and evaluate
Received: 11 July 2018 the first 2 years of the programme. This was a naturalistic prospective observational
Accepted: 12 October 2018
whole cohort study.
Published: 02 November 2018

Citation: Results: The PReP Programme supported 43 mentally disordered offenders,

Smith D, Harnett S, Flanagan A, representing 13.7%, (43/313) of all new assessments by the prison’s inreach
Hennessy S, Gill P, Quigley N,
Carey C, McGhee M, McManus A,
mental health service during the 2 years study period. When compared with that
Kennedy M, Kelly E, Carey J, reported at time of reception at the prison, gains were achieved in level of mental
Concannon A, Kennedy HG and
health support (FET p < 0.001) and security and quality of accommodation (FET
Mohan D (2018) Beyond the Walls: An
Evaluation of a Pre-Release Planning p < 0.001) upon release. Of those participants seen by the PReP Programme,
(PReP) Programme for Sentenced 20 (46.5%, 20/43) were returned to prison during the 2-years study period.
Mentally Disordered Offenders.
Front. Psychiatry 9:549.
There was no significant relationship between re-imprisonment and gains made
doi: 10.3389/fpsyt.2018.00549 in mental health support (FET p = 0.23) or accommodation (FET p = 0.23).

Frontiers in Psychiatry | www.frontiersin.org 1 November 2018 | Volume 9 | Article 549

Smith et al. Beyond the Walls: A Pre-Release Planning Programme

Conclusions: We have shown that compared to that reported at time of reception

at prison, the level of mental health support and the security of tenure and quality
of accommodation both improved upon release following the intervention of the
programme. Improved mental health support and accommodation were not associated
with lower rates of re-imprisonment.

Keywords: prison, mental health, homeless, continuity of care, transition, participatory action research

INTRODUCTION teams, and early identification of needs prior to release, can

promote involvement of community based supports and assist
Prevalence rates for severe and enduring mental illnesses are in achieving continuity of care (22–24). These recommendations
significantly higher among sentenced prisoners than their peers are echoed in Human Rights legislation. Of particular relevance
in the general population (1–3). Mentally disordered offenders is Rule 107 of the United Nations Standard Minimum
tend to have more complex health and social needs than non- Rules for the Treatment of Prisoners (The Nelson Mandela
mentally disordered offenders (4, 5). Rules), which highlights the importance of maintaining or
Over the last decade, our service has developed a number of establishing “relations with persons or agencies outside the prison
initiatives aimed at addressing the needs of mentally disordered as may promote the prisoner’s rehabilitation” (25). However,
offenders in remand (6, 7) and sentenced (8) prisons. These efforts to establish and maintain relations with “persons or
projects have been successful in improving the quality of care agencies outside the prison” can be challenging. The double
provided to these individuals at the initial point of contact with stigma of being mentally ill and a convicted offender, along
the prison system and within the prison environment itself. with high rates of substance misuse and homelessness (5,
The immediate post-release period however, is a time 26), can act as barriers to engagement with community
which poses increased risks for all prisoners, but especially based healthcare. It could also be argued that due to the
those with a history of mental illness (9), including an complex social needs of mentally disordered offenders, that
increased risk of morbidity, mortality and homelessness (10– coordination of robust and holistic care plans should routinely
12). Moreover, in the context of the current homeless and be incorporated into prison inreach mental health services
housing crisis (13, 14) this vulnerable group are likely to be (27).
further marginalized and exposed to these adverse outcomes. Various models have been proposed to overcome these
Rates of re-imprisonment are high for all offenders both challenges, most of which involve case management in the pre-
in Ireland (15) and worldwide (16). In relation to those and post- release periods for varying amounts of time (24).
offenders with a mental illness, rates of re-imprisonment Assertive Community Treatment (ACT) has been utilized to
are increased when compared with non-mentally disordered provide intensive case management for up to 1 year in the post-
offenders (17). release period (28). This intervention tends to be expensive and
When prisoners near the end of their sentence, a number therefore more time limited approaches have been developed.
of potential supports are available to them both internal and Mckenna et al. have shown that a time limited intervention
external to the prison. These are provided by the criminal in the pre-release period based on the principles of ACT can
justice and public health systems, as well as non-governmental improve engagement with community mental health services in
organizations and the person’s family network. These supports the post-release period (29).
however, are typically fragmented and independent of one Critical Time Intervention (CTI) is a holistic approach to
another, risking the individual falling through the gaps between case management in the pre- and post-release period, which has
services upon their release (18). demonstrated benefits in assisting mentally disordered offenders
The World Health Organization has outlined a framework to engage with healthcare supports in the post-release period
for patient-centered, integrated healthcare provision (19). This (22, 30, 31). CTI case managers aim to establish effective and
model emphasizes the need for collaboration between agencies trusting relationships with service users prior to their release
and disciplines to improve patient outcomes and experiences, from an institution in order to identify and ameliorate potential
particularly for those with complex needs. These principles barriers to community reintegration (32, 33). Thereafter, they
have been embedded in healthcare policy across the UK (20) provide a time-limited period of support in the post-release
and Ireland (21). Despite their complex healthcare needs, period to help achieve transfer of care. In a randomized control
programmes for prison populations are conspicuous by their trial of CTI within a prison setting, Jarrett et al. reported that
absence in these clinical strategies. It has been suggested that the majority of the case manager’s work in establishing support
enhanced coordination between medical and mental health systems was performed within the prison, prior to the prisoner’s
release. Jarrett et al. also suggested that social workers may be best
placed to fulfill the role of case manager due to the complex social
Abbreviations: PReP, pre-release planning; CMHT, community mental health
team; GP, general practitioner; CTI, critical time intervention; ACT, assertive problems faced by these individuals and the knowledge of local
community treatment. services and agencies needed to engage community supports (22).

Frontiers in Psychiatry | www.frontiersin.org 2 November 2018 | Volume 9 | Article 549

Smith et al. Beyond the Walls: A Pre-Release Planning Programme

The objective of this project was to establish a new Pre- social workers, the PReP programme was supplemented by other
Release Planning (PReP) programme involving case management members of the existing Mountjoy Prison Inreach Mental Health
by mental health social workers, to enhance interagency Service, which included two full time community forensic mental
collaboration and improve continuity of care for sentenced health nurses, a visiting consultant forensic psychiatrist, and 1–2
mentally disordered offenders as they transition from prison to visiting psychiatric trainees.
the community. This was a naturalistic prospective observational whole cohort
We aimed to evaluate the first 2 years of the PReP Programme study. The intervention of the programme was provided to all
by measuring its success at improving health and social outcomes individuals on the inreach mental health service caseload within
for released mentally disordered offenders. In particular we 12 months of their earliest date of release. Since its inception,
aimed to explore for gains achieved in the level of mental health the key interventions of the programme have evolved based upon
support and the security and quality of accommodation achieved feedback received from service users and family members at pre-
by participants upon release in comparison to that reported at release planning (PReP) meetings held prior to an individual’s
time of imprisonment. Finally, we aimed to explore the impact of release. In addition stakeholders were afforded the opportunity to
these outcomes on rates of re-imprisonment. participate in critical reflection at weekly multiagency meetings.

METHODS Interventions of the PReP Programme:

1. Establishing trusting professional relationships with
Setting mentally disordered offenders in the pre-release period.
This study took place in Ireland’s oldest penal institution,
2. Liaison with mental health and other support agencies—
Mountjoy Prison, which was opened in 1850. Mountjoy Prison
Establishing or maintaining relationships with community
is a closed, medium secure prison for adult males, and is the
based mental health teams and other support agencies
main committal prison for sentenced prisoners in Dublin city
including: general practitioners, addiction services,
and county. It has capacity for 630 prisoners. The prison complex
intellectual disability services, accommodation providers,
consists of the main prison, a training unit and a 10-bed High
homeless support agencies and vocational programmes.
Support Unit (8).
3. Advocacy—Addressing queries and concerns raised by
community based mental health teams and other support
Study Design agencies. In addition the programme advocated on behalf of
A process of participatory action research was chosen to design
participants to ensure social welfare payments and medical
and develop the PReP Programme. Action research is described
payment schemes were in place upon their release.
as a process involving a spiral of steps, each of which is composed
4. Family support—Providing information regarding
of a cycle of planning, action and critical reflection (34). This
diagnosis, treatment needs and relapse prevention.
process can result in organizational change and development.
Exploring risks concerning the person on their return
The authors have previously used this method to develop prison
to the community including child protection issues and
inreach mental health services (7, 8).
suitability of accommodation. This was of particular relevance
The initial “planning” step involved a literature review and
for participants who planned to live with a family member on
was followed by an iterative process of identifying and consulting
their release.
stakeholders then drafting and re-drafting the new model of
5. Release planning—Coordinating robust, holistic care plans
care until there was sufficient support for the change process
prior to the person’s release from custody. In most cases
to proceed. Stakeholders included managers from the National
release plans were informed by multiagency, multidisciplinary
Forensic Mental Health Service (a specialist tertiary mental
pre-release planning (PReP) meetings held within 1 month of
health service funded and managed by the state health service),
the person’s release from prison. Figure 1 displays examples
the Irish Prison Service, Probation Services, community based
of the various stakeholders invited to attend PReP meetings.
homeless support agencies, service users (prisoners availing of
There was no statutory requirement for any stakeholder to
the support of the existing prison inreach mental health service)
attend pre-release meetings. Written release plans containing
and their families. This series of stakeholder meetings and
details of all relevant supports, contact details of key persons
consultations led to the interactive development of a protocol
in the community, and accommodation arrangements were
for case finding and engagement, multi-agency liaison and
provided to all participants supported by the programme.
interventions including the need for an integrated approach to
6. Post-release support—Providing time limited telephone
release planning for mentally disordered offenders. Given the
support for service users, family members and receiving
complex mental health and social needs of these individuals,
services, to ensure adequate handover and aid transition of
social work expertise was identified as a vital, yet missing
care.
component of the exisiting inreach mental health service.
7. Service evaluation through data collection and analysis.
Subsequently, in March 2015, a social worker was redeployed
from inpatient services at the National Forensic Mental Health
Service, and the PReP Programme was established. A second Referral Process and Participants
social worker was added in November 2015 providing a 1.5 full During the study period, referrals to the Moutjoy Inreach Mental
time equivalent resource. Although case management was led by Health Service were received through multiple sources.

Frontiers in Psychiatry | www.frontiersin.org 3 November 2018 | Volume 9 | Article 549

Smith et al. Beyond the Walls: A Pre-Release Planning Programme

FIGURE 1 | Examples of stakeholders invited to attend Pre-Release Planning (PReP) meetings prior to the individual’s release. CMHT, Community Mental health
Team; IPS, Irish Prison Service; GP, General Practitioner.

Upon reception at the prison, all newly received prisoners prison psychology, prison general practitioner, and chaplaincy.
were screened by prison general nursing staff for a history of Finally, family members and prisoners themselves also initiated
mental illness, active signs of mental illness and risk of harm referrals.
to self or others. If a need for increased levels of observation In the first instance all new referrals were assessed by
was identified, the prisoner could be placed directly in the high the inreach mental health service’s community forensic mental
support unit. In the event of a prisoner being placed in the high health nurses then triaged at weekly multi-agency meetings and
support unit, members of the Moutjoy Prison Inreach Mental appropriate follow up arranged.
Health Service aimed to assess them on the following working As the PReP Programme social workers were an integral part
day. of the Moutjoy Inreach Mental Health Service no formal referral
All new committals were assessed by a prison general was required. They engaged with any patient on the inreach
practitioner (GP) within 24 h of reception, and a referral team’s caseload within 12 months of their earliest date of release.
generated to the Moutjoy Prison Inreach Mental Health Service Participants on the programme were all those individuals on
if deemed necessary. Referrals were also received from other the Moutjoy Prison Inreach Mental Health Service caseload who
sentenced or remand prisons in the event of a prisoner with were released to the community within the 2 years study period
identified mental health needs being transferred to Moutjoy from 1st March 2015 to 28th February 2017.
Prison.
Additionally, referrals of prisoners already allocated within Variables, Data Sources and
the prison were received at weekly multiagency meetings Measurements
chaired by the visiting consultant forensic psychiatrist and For all participants demographic and clinical information was
attended by the healthcare prison governor, the prison chief routinely collected by members of the PReP Programme based
nurse officer, general prison nursing staff, probation services, on assessment and information gathered from electronic prison

Frontiers in Psychiatry | www.frontiersin.org 4 November 2018 | Volume 9 | Article 549

Smith et al. Beyond the Walls: A Pre-Release Planning Programme

medical records and collateral sources. Binary measures were patient placement from a risk-need appropriateness perspective
used when possible to aid with data analysis. Variables included to ensure proportionality and safety. The DUNDRUM-1 and
age, nationality, offense type, homeless status, accommodation DUNDRUM-2 have previously been used for this purpose in a
at time of reception to the prison, prior engagement with remand prison setting (7).
community mental health teams, lifetime history of self-harm, DUNDRUM-1 and DUNDRUM-2 mean scores were
lifetime history of polysubstance abuse, lifetime history of calculated by members of the Moutjoy Prison Inreach Mental
psychosis, active psychosis at time of first assessment and ICD- Health Service for all participants in the week prior to their
10 (35) diagnosis at time of release. Diagnoses were documented release from custody.
by the Mountjoy Inreach Mental Health Service and PReP
Programme based on serial clinical interviews and review of past Ethical Approval
medical and psychiatric case records from prison and community The study protocol was approved by the National Forensic
sources. All diagnoses were validated by a Consultant Forensic Mental Health Service Research, Audit, Ethics and Effectiveness
Psychiatrist. Committee and by the Irish Prison Service Research Ethics
Offense type related to the most serious index offense on Committee as a service evaluation project (39). In accordance
reception at the prison and was classified as violent or non- with internationally recognized ethical principles, service
violent. A violent offense was defined as an act of physical evaluation studies do not require signed informed individual
violence on a person and included homicide, assault, robbery, consent for all patients assessed and participating. Service
aggravated burglary, contact sexual offenses, false imprisonment, evaluation is an ethical obligation in order to ensure appropriate
driving offenses involving injury to others and arson where there use of resources, appropriate quality and standards for patients
was a possibility of injury to others. and continuous learning at the systems level. All patients
Homelessness was defined as rough sleeping or residence therefore benefit. Nonetheless all participants gave written
in homeless shelters reported at the time of committal. Rough informed consent to participate in the programme. No
sleeping was defined as sleeping outside on the street or in randomization procedure was used for allocation to the PReP
other open spaces. Those individuals staying with family or Programme. All data collected were anonymized and no
friends, or in long term placements were not included in the individual patient data have been presented.
definition of homelessness for the purposes of this study. More
detailed information about the security of tenure and quality of Data Analysis
accommodation at time of reception and upon release was also Anonymized data were analyzed using IBM SPSS version 24.
captured. We used Chi-square tests to explore the relationship between
Regarding outcome measures, the mental health/healthcare categorical variables. A Fisher Exact test was used when there was
support and accommodation achieved on day of release was an expected count of <5 in any of the groups. We used t-tests to
recorded. This information was gathered from interviews, compare continuous variable means between two groups and a
collateral sources, electronic prison medical records and one-way analysis of variance (ANOVA) when comparing means
correspondence with receiving community based supports. In between multiple groups.
order to explore whether or not gains had been achieved
following the intervention of the PReP Programme, in terms RESULTS
of level of mental health support and security of tenure and
quality of accommodation, these outcomes were compared Figure 2 displays the pathway from point of reception at the
before and after the period of imprisonment. If a participant prison to mental health outcome on day of release for all 3,010
of the programme was re-imprisoned within the 2 years study committals to Mountjoy Prison, from 1st March 2015 to 28th
period this was identified and recorded. February 2017. Of these, 2,697 committals (89.6%, 2697/3010)
The DUNDRUM Toolkit (36) was used to assess the were deemed not to require psychiatric assessment following
risk-appropriateness (whether transfer to a particular level of screening of referrals by the Mountjoy Prison Inreach Mental
therapeutic security is necessary) of the mental health outcomes Health Service. The remaining 313 (10.4%, 313/3010) committals
achieved upon release. DUNDRUM-1 (37) assesses level of were taken onto the caseload; 43 (13.7%, 43/313) of whom were
security required. The DUNDRUM-2 (38) rates urgency of need subsequently supported by the PReP Programme as they were
for admission. The sum score of the DUNDRUM-1 is divided by expected to be released within 12 months. This represented
the number of items to provide a mean score which is always 40 individuals as one participant was imprisoned at Mountjoy
between zero and four. A mean DUNDRUM-1 score >3 would Prison twice and another three times, during the study period.
guide a need for high therapeutic security, between 2 and 3 For this group, the median duration from date of initial
would guide toward medium therapeutic security and between reception at any prison during the relevant committal episode to
1 and 2 would guide toward acute low therapeutic security, date of release was 516 days (N = 43, mean 672.9 days SD 772.0),
often referred to as Psychiatric Intensive Care Unit. Scores and from date of committal to Mountjoy Prison to date of release
lower than one indicate an open hospital ward or community was 259 days (N = 43, mean 534.6 days SD 722.7). The median
setting would be appropriate. These scores are not binding but duration from date of committal at Mountjoy Prison to date of
assist the clinical decision maker for individual cases. The mean first assessment by the Inreach Mental Health Service was 6 days
scores for groups are useful guides to the appropriateness of (N = 43, mean 54.8 days SD 164.7). The median duration from

Frontiers in Psychiatry | www.frontiersin.org 5 November 2018 | Volume 9 | Article 549

Smith et al. Beyond the Walls: A Pre-Release Planning Programme

FIGURE 2 | Consort diagram displaying mental health outcomes on day of release for all those seen by the Mountjoy Prison Ireach Mental Health Service and PReP
Programme from 1st March 2015 to 28th February 2017. PReP, Pre-Release Planning; CMHT, Community Mental Health Service; GP, general practitioner; MHA
2001, Mental Health Act 2001.

date of committal at Mountjoy Prison to date of first assessment individuals had a median duration in Mountjoy Prison of 5.5
by the PReP Programme was 124.0 days (N = 43, mean 380.1 days (mean 15.9 days SD 18.1). Despite spending only a brief
days SD 696.3). The median duration from date first seen by the period in Mountjoy Prison the majority (87.5%, 7/8) of this group
PReP Programme to date of release was 123 days (N = 43, mean were referred for healthcare follow up upon release by the inreach
154.4 days SD 149.2). mental health service.
Mental health outcomes for the eight individuals on the A pre-release planning (PReP) meeting was convened prior to
caseload who were released before being seen by the PReP release for 32 of those availing of the support of the programme
programme are also displayed in Figure 2. For this group the (74.4%, 32/43). Ten (31.3%, 10/32) of these meetings were
median duration from date of committal to Mountjoy Prison to attended by community mental health teams, 17 (53.1%, 17/32)
date of first assessment by the Mountjoy Prison Inreach Mental were attended by a family member/spouse, and nine (28.1%,
Health Service was 2.5 days (N = 8, mean 3.9 days SD 4.5). These 9/32) were attended by the patient themselves.

Frontiers in Psychiatry | www.frontiersin.org 6 November 2018 | Volume 9 | Article 549

Smith et al. Beyond the Walls: A Pre-Release Planning Programme

A meeting was not convened for the remaining 11 committals TABLE 1 | Comparison of demographic, legal and clinical characteristics of
for the following reasons: a release plan had already been agreed participants and non-participants of the PReP Programme.

by all parties (N = 7); the patient was unexpectedly released (N = Participants Non-Participants Statistical p-value
2); no severe mental illness (defined as major depressive disorder, (N = 43) (N = 8) test of
hypomania, bipolar disorder and/or any form of psychosis difference
including schizophrenia, schizoaffective disorder and any other N (%) Mean (SD) N (%) Mean (SD)
non-affective, non-organic psychosis) was identified following
serial assessments by the team (N = 2). The mental health Age 35.67 (8.02) 32.88 (7.75) t = 0.91 0.37
outcomes for these 11 patients are shown in Figure 2. Ten of NATIONALITY
these individuals had healthcare support arranged on the day Irish 41 (95) 6 (75) FET 0.11
of their release despite no formal meeting having been held. Non-Irish 2 (5) 2 (25)
No healthcare input was arranged for the remaining individual HOMELESS ON RECEPTION
Yes 21 (49) 2 (25) FET 0.27
as they were found not to meet criteria for a severe mental
No 22 (51) 6 (75)
illness.
OFFENSE TYPE
All 43 committals seen by the PreP Programme were
Violent 22 (51) 0 (100) FET 0.02
issued with a written release plan, the contents of which
Non-violent 21 (49) 8 (0)
are described in the methods section of this article. In TRANSFERRED FROM ANOTHER PRISON
the event of healthcare follow up being arranged a written Yes 31 (72) 0 (0) FET <0.001
release plan was also forwarded to the receiving healthcare No 12 (28) 8 (100)
provider. PREVIOUS ADMISSION TO SECURE HOSPITAL
Yes 18 (42) 0 (0) FET 0.04
Case Description No 25 (58) 8 (100)
Demographic, legal and clinical characteristics for those who PSYCHOTIC AT FIRST ASSESSMENT
availed of the support of the PReP Programme (N = 43) Yes 16 (37) 5 (62) FET 0.25
and those who did not (N = 8) are displayed in Table 1. No 27 (63) 3 (38)
Participants and non-participants did no differ significantly LIFETIME PSYCHOSIS
in relation to age, nationality, homeless status at time of Yes 33 (77) 6 (75) FET 1.00

reception or clinical variables. Participants however, were No 10 (23) 2 (25)

HISTORY OF PSA
more likely to have been charged with a violent offense,
Yes 39 (91) 7 (88) FET 1.00
to have been transferred from another prison and to have
No 4 (9) 1 (12)
had a previous admission to a secure forensic psychiatric
HISTORY OF SELF-HARM
hospital. Yes 26 (60) 4 (50) FET 0.70
No 17 (40) 4 (50)
Demographics PREVIOUS CONTACT WITH CMHT
Of the 43 committals seen by the PReP Programme, all were male, Yes 30 (70) 7 (88) 0.42
and 41 (95.3%, 41/43) identified themselves as Irish, with the No 13 (30) 1 (12) FET
remaining two individuals identifying as Non-Irish Europeans.
PSA, polysubstance abuse; CMHT, Community Mental Health Team; FET, Fisher’s exact
The mean age at time of first assessment by the Mountjoy Prison test.
Inreach Mental Health Service was 36 years (SD 8.0, range
21–63).
As a result of concerns regarding risk posed to children in the
Offense Type event of release, a total of seven referrals were made to Tusla,
Regarding the nature of the most serious index offense, of Ireland’s Child and Family Agency, by members of the PReP
those seen by the PReP Programme 48.8% (21/43) were Programme in keeping with their obligations under Ireland’s
charged with a violent offense, that is one involving physical child protection legislation.
violence to another person. The remaining 51.2% (22/43)
were charged with non-violent offenses. Thirty-one (31/43, CLINICAL CHARACTERISTICS
72.1%) of those supported by the PReP Programme were
transferred from another remand or sentenced prison Primary ICD-10 Diagnoses, Active and
to Mountjoy Prison. Two were re-patriated from prisons Lifetime Psychosis
abroad. Table 2 displays the primary ICD-10 diagnosis at the time of
release for all those seen by the PReP Programme. Almost
Contact With Children and Child Protection two thirds of those seen had primary ICD-10 diagnoses of
Issues Schizophrenia, Schizotypal and Delusional Disorders (58.1%,
Sixteen of the 43 committals seen by the PReP Programme 25/43) or Bipolar Affective Disorder (2.3%, 1/43). An additional
reported having children. Of these, 14.0% (6/43) reported that 16.3% (7/43) had a primary diagnosis of a drug induced psychotic
they had contact with their children prior to reception at prison. episode. At the time of initial assessment by the Mountjoy Prison

Frontiers in Psychiatry | www.frontiersin.org 7 November 2018 | Volume 9 | Article 549

Smith et al. Beyond the Walls: A Pre-Release Planning Programme

Inreach Mental Health Service, 37.2% (16/43) of those seen and two had been repatriated from international prisons (4.7%,
following screening and referral were assessed as being actively 2/43).
psychotic. Based on information from interview and collateral Regarding compliance with prescribed psychiatric
sources, just over three quarters of individuals seen by the PReP medications, of the 25 (58.1%, 25/43) committals prescribed such
Programme had a lifetime history of a psychotic illness (76.7%, treatment prior to their imprisonment, 14 reported being fully
33/43). compliant (56.0%, 14/25), seven (28.0%, 7/25) reported being
partially compliant and four reported being non-compliant
Co-morbidity and Self-Harm History (16.0%, 4/25).
Almost all individuals supported by the programme had a
lifetime history of polysubstance abuse (90.7%, 39/43). Based Outcomes Following the Intervention of
upon collateral information, one quarter (25.6%, 11/43) had a the PReP Programme:
co-morbid diagnosis of a personality disorder. Of all those seen, 1. Mental health outcomes:
60.5% (26/43) had a lifetime history of deliberate self-harm. Mental health supports arranged on day of release for all those
seen by the PReP Programme are displayed in Figure 2.
Previous Contact and Engagement With Of the 43 committals seen by the programme, 35 (81.4%,
35/43) were referred for community mental health team
Community Mental Health Teams and follow up upon release, of which 82.9% (29/35) were accepted.
Other Healthcare Supports Fifteen (51.7%%, 15/29) of these accepted referrals, were
The majority of those seen by the PReP Programme (69.8%, initially declined. In these cases further efforts were made by
30/43) reported prior contact with a community mental health the PReP Programme to liaise with the receiving service to
team at some point before their reception at the prison. Eighteen address their concerns so that the referral process could be
individuals (41.9%, 18/43) had previously been admitted to the completed.
Central Mental Hospital, the Republic of Ireland’s only secure Table 3 displays a comparison between the level of
forensic hospital. healthcare support at time of reception at prison compared
Regarding level of engagement with mental health supports with that arranged on day of release following the intervention
prior to reception, 14 (32.6%, 14/43) had no contact with any of the PReP Programme. A Fisher Exact Test indicated that
mental health supports; six (14.0%, 6/43) were attending a general the level of mental health support significantly improved
practitioner alone; 20 (47%, 20/43) were attending outpatient upon release from prison, following the intervention of the
services (community mental health team, addiction services or programme (FET p < 0.001).
intellectual disability services), one was in hospital (2.3%, 1/43) Regarding post-release engagement with arranged mental
health supports, the PReP Programme confirmed that 89.7%
of those accepted by community mental health teams (26/29)
TABLE 2 | Primary ICD-10 diagnosis at time of release for all those seen by the attended their first appointment in the post-release period. Of
PReP Programme (N = 43). these, 27.6% (8/29) were admitted involuntarily to a general
psychiatric hospital under the Mental Health Act 2001.
Primary ICD-10 diagnosis N %
Receiving mental health services were then contacted
in the post-release period to confirm if the referred
F00-09 Organic disorders 1 2.3
individual remained engaged following attendance at their
- Alcohol related dementia
first appointment. The median duration of post-release follow
F10-19 Substance use disorder up was 20.5 days (mean 61.31 days, SD 104.09). At time of
- Drug induced psychosis 7 16.3
- Polysubstance abuse only 1 2.3
F20-29 Schizophreniform disorders TABLE 3 | Comparison of level of healthcare support at time of reception to
- Schizophrenia 18 41.9 prison with that on day of release, following the intervention of the PReP
- Schizoaffective disorder 5 11.6 Programme (N = 43).

- Delusional disorder 2 4.7 Healthcare support Total

F30-39 Mood disorder
None GP Outpatient services Prison Hospital
- Manic episode 1 2.3
(CMHT, Addiction
- Depressive episode 3 7.0 services, ID services)
F60-69 Personality disorder
Prior to 14 6 20 2 1 43
- Emotionally unstable personality disorder 3 7.0
reception at
F70-79 Mild intellectual disability 2 4.7 prison (N)
Total 43 100 On day of 2 11 22 0 8 43
release (N)
ICD-10, International Statistical Classification of Diseases and Related Health Problems,
10th Revision. CMHT, Community Mental Health Team; ID, intellectual disability; GP, general practitioner.

Frontiers in Psychiatry | www.frontiersin.org 8 November 2018 | Volume 9 | Article 549

Smith et al. Beyond the Walls: A Pre-Release Planning Programme

follow up, 20 individuals (76.9%, 20/26) remained engaged tenure and quality of accommodation significantly improved
with community mental health teams, of whom four were upon release from prison following the intervention of the
inpatients, and none had returned to prison. PReP Programme (FET p < 0.001).
Risk-appropriateness of arranged mental health supports: 3. Re-imprisonment:
Mean DUNDRUM-1 triage security and DUNDRUM-2 triage Of those participants seen by the PReP Programme, 20 (46.5%,
urgency scores for those seen by the PReP Programme 20/43) were returned to prison during the 2-years study
(N = 43) released to community inpatient (N = 8), outpatient period. The median duration from date of release to end of the
services (community mental health team, addiction services, study period was 274.0 days (mean 314.0 days SD 185.9 days).
intellectual disability services) (N = 22), general practitioner Fifteen individuals (34.9%, 15/43) were under the supervision
(N = 11) and no healthcare follow up (N = 2) are summarized of probation services when initially released, 7 (46.7%, 7/15)
in Table 4. of who were re-imprisoned during the 2-years study period.
Mean DUNDRUM-1 triage security scores (ANOVA Table 6 displays rates of re-imprisonment for all those
F = 1.99, between groups df = 3, within groups df = 39, supported by the PReP Programme according to the level
p = 0.13) and DUNDRUM-2 triage urgency scores (ANOVA of mental health support and accommodation achieved on
F = 1.87, between groups df = 3, within groups df = 39, day of release. There was no significant relationship between
p = 0.15), although not significant, tended to be higher for re-imprisonment and gains made in level of mental health
those transferred to higher levels of mental health support. support (FET p = 0.23) or accommodation (FET p = 0.23)
2. Accommodation outcomes: following the support of the PReP Programme, however the
Twenty one (48.8%, 21/43) committals seen by the PReP duration of follow up was relatively short (median 274.0 days).
Programme were homeless at the time of their reception
to prison. This included five (23.8%, 5/21) who reported Secondary Analysis
rough sleeping, 13 (61.9%, 13/21) who reported staying in For the reasons outlined above, eleven participants availed
emergency “night to night” homeless shelters and two (9.5%, of the support of the PReP Programme but did not have a
2/21) who reported staying in short term, “week to week” pre-release planning (PReP) meeting prior to their release. A
homeless shelters. The remaining individual (4.8%, 1/21) was secondary analysis was performed to explore if a meeting was
an inpatient in a general psychiatric hospital prior to reception associated with any difference in outcome measures. There was
at prison, but had no regular accommodation before this and no significant difference found between those who had a meeting
reported staying in emergency homeless shelters. Twenty-one (N = 32) and those who did not (N = 11) in relation to mental
participants (48.8%, 21/43) continued to meet the definition health outcomes (FET p = 0.24), security of tenure and quality of
of homelessness at the time of release. No individuals were accommodation achieved upon release (FET p = 0.74) and rates
released to rough sleeping. of re-imprisonment (X 2 = 0.38, df = 2, p = 0.72).
Table 5 displays a comparison between accommodation
at time of reception at prison compared with that achieved DISCUSSION
on day of release following the intervention of the PReP
Programme. A Fisher Exact Test indicated that the security of We have followed a participatory action research design to
introduce a new service for mentally disordered offenders as they
transition from prison to the community. We have completed an
TABLE 4 | Risk-appropriateness of mental health outcomes for all those seen by
evaluation of the first 2 years of the project to examine whether
PReP Programme (N =43). the goals of the service were achieved. In particular whether
those referred to the PReP Programme had improved levels
N (%) D-1 triage D-2 triage of mental health support and improved security of tenure and
security score urgency score
quality of accommodation upon their release in comparison to
Mean (SD) 95% CI Mean (SD) 95% CI that reported at time of imprisonment. During the period of this
study, there were no other major changes in the organization,
Psychiatric 8 (19) 2.11 (0.60) 1.61–2.62 2.05 (0.71) 1.46–2.64 management or delivery of prison in-reach services nor was there
admission
any major change in the organization, management or delivery of
Outpatient 22 (51) 1.64 (0.84) 1.27–2.01 1.45 (1.00) 1.01–1.89
prison and criminal justice services.
Services
(CMHT,
Addiction Summary of Findings
services, ID We have shown that compared to that reported at time of
services) imprisonment, the level of mental health support and the
GP 11 (25) 1.54 (0.80) 1.00–2.07 1.32 (1.16) 0.54–2.10 security of tenure and quality of accommodation both improved
No healthcare 2 (5) 0.70 (0.57) −4.38–5.78 0.35 (0.21) −1.56–2.26 following the intervention of the PReP Programme. In the
follow-up absence of a control group we cannot show that the PReP
D-1, DUNDRUM-1; D-2, DUNDRUM-2; SD, standard deviation; 95% CI, 95% confidence
programme caused this effect, but we believe this is so. Higher
interval; CMHT, Community Mental Health Team; ID, intellectual disability; GP, general levels of mental health support and improved accommodation
practitioner. were not associated with lower rates of re-imprisonment within

Frontiers in Psychiatry | www.frontiersin.org 9 November 2018 | Volume 9 | Article 549

Smith et al. Beyond the Walls: A Pre-Release Planning Programme

TABLE 5 | Comparison of accommodation at time of reception to prison with that on day of release, following the intervention of the PReP Programme (N = 43).

Accommodation Total

Rough Emergency/Short Long term hostel, secure Hospital Prison

sleeping term hostel tenancy, living with family

Prior to 5 15 20 1 2 43
reception at
prison (N)
On day of 0 16 19 8 0 43
release (N)

Emergency Hostel Accommodation, in a homeless shelter booked on a nightly basis; Short Term Hostel, accommodation in a homeless shelter booked on a weekly basis; Long Term
Hostel, accommodation in a homeless shelter booked for 6 months or longer; Secure Tenancy, private rented accommodation or own home.

TABLE 6 | Impact of level of mental health support and accommodation outcomes on rates of re-imprisonment, following the intervention of the PReP Programme
(N = 43).

Re-imprisoned? Healthcare support on day of release Total

None GP Outpatient Services Involuntary Hospital

(CMHT, Addiction admission under MHA
services, ID Services) 2001

Yes (N) 2 4 12 2 20
No (N) 0 7 10 6 23

Accommodation on day of release

Rough Emergency/ Long term hostel, secure Involuntary hospital

sleeping short term tenancy, living with family admission under MHA
hostel 2001
Yes (N) 0 10 8 2 20
No (N) 0 6 11 6 23

MHA 2001, Mental Health Act 2001; CMHT, Community Mental Health Team; ID, intellectual disability; GP, general practitioner.

the 2 years study period however the follow up period was them to build trusting relationships with mentally disordered
relatively short. We were not able to further analyse relationships offenders in the pre-release period. Practical supports offered
between variables and outcomes owing to lack of statistical by the programme, including liaison with family members and
power. assistance in accessing accommodation and social welfare may
have acted as incentives for engagement before and after release.
Strengths and Limitations This may have been reflected by the high rates of engagement
This project, the first of its kind in Ireland, embodies with arranged mental health appointments immediately after
the principles of integrated and multidisciplinary healthcare release (89.7%, 26/29).
provision. Post-release mental health and accommodation The main focus of the programme was to improve pre-
outcomes were mapped for all those seen by the PReP release planning and manage transfer of care to community based
programme. Healthcare outcomes were also mapped and supports. Social workers from the programme subsequently
presented for eight patients on the inreach mental health team’s offered time limited telephone support to service users, family
caseload who were released prior to availing of the support of the members and receiving services. This correspondence revealed
PReP programme. that the majority of those receiving mental health follow up from
Prior to the development of the PReP Programme, release community mental health services remained engaged at a median
planning in the prison studied was performed by a medically duration of 3 weeks following their release (76.9%, 20/26).
focused inreach mental health service comprised of doctors and Unlike Assertive Community Treatment (ACT) and Critical
nurses. The addition of mental health social work expertise Time Intervention (CTI), the programme did not provide case
enhanced the ability of the team to develop robust release management in the post-release period. Although this may be
plans in collaboration with community based supports. As viewed as a limitation of the PReP Programme, previous studies
suggested by Jarrett et al. (22) social workers might be best (22, 29, 40) and a recent systematic review (24), have highlighted
placed to coordinate such care plans given their knowledge of the importance of pre-release planning in any intervention
local services and support agencies. The social workers of the to aid the transition for mentally disordered offenders. We
Pre-Release Planning (PReP) Programme were based within the acknowledge that the less intense follow up provided by our
prison as part of the inreach mental health team. This allowed programme results in difficulty determining the quality of

Frontiers in Psychiatry | www.frontiersin.org 10 November 2018 | Volume 9 | Article 549

Smith et al. Beyond the Walls: A Pre-Release Planning Programme

engagement with mental health and other supports in the post- the addition of a meeting as outlined in previous studies of
release period. Future projects will focus on assessing whether or this kind (40). This may be a focus of future research by our
not the achievements of the PReP programme translate into long service.
term sustained improvements in engagement with mental health In the event of a pre-release planning meeting being
supports, accommodation and legal outcomes. held, attendance by community mental health teams, families
Homelessness is one of the greatest challenges facing released and service users was relatively poor. In Ireland, as in
prisoners (23) and may act as an impediment to engaging many developed countries, there is no statutory requirement
with healthcare supports (41). These individuals may be for any agency to attend pre-release planning meetings.
further marginalized losing out on available accommodation to Unfortunately community mental health teams were often
family’s and non-mentally ill persons experiencing homelessness. unable to attend due to scheduling problems and on occasion
Although rates of broadly defined homelessness were not due to reluctance to accept the individual until late in the
reduced following the intervention of the programme (N prisoner’s sentence. Despite having the support of prison
= 21 on reception vs. N = 21 on day of release), there authorities it often proved difficult to transfer prisoners
was evidence of improvements in the security of tenure and from their location in the prison to the site of pre-release
quality of accommodation obtained upon release. Moreover, the planning meetings. This occurred mainly due to prison
fact that more individuals were not released to homelessness officer shortages or because the prisoner was too unwell to
may represent an improved outcome, given that previous attend.
studies have highlighted an increased risk of homelessness All mentally disordered offenders on the inreach mental
and unstable housing upon release from prison (42). Despite health services’ caseload released within the period studied
improvements in both the level of healthcare support and were eligible for support by the programme. This inclusive
accommodation achieved following imprisonment and the approach did not permit the creation of a comparable control
intervention of the PReP Programme, 46.5% (20/43) of those group, which would have allowed for more rigorous analysis
supported by the intervention were re-imprisoned within the regarding the effectiveness of the intervention. Additionally,
2 years study period. Although disappointing, this rate of the service has been operational for 2 years, therefore we
re-imprisonment is consistent with that reported for general were not able to further analyse relationships between variables
prison populations in our jurisdiction (15). Gains made in and outcomes owing to a lack of statistical power. Also, the
healthcare and accommodation outcomes were not associated inclusive and real-world nature of this project resulted in
with reduced rates of re-imprisonment during a relatively some participants availing of the support of the programme
short follow up period. This finding may not be surprising despite not meeting criteria for a mental illness at the time of
as a number of more intensive post-release case management release.
models have found an association with increased rates of This project was set in an all male sentenced prison and its
re-imprisonment through the increased level of monitoring findings may not be transferable to female prison populations.
provided by these interventions in the post-release period Future plans by our service include the establishment of a
(24). Regrettably, information was not available regarding the similar social work-led PReP Programmes in a number of
status of participant’s mental illness and level of engagement Ireland’s other sentenced prisons, including its main female
with community mental health supports at the time of re- prison.
imprisonment.
A process of participatory action research was used to
design, develop and evaluate the PReP programme. This design
CONCLUSIONS
meant that the programme could be implemented without delay We have shown that compared to that reported at time
following the identification of a need by stakeholders within the of imprisonment, the level of mental health support and
prison. Although this creates practical advantages for service the security of tenure and quality of accommodation both
development, it may result in difficulty identifying the specific improved at time of release, following the intervention of the
variables asscoiated with achieved outcomes. PReP Programme. Higher levels of mental health support and
At the planning stage of the project, a multidisciplinary, improved accommodation were not associated with lower rates
multiagency pre-release planning (PReP) meeting was envisaged of re-imprisonment within the 2 years study period.
to be a central component of the intervention provided by the
PReP Programme. Despite this not all of those supported by
the programme had a pre-release planning meeting. We have DATA AVAILABILITY
outlined reasons why meetings were not convened for eleven
of the forty-three participants. We also performed a secondary The raw data supporting the conclusions of this manuscript will
analysis to explore if a meeting was associated with improved be made available by the authors, without undue reservation, to
outcomes and found that it was not. We believe this is an any qualified researcher.
interesting observation. It implies that the networking and liaison
work carried out by PReP team members is as effective as a AUTHOR CONTRIBUTIONS
meeting arranged in addition to that liaison work, at least from
a quantitative, outcomes point of view. It remains possible that DS completed the first and revised drafts of the
better qualitative outcomes and experiences would result from manuscript, which were then edited by SHa, AF, SHe,

Frontiers in Psychiatry | www.frontiersin.org 11 November 2018 | Volume 9 | Article 549

Smith et al. Beyond the Walls: A Pre-Release Planning Programme

NQ, CC, DM and HK who also assisted with data ACKNOWLEDGMENTS

analysis. The intervention was designed and developed
by DM, SHa, AF, SHe and PG, with assistance from We would like to acknowledge and thank the clinical and
all stakeholders including prisoners and their families. custodial staff of Mountjoy Prison. In addition, we’d like to thank
All authors contributed to the participatory action the prisoners, family members, prison services and community
research process. All authors read and approved the final based support agencies that engaged with the Pre-Release
manuscript. Planning Programme.

REFERENCES 18. Till A, Exworthy T, Forrester A. Integration and offender

mental health. J Forensic Psychiatry Psychol. (2014) 26:11–21.
1. Duffy D, Linehan S, Kennedy HG. Psychiatric morbidity in the male doi: 10.1080/14789949.2014.960440
sentenced Irish prisons population. Irish J Psychol Med. (2006) 23:54–62. 19. World Health Organisation (WHO). Framework on Integrated, People-
doi: 10.1017/S0790966700009587 Centred Health Services (2016).
2. Curtin K, Monks S, Wright B, Duffy D, Linehan S, Kennedy HG. Psychiatric 20. Department of Health UK. NHS Outcomes Framework 2016 to 2017 (2016).
morbidity in male remanded and sentenced committals to Irish prisons. Irish 21. Department of Health Ireland. Statement of Strategy 2016–2019 (2016).
J Psychol Med. (2014) 26:169–173. doi: 10.1017/S079096670000063X 22. Jarrett M, Thornicroft G, Forrester A, Harty M, Senior J, King C,
3. Fazel S, Seewald K. Severe mental illness in 33,588 prisoners worldwide: et al. Continuity of care for recently released prisoners with mental
systematic review and meta-regression analysis. Br J Psychiatry (2012) illness: a pilot randomised controlled trial testing the feasibility of
200:364–73. doi: 10.1192/bjp.bp.111.096370 a critical time intervention. Epidemiol Psychiatr Sci. (2012) 21:187–93.
4. Maden T. Psychiatric morbidity among prisoners in england and wales: Brit. doi: 10.1017/S2045796011000783
J. Psychiatry (2000) 176:503–503. doi: 10.1192/bjp.176.5.503 23. Binswanger IA, Nowels C, Corsi KF, Long J, Booth RE, Kutner J, et al. From the
5. Durcan G. From the Inside: Experiences of Prison Mental Health Care. prison door right to the sidewalk, everything went downhill a qualitative study
Sainsbury Centre for Mental Health (2008). of the health experiences of recently released inmates. Int J Law Psychiatry
6. McInerney C, Davoren M, Flynn G, Mullins D, Fitzpatrick M, Caddow (2011) 34:249–55. doi: 10.1016/j.ijlp.2011.07.002
M, Caddow F, et al. Implementing a court diversion and liaison scheme 24. Hopkin G, Evans-Lacko S, Forrester A, Shaw J, Thornicroft G. interventions
in a remand prison by systematic screening of new receptions: a 6 year at the transition from prison to the community for prisoners with mental
participatory action research study of 20,084 consecutive male remands. Int illness: a systematic review. Adm Policy Ment Health (2018) 45:623–34.
J Mental Health Syst. (2013) 7:18. doi: 10.1186/1752-4458-7-18 doi: 10.1007/s10488-018-0848-z
7. O’Neill C, Smith D, Caddow M, Duffy F, Hickey P, Fitzpatrick M, 25. Assembly UG. UN General Assembly, United Nations Standard Minimum
et al. STRESS-testing clinical activity and outcomes for a combined Rules for the Treatment of Prisoners (the Nelson Mandela Rules). (2016).
prison in-reach and court liaison service: a 3-year observational study of Available online at: http://www.refworld.org/docid/5698a3a44.html
6177 consecutive male remands. Int J Mental Health Syst. (2016) 10:67. 26. Seymour M, Costello L. A Study of the Number, Profile and Progression
doi: 10.1186/s13033-016-0097-z Routes of Homeless Persons before the Court and in Custody. Research Report
8. Giblin Y, Kelly A, Kelly E, Kennedy HG, Mohan D. Reducing the use of Commissioned by the Department of Justice, Equality and Law Reform. (2015).
seclusion for mental disorder in a prison: implementing a high support unit 27. Fazel S, Hayes AJ, Bartellas K, Clerici M, Trestman R. Mental health of
in a prison using participant action research. Int J Mental Health Syst. (2012) prisoners: prevalence, adverse outcomes, and interventions. Lancet Psychiatry
6:2. doi: 10.1186/1752-4458-6-2 (2016) 3:871–81. doi: 10.1016/S2215-0366(16)30142-0
9. Pratt D, Appleby L, Piper M, Webb R, Shaw J. Suicide in recently 28. Solomon P, Draine J. One-year outcomes of a randomized trial of case
released prisoners: a case-control study. Psychol Med. (2009) 40:827–35. management with seriously mentally ill clients leaving jail. Eval Rev. (2016)
doi: 10.1017/S0033291709991048 19:256–73. doi: 10.1177/0193841X9501900302
10. Pratt D, Piper M, Appleby L, Webb R, Shaw J. Suicide in recently released 29. McKenna B, Skipworth J, Tapsell R, Madell D, Pillai K, Simpson A, et al.
prisoners: a population-based cohort study. Lancet (2006) 368:119–23. A prison mental health in-reach model informed by assertive community
doi: 10.1016/S0140-6736(06)69002-8 treatment principles: evaluation of its impact on planning during the pre-
11. Farrell M, Marsden J. Acute risk of drug-related death among newly release period, community mental health service engagement and reoffending.
released prisoners in England and Wales. Addiction (2008) 103:251–5. Criminal Behav Mental Health (2015) 25:429–39. doi: 10.1002/cbm.1942
doi: 10.1111/j.1360-0443.2007.02081.x 30. Draine J, Herman DB. Critical time intervention for reentry from
12. Lennox C, Senior J, King C, Hassan L, Clayton R, Thornicroft G, prison for persons with mental illness. Psychiatr Serv. (2007) 58:1577–81.
Shaw J. The management of released prisoners with severe and doi: 10.1176/ps.2007.58.12.1577
enduring mental illness. J Forensic Psychiatry Psychol. (2012) 23:67–75. 31. Shaw J, Conover S, Herman D, Jarrett M, Leese M, McCrone P, et al. Critical
doi: 10.1080/14789949.2011.634921 time Intervention for Severely mentally ill Prisoners (CrISP): a randomised
13. McMahon A. Homeless crisis. Irish Times (2017) Available online at: https:// controlled trial. Health Serv Deliv Res. (2017) 5:1–138. doi: 10.3310/hsdr05080
www.irishtimes.com/news/social-affairs/homeless-crisis-nearly-9-000- 32. Herman DB. transitional support for adults with severe mental illness: critical
destitute-in-november-1.3333124 time intervention and its roots in assertive community treatment. Res Social
14. Butler P. One in every 200 people in UK are homeless, according to Work Pract. (2014) 24:556–63. doi: 10.1177/1049731513510976
Shelter. The Guardian (2017). Available online at: https://www.theguardian. 33. Chen F-P, Ogden L. A Working relationship model that reduces homelessness
com/society/2017/nov/08/one-in-every-200-people-in-uk-are-homeless- among people with mental illness. Qualitat Health Res. (2011) 22:373–83.
according-to-shelter doi: 10.1177/1049732311421180
15. Central Statistics Office. Prison Recidivism: 2008 Cohort. Dublin (2013). 34. Lewin K. Action research and minority problems. J Social Issues (1946)
16. Fazel S, Wolf A. A systematic review of criminal recidivism rates 2:34–46. doi: 10.1111/j.1540-4560.1946.tb02295.x
worldwide: current difficulties and recommendations for best 35. World Health Organisation (WHO). International Statistical Classification of
practice. PLoS ONE (2015) 10:e0130390. doi: 10.1371/journal.pone. Diseases and Related Health Problems, 10th Revision (ICD-10). Geneva (1992).
0130390 36. Kennedy HG, O’Neill C, Flynn G, Gill P. Dangerousness Understanding,
17. Fazel S, Yu R. Psychotic disorders and repeat offending: systematic review and Recovery and Urgency Manual (The Dundrum Quartet). 1st ed. (2010).
meta-analysis. Schizophr Bull. (2011) 37:800–10. doi: 10.1093/schbul/sbp135 Available online at: http://www.tara.tcd.ie/bitstream/handle/2262/39131/

Frontiers in Psychiatry | www.frontiersin.org 12 November 2018 | Volume 9 | Article 549

Smith et al. Beyond the Walls: A Pre-Release Planning Programme

THE%20DUNDRUM%20TOOLKIT%20V1%200%2021%201304101.pdf? 42. Metraux S, Culhane DP. Homeless shelter use and

sequence=1&isAllowed=y reincarceration following prison release. Criminol Public
37. Flynn G, O’Neill C, McInerney C, Kennedy HG. The Dundrum-1 structured Policy (2004) 3:139–60. doi: 10.1111/j.1745-9133.2004.tb0
professional judgment for triage to appropriate levels of therapeutic 0031.x
security: retrospective-cohort validation study. BMC Psychiatry (2011) 11:43.
doi: 10.1186/1471-244X-11-43 Conflict of Interest Statement: The authors declare that the research was
38. Flynn G, O’Neill C, Kennedy HG. Dundrum-2: Prospective validation of a conducted in the absence of any commercial or financial relationships that could
structured professional judgment instrument assessing priority for admission be construed as a potential conflict of interest.
from the waiting list for a forensic mental health hospital. BMC Res Notes
(2011) 4:230. doi: 10.1186/1756-0500-4-230 Copyright © 2018 Smith, Harnett, Flanagan, Hennessy, Gill, Quigley, Carey,
39. NHS HRA. Defining Research (2013). McGhee, McManus, Kennedy, Kelly, Carey, Concannon, Kennedy and
40. Hancock N, Smith-Merry J, Mckenzie K. Facilitating people living with severe Mohan. This is an open-access article distributed under the terms of the
and persistent mental illness to transition from prison to community: a Creative Commons Attribution License (CC BY). The use, distribution or
qualitative exploration of staff experiences. Int J Mental Health Syst. (2018) reproduction in other forums is permitted, provided the original author(s)
12:361. doi: 10.1186/s13033-018-0225-z and the copyright owner(s) are credited and that the original publication in
41. Tobin Tyler L, Brockmann B. Returning home: incarceration, reentry, stigma this journal is cited, in accordance with accepted academic practice. No use,
and the perpetuation of racial and socioeconomic health inequity. J Law distribution or reproduction is permitted which does not comply with these
Medicine Ethics. (2018) 45:545–57. doi: 10.1177/1073110517750595 terms.

Frontiers in Psychiatry | www.frontiersin.org 13 November 2018 | Volume 9 | Article 549

View publication stats