Forensic Mental Health Services Managed Care Network
Psychological Therapies: Violence
DRAFT FOR CONSULTATION
(3 October – 30 November 2008)
The Forensic Network Advisory Board considers this to be an authoritative report that will be helpful in informing future service developments. The Board would like your views on all aspects of the three reports and therefore they are subject to a two month consultation period from 3
October until 30 November 2008.
We would particularly welcome your views on the following questions:
1. Do you think the report is helpful as a guideline for the
development of a treatment protocol? Are there any gaps in the
scope of each of the reports, for example are you aware of any
existing programmes or best practice guidelines that might be
2. What issues do you think need to be addressed to successfully
implement such protocols across the Forensic Network, taking
cognisance of the different levels of security and the complexity
of patient pathways?
3. Do you have any further suggestions on how to progress this
Comments should be submitted, clearly stating which report they relate to, by email to:
Vivienne Gration, Forensic Network Manager at
Vivienne.firstname.lastname@example.org by 30 November 2008.
Authors & Membership of The Group 3
Executive Summary 4
Needs Analysis 6
Review of Psychological approaches for the treatment of 6 violence
; Overview of literature
; Static factors
; Dynamic factors
Best Practice Guidelines 15
Assessment Procedures: Criteria for treatment 19
Interventions for Violence 22
Outcome Measures: Clinical Effectiveness 26
Resource Requirements 27
Monitoring of service delivery and Evaluation of Service 29
Staff Supervision and Training Strategy 30
Development of service 32
Conclusions and recommendations 34
Dr Lisa A. Marshall Lead Consultant Clinical Forensic Psychologist, The State Hospital, Carstairs. Associate Programme Organiser MSc Psychology & Mental Health, Glasgow Caledonian University.
Dr Lorraine Johnstone, Clinical Forensic Psychologist, Directorate of Forensic Mental Health, Glasgow.
Stephan McAlpine, Principal Psychologist, Scottish Prison Service.
Phil Kennedy, Forensic Psychologist, The State Hospital, Carstairs
Andrew McFarlane, Criminal Justice Social Worker, FCMHT
Cheryl McMorris, Head II OT, Directorate of Forensic Mental Health & Learning Disabilties. NHS Greater Glasgow & Clyde.
Catherine Totten, Clinical Specialist Occupational Therapist, Discharge Liaison, The State Hospital, Carstairs.
David C Langton, Nurse Consultant, Forensic Network, 110 Lampits Road Carstairs, Lanark
Dr Rhiannon Pugh, Consultant Psychiatrist in Psychotherapy, The State Hospital, Carstairs
Membership of Group:
Dr Lisa A. Marshall Lead Consultant Clinical Forensic Psychologist, The State Hospital, Carstairs. Associate Programme Organiser MSc Psychology & Mental Health, Glasgow Caledonian University. Dr Lorraine Johnstone, Clinical Forensic Psychologist, Glasgow. Stephan McAlpine, Principal Psychologist, Scottish Prison Service. Phil Kennedy, Forensic Psychologist, The State Hospital, Carstairs Andrew McFarlane, Criminal Justice Social Worker, FCMHT
David C Langton, Nurse Consultant, Forensic Network, 110 Lampits Road Carstairs, Lanark
Cheryl McMorris, Head II OT, Directorate of Forensic Mental Health & Learning Disabilities. NHS Greater Glasgow & Clyde. Catherine Totten, Clinical Specialist Occupational Therapist, Discharge Liaison, The State Hospital, Carstairs.
Dr Rhiannon Pugh, Consultant Psychiatrist in Psychotherapy, The State Hospital, Carstairs
Dr John Marshall, Head of Forensic Child and Adolescent Mental Health Service, Glasgow.
The assessment of risk for future violence has been a significant focus of debate and research in the forensic field for at least a decade with use of tools such as the HCR-20 becoming commonplace. However it is only more recently that recognition has been given to the notion that any attempts to ameliorate the risk for violence, should focus on the treatment of dynamic risk factors for violent behaviour. While treatment programmes for problems that co-occur with violent behaviour such as anger and substance misuse are well established, there has been minimal progress with regard to treatments that specifically target reducing future violent behaviour.
The report begins with a needs analysis to illustrate the rate of violent behaviour in the community and within our institutions before moving on to examine current knowledge on the assessment and treatment of violent behaviour. The Needs Analysis demonstrates that a substantial amount of violent incidents occur in forensic mental health units and indicate a requirement for violence reduction programmes to be developed and implemented throughout the network.
Treatment programmes that have a theoretical basis are more effective than those that do not. While no single theoretical approach has been developed that explains all forms of violent behaviour, cognitive behavioural theory, social learning theory and the transtheoretical model of change have been consistently utilised in the treatment programmes currently available.
A range of criminogenic needs that have been identified for mentally disordered violent offenders but more research is needed to clarify their role. The Risk, Need and Responsivity principles should be central to the development of a treatment for violent behaviour. Adherence to a clear theoretical approach, use of standardised, reliable assessment and evaluation tools, use of a treatment manual and recognition of treatment integrity issues should also be at the core of any programme development.
The Violent Reduction Scale (VRS) is currently the only validated measure of violent behaviour that highlights criminogenic needs, treatment readiness and treatment effects. Given the limited range of measures available to assess violence, consideration should be given to utilising assessment measures for the dynamic factors targeted in treatment.
A detailed formulation for each participant enables the identification of key criminogenic needs for each participant that can be addressed in treatment. The Violence Reduction Programme (VRP), Violence Prevention Programme (VPP) and Violent Offender Treatment Programme for Mental Illness (VOTPmi) appear to be the most relevant violent treatment programmes for forensic establishments in Scotland. Programme delivery may be on an individual or a group basis with a „rolling‟ programme which
enables treatment to be tailored to individual needs also an option.
Clinical effectiveness can be assessed by both objective and subjective assessment tools. Pre- and post-treatment assessments can be utilised to demonstrate treatment effects. Institutional involvement and commitment are essential for overcoming barriers to implementation. Monitoring and Evaluation are essential components to the development of any treatment programme. The dearth of literature in this field enhances the need for treatment outcomes to be disseminated.
The selection, training and supervision of staff are particularly vital in the development of a new programme. The complex nature of the assessment and formulation stages of violence treatment suggests that the programme should be co-facilitated, with at least one of the facilitators being a Chartered Clinical and/or Forensic Psychologist. Non-programme staff can also play a key role in supporting treatment gains and facilitating participants‟ engagement.
User involvement in the development would be recommended and a programme that can be adapted across settings to enhance homogeneity would be beneficial.
The assessment of risk for future violence has been a significant focus of debate and research in the forensic field for at least a decade with use of tools such as the HCR-20 becoming commonplace (Webster, Douglas, Eaves & Hart, 1997). However it is only more recently that recognition has been given to the notion that any attempts to ameliorate the risk for violence, should focus on the treatment of dynamic risk factors for violent behaviour (Douglas & Skeem, 2005). While treatment programmes for problems that co-occur with violent behaviour such as anger and substance misuse are well established, there has been minimal progress with regard to treatments that specifically target reducing future violent behaviour. This report will begin with a needs analysis to illustrate the rate of violent behaviour in the community and within our institutions before moving on to examine current knowledge on the assessment and treatment of violent behaviour.
1. Needs Analysis:
Violence is a significant problem for Scotland‟s population. For example, a recent finding reported by Crawford (2006) indicated that for young men under the age of 40 years, stabbing is the second most common cause of death. More than four of every ten convictions in Scotland are for violent offences, with almost one-third of offences for serious acts of violence, i.e. murder, attempted murder, or serious assault (see Table 1). Although these figures are higher for men than for women, violence represents a significant problem within both groups.
1Table 1: Conviction statistics – by crime – in Scottish prisons
Convicted Offenders: 30 June 2000
Male Female Total
Total crimes 4855 157 5012
Non-sexual crimes of 2104 (43%) 50 (32%) 2154
Murder 665 (14%) 20 (13%) 685 (14%)
Attempted murder / 718 (15%) 13 (8%) 731 (15%)
It is widely accepted that formal figures on violence are likely to grossly underestimate the actual incident rate. For example, many crimes are unreported including those that involve minor assaults and for those more serious incidents that require medical attention. A recent injury surveillance study indicated that 70% of assaults that require medical attention are not reported to police (Violence Reduction Unit, 2006).
Drawing on a range of formal and informal data, the figures reveal high rates of violent offending. Statistics from both criminal justice and health
1 Figures taken from Prison Statistics Scotland (2000), published in November 2001. Note that the figures relate to the ‘main’ crime for which the individual has been sentenced.
confirm this. For example, the Scottish Crime Survey (McVie et al., 2003) found that 24% of all crimes reported involved violence (i.e. robbery and/or assault). In 2002, there was an estimated 174,000 incidents of minor incidents and 46,000 serious incidents of violent offending. Since 1992, the figures suggested that there has been an increase in violent offending of approximately 45%. According to Prison statistics, of those prisoners in custody on 30 June 2005, 22% were remanded and 40% convicted of non-sexual crimes of violence
Considering data from mental health services, there are currently 272 restricted status patients detained in hospital and 47 in the community. Anecdotal evidence suggests that there are larger numbers of patients who pose a risk of violence but who are not restricted and are being managed on an informal basis in the community. For example, in Glasgow alone a service for mentally disordered patients with co-occurring „severe/serious‟ violence risks, has an estimated 480 out-patients being
It has been estimated that the treatment of violence costs the Scottish NHS between ?258 million and ?517 million per annum. While there are no absolute figures on the cost of violence to the Scottish economy each year, one economist has suggested the figure could be as high as ?3 billion (Violence Reduction Unit, 2006).
Needs Analysis for Forensic Mental Health Units.
Six forensic units in Scotland were contacted to gain statistics on reported violent incidences occurring on the wards. The units contacted were of varying levels of security ranging from low to high and included the private sector. Four responses were received and the figures are presented below.
VIOLENCE RELATED INCIDENTS OVER A THREE YEAR PERIOD
2004-2006 2004-2006 2003-2005 2003-2006
High Low Low Low
Security Security Security Security
(230) (41) (27) (12)
Violent 769 43 70 98
Assault 1368 34 45 27
Other 1148 46 - -
Total 3285 123 115 125
A more detailed breakdown of these figures is provided in Appendix 1.
Violence has also been found to be the most common reason for re-admission to high security hospital in Scotland with almost half of re-admissions being due to violent behaviour (n =59, 47.9%) (Duncan et al. 2002).
The figures do not take account of violent index offences or histories of violent behaviour outwith the secure environment. It is recognised that not all violent incidents (especially verbal abuse) are reported and the true extent of these figures may be much higher. These figures also do not reflect incidents occurring with community forensic teams. Moreover, the figures do not identify the proportion of patients who are accountable for the figures. Research has indicated that most violent incidents in institutions are perpetrated by a small proportion of patients/offenders. Gardner, Lidz, Mulvey & Shaw (1996) for example reported that 45% of all violent incidents in a psychiatric hospital were perpetrated by only 5% of patients. A similar picture has been reported at the State Hospital, Carstairs.
Given the level of violent incidents within our communities and our institutions it is surprising that when this Reference Group asked Forensic Network members about its assessment and treatment of violent behaviours, it was found that only the Scottish Prison Service currently has a formal treatment programme whose specific focus is preventing future violence. However, this programme is designed for offenders and is not suitable for those with a major mental illness. Anecdotal evidence suggests that mentally disordered patients with violent behaviour problems are more likely to receive individual treatment from clinicians based in forensic mental health services. Unfortunately, in the absence of any systematic research concerned with the treatment model, modality or effectiveness, it is not possible at this stage to comment with any degree of certainty what the service provision in Scotland actually is.
The figures demonstrate that a substantial amount of violent incidents occur in forensic mental health units and indicate a requirement for violence reduction programmes to be developed and implemented throughout the network.
2. Review of Psychological Approaches for the Treatment of
Traditionally the assessment and treatment of violent offenders has focused on a unidimensional criminogenic variable namely poor anger control (Michie & Cooke, 2006). This implies an intrinsic link between anger and violence. While anger can clearly be a component of violence
(Jones & Hollin, 2004), “the emotion of anger is neither a necessary nor a sufficient condition of human aggression and violence” (Howells, 2004, pg.
189). Indeed for some offenders their violent acts are characterised by the absence of anger in that they are wholly instrumental in nature. This review will begin by defining the concept under discussion before briefly reviewing some of the key theoretical and empirical literature.
The extent to which violence has become embedded within the anger management literature is such that the terms “violence” and “anger” are often used interchangeably (Howells, 2004). This paper would argue that there are vital differences.
For the purposes of this paper, the definition of violence used has been taken from the manual for the HCR-20 which defines violence as “actual,
attempted, or threatened harm to a person or persons….Violence is behaviour which is obviously likely to cause harm to another person or persons” (Webster, Douglas, Eaves & Hart, 1997, pg. 24). The definition includes behaviours such as stalking but does not include violent acts against animals or acts of reasonable self defence. Of note, it plainly defines violence as a behaviour, not an emotion such as anger. This definition was chosen because it enables a clear link to be made between violence risk assessment (through the HCR-20) and the treatment of violence.
For the purposes of this report it was decided that domestic violence would not be included. While a very significant problem, current research suggests that domestic violence requires a specialist intervention rather than a general intervention for violent offending (Wong & Gordon, 2000a). Similarly terrorism, suicide/self-harm and sexual violence (which is being covered by another Reference group) was considered to be beyond the remit of this panel. While females and adolescents contribute to the levels of violence found within our communities and institutions, the panel focused primarily on adult males as this group is responsible for a significant proportion of detected violence and has the most developed treatment programmes available. A recent report on the needs of young people with mental health problems and at risk of offending, has confirmed the need to tailor assessments and interventions to the specific needs of young people rather than merely adapting adult programmes (Marshall & Irvine, 2006).
Overview of the Literature
A thorough review of OVID databases was carried out to identify pertinent research papers examining the psychological interventions for violence. The search highlighted the general paucity of research in this area and the absence of any „gold standard‟ randomised control trials (RCTs) for violence treatment. This review will therefore focus on meta-analyses and
key theoretical and empirical studies in an attempt to highlight the salient aspects of the current research in this area.
It is widely recognised that treatment programmes for offenders that rely on any theoretical principal are five times more effective than those with no specific theoretical basis (Izzo & Ross, 1990). However as yet no single theory or amalgamation of theories has been presented that explains the development and maintenance of violent behaviour. The heterogeneity of factors identified as being related to violent behaviour suggests the identification of a single guiding theory is unlikely (Howells & Day, 2002; Polaschek & Collie, 2004). Instead research has suggested that an amalgamation of several theories might be useful. Theories that have been formulated in relation to violence include social learning, cognitive behavioural, and attachment theories (Timmerman & Emmelkamp, 2005).
The only violence programme currently operating in Scotland is the Violence Prevention Programme (VPP) in the Scottish Prison Service. A detailed description of this will be provided below but the theoretical background to it is presented here as it is broadly similar to the theoretical background of the other major programmes identified by the panel. The VPP model of intervention is based on several assumptions about human behaviour. These assumptions are drawn from social learning theory and cognitive-behavioural approaches. Research has consistently demonstrated that cognitive-behavioural interventions for changing criminal behaviour are the most effective and efficient (Glick, 2003). The model of change utilised in the VPP is augmented by the therapeutic methods of motivational interviewing (Miller & Rollnick, 2002), relapse prevention (Marlatt & Donovan, 2005) and self-management (Ward and Hudson, 2000). All of these psychological theories and therapeutic approaches are integrated with the principles of effective correctional treatment (Andrews, Bonta, and Hoge, 1990; Andrews & Bonta, 2003).
The principal theoretical orientation of the VPP is social learning theory. Social learning theorists contend that violent behavioural tendencies are learned through mechanisms such as observation, vicarious experiences (seeing others rewarded for aggression) or direct experiences (being aggressive and receiving rewards). Implicit in this model is that violent behaviour is subject to modification in the same manner as any other behaviour. Both pro-social and violent behaviour are assumed to be developed and maintained by external situational events, external reinforcement, and by internal processes such as attitudes, appraisals, and emotions. The VPP focuses on the interactions between the individual and the environment in shaping the pattern of acquiring, maintaining and instigating violent behaviour.
To effect self-improvement, the VPP utilises the principles, methods and procedures of cognitive-behavioural treatment (Beck, 1976; Ellis, 1977; Goldstein & Glick, 1987; Deffenbacher, 1992, Glick, 2003). Cognitive behaviour therapy includes a variety of techniques and strategies that are