Home > Clinical Guides > Antisocial Personality Disorder:... > INTERVENTIONS FOR PEOPLE WITH ANTISOCIAL...

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

National Collaborating Centre for Mental Health (UK). Antisocial Personality Disorder: Treatment, Management and Prevention. Leicester (UK): British Psychological Society; 2010. (NICE Clinical Guidelines, No. 77.)

7INTERVENTIONS FOR PEOPLE WITH ANTISOCIAL PERSONALITY DISORDER AND ASSOCIATED SYMPTOMS AND BEHAVIOURS

7.1. INTRODUCTION

Both psychological and pharmacological interventions for people with antisocial personality disorder are poorly researched and direct evidence on the treatment of this population is scarce. Three relatively recent reviews failed to identify any high-quality evidence for people receiving treatment for antisocial personality disorder (Salekin, 2002; Warren et al., 2003; Duggan et al., 2007a).

A number of approaches have been adopted to address this problem: the use of lower quality evidence, including evidence such as case studies and case series (for example, Salekin, 2002); the use of research on other personality disorders or mixed populations of people with personality disorder including a proportion with antisocial personality disorder, usually a relatively small proportion (for example, Warren et al., 2003); and consideration of the impact of treatments for comorbid problems (such as drug misuse) in antisocial personality disorder populations (Duggan et al., 2007a). All three approaches are problematic in guiding treatment choice for antisocial personality disorder because of difficulties understanding causality (Salekin, 2002), lack of generalisability (Warren et al., 2003), and the lack of direct evidence for the treatment of the disorder itself (Duggan et al., 2007a).

In order to address these limitations, the GDG opted to identify the best available evidence on:

  1. the treatment of people with antisocial personality disorder – this was to ensure that new studies or studies excluded by other reviews could be considered
  2. the treatment of specific components of the diagnostic construct of antisocial personality disorder (for example, impulsivity and aggression) – this was to include important evidence on the treatment of a particular aspect of antisocial personality disorder
  3. interventions for offenders that aim to reduce re-offending – this was considered important because offending and related behaviours are both key to the difficulties associated with antisocial personality disorder.

The GDG recognised that the inclusion of evidence for offending behaviour was potentially controversial and that offending behaviour might be seen as a poor proxy outcome in the treatment of antisocial personality disorder. The rationale for using offending behaviour as a proxy for a diagnosis of antisocial personality disorder (where the latter has not been recorded) is threefold. First, a history of antisocial behaviour is a particular feature of antisocial personality disorder in the DSM-IV diagnostic system (APA, 2000), specifically the ‘failure to conform to social norms with respect to lawful behaviours as indicated by repeatedly performing acts that are grounds for arrest’. Second, interventions aimed at reducing offending behaviour often focus on other diagnostic criteria of antisocial personality disorder as mediating variables in the treatment process. To date, such work has included studies of impulsivity, aggressiveness, and lack of remorse as ‘treatment targets’. Therefore, evidence that has a bearing on the amelioration of these factors is also potentially relevant to the treatment of antisocial personality disorder. Third, surveys of offenders very often find high rates of personality disorder that are significantly above the levels found in community-based studies of prevalence, in particular among those who are imprisoned and those with entrenched patterns of more serious offences. As noted earlier, a survey for the UK Office for National Statistics interviewed 3,142 prisoners and found that 49% of male sentenced prisoners, 63% of males on remand, and 31% of female prisoners met criteria for diagnosis of antisocial personality disorder (Singleton et al., 1998).

7.1.1. Treatment of comorbid disorders

Given the limited evidence for the treatment of antisocial personality disorder and that guidance on disorders commonly comorbid with antisocial personality disorder generally does not consider the impact of antisocial personality disorder on treatment recommendations, the GDG decided to review the evidence for the treatment of comorbid disorders. The evidence on the treatment of comorbid disorders was restricted to populations with antisocial personality disorder, and evidence was not extrapolated from studies of offenders or other populations. In the review of interventions for offending behaviour, the GDG also decided to include studies of interventions for drug and alcohol misuse and dependence in offender populations where such studies met quality criteria.

7.2. PSYCHOLOGICAL INTERVENTIONS FOR ANTISOCIAL PERSONALITY DISORDER

7.2.1. Introduction

There has been little formal development of psychological interventions specifically for the treatment of antisocial personality disorder with considerably more emphasis placed on the psychological treatment of other personality disorders, primarily borderline personality disorder (for example, Kernberg, 1984; Linehan, 1997). As with personality disorder more generally, psychoanalytic approaches to treatment held sway initially (Cordess & Cox, 1998); more recently developments in cognitive behavioural treatments have emerged but such approaches in antisocial personality disorder are not supported by a strong evidence base (Duggan et al., 2007a).

Psychological interventions for comorbid disorders are, by contrast, well developed and are as effective or more effective than pharmacological treatments for common mental disorders (for example, NCCMH, 2005a, 2005b, 2006). This suggests that such interventions may have a significant role to play in the treatment of comorbid disorders in antisocial personality disorder. Similarly effective psychological treatments for drug and alcohol disorders have also been developed (NCCMH, 2007a) and may again be of benefit to people with antisocial personality disorder with comorbid drug and alcohol problems.

Although psychological interventions specifically for antisocial personality disorder are limited, interventions for some of the components of the antisocial personality disorder diagnostic construct have been better developed, principally for the treatment or management of aggression. However, the relevance of anger management as an intervention for an aspect of the antisocial personality disorder diagnostic construct may be limited. Anger is not explicitly included in the diagnostic criteria for antisocial personality disorder and while anger may be related to impulsivity and aggression, reducing anger may not reduce impulsivity and aggression. Equally, when delivered to offenders, anger management interventions may reduce levels of anger without having an impact on offending, aggressive or violent behaviours if the causes of those behaviours in an individual are unrelated to anger. The majority of literature on anger management has focused on populations of college students (Edmonson & Conger, 1996; Del Vecchio & O’Leary, 2004; Beck & Fernandez, 1998). The GDG felt that it would not be appropriate to extrapolate from college students with elevated levels of anger to people with antisocial personality disorder. As a consequence, this review is not concerned with the efficacy of anger management in these populations.

In contrast to the limited development of specific treatment for antisocial personality disorder, there has been considerable development of interventions aimed at reducing offending behaviour. These include a wide range of cognitive and behavioural interventions (for example, Landenberger & Lipsey, 2005; Lipsey et al., 2001, 2007; Lipton et al., 2002; Tong & Farrington, 2006; Wilson et al., 2005), and to a lesser extent therapeutic communities (Lees et al., 2003). Within the UK criminal justice system the use of cognitive and behavioural interventions such as Reasoning and Rehabilitation (for example, Cann et al., 2003) and Enhanced Thinking Skills (for example, Friendship et al., 2002) is widespread.

Current practice

Healthcare services

Most people with antisocial personality disorder in the community remain undiagnosed and untreated (Department of Health, 2003). They do not come into contact with mental health services and often do not perceive any need for treatment of their personality problems. Some people with the disorder may seek treatment for comorbid mental health disorders, including anxiety and depression, but whether they have a formal diagnosis of antisocial personality disorder or not, they may nevertheless be excluded from services because of their personality disorder or the mistaken belief that they will not be able to benefit from treatment. People with antisocial personality disorder may also make limited use of inpatient services in a crisis but are unlikely to be offered or engage in long-term treatment.

In contrast to mental health services, a significant number of people with antisocial personality disorder are treated by drug and alcohol services in both the statutory and non-statutory sector. Here the focus on treatment will be on the drug or alcohol misuse not the personality problem.

Health services treating people specifically for their antisocial personality disorder are largely limited to specialist healthcare services such as forensic services. However, even within forensic services specific provision for antisocial personality disorder is underdeveloped. At the very severe end of the spectrum the recent development of the Dangerous and Severe Personality Disorder Service (Home Office, 1997) has seen the establishment of new units in two special hospitals (Rampton and Broadmoor) and two high secure prisons (HMP Frankland and HMP Whitemoor).

The criminal justice system

The large majority of people receiving interventions for antisocial personality disorder and related problems will be in the criminal justice system, with the interventions provided either by the probation or prison services. The explicit aim of these interventions is to reduce offending behaviour. These interventions are highly manualised and subject to stringent quality assurance and auditing (T3 Associates, 2003). Whether individuals in the criminal justice system receive interventions will depend on a range of factors including the availability of places on offending behaviour programmes in the institution or probation service, the type and length of their sentence (as this may or may not facilitate their enrolment in a programme), and, if they are in prison, whether they voluntarily choose to enrol on a programme.

The majority of psychological interventions delivered in the criminal justice system are cognitive behavioural and largely based on social learning theory, a development of behavioural learning models that has been adapted to take account of findings from cognitive and developmental psychology (Bandura, 2001). These interventions include: behaviour modification; relaxation training; systematic desensitisation; social skills training; problem-solving therapy; cognitive therapy; and moral reasoning or moral reconation therapy. Virtually all of these methods have been employed in efforts to reduce offending behaviour and this represents the largest research base of evidence for interventions with offenders. The literature has been reviewed in a number of meta-analyses (for example, Lipton et al., 2002; Landenberger & Lipsey, 2005; Tong & Farrington, 2006; Lipsey et al., 2007).

Beyond the health and criminal justice system interventions, the provision of care and support for people with antisocial personality disorder is also very limited. As they may cause disruption and a threat to staff or other services users, people with antisocial personality disorder may find themselves excluded from a range of services that might otherwise support them in the community (including during transition from the care of the criminal justice system to the community), such as housing, welfare and employment services.

7.2.2. Definition and aim of review

The review considered psychological interventions for antisocial personality disorder and its constructs. This included interventions for people specifically diagnosed with antisocial personality disorder, but also interventions for the symptoms or behaviours associated with this diagnostic construct including anger, impulsivity and aggression. However, studies of populations with diagnoses of serious mental illness (including schizophrenia) were excluded. In addition, interventions for offending behaviour in people without a diagnosis of antisocial personality disorder were considered, including for offenders with substance misuse problems.

Outcomes

For the review of the effectiveness of interventions for adults with antisocial personality disorder, the GDG chose re-offending as the primary outcome. There are a number of measures of re-offending including conviction, arrest, breaches of conditions attached to parole or probation, re-incarceration and recidivism. Conviction was considered the most robust measure but where this was not reported other re-offending outcomes were extracted in the order of priority listed above.

7.2.3. Databases searched and inclusion/exclusion criteria

Information about the databases searched and the inclusion/exclusion criteria used for this section of the guideline can be found in Table 28 (further information about the search for health economic evidence can be found in Appendix 11.)

Table 28. Databases searched and inclusion/exclusion criteria for clinical evidence.

Table 28

Databases searched and inclusion/exclusion criteria for clinical evidence.

The review team conducted a series of systematic searches for RCTs that assessed the efficacy and cost effectiveness of psychological interventions specifically for the treatment of antisocial personality disorder, behaviours or symptoms associated with the antisocial personality disorder construct, and offending behaviour (see Table 31).

Table 31. Study information table for group-based cognitive and behavioural intervention compared with non-treatment control for substance misuse offenders.

Table 31

Study information table for group-based cognitive and behavioural intervention compared with non-treatment control for substance misuse offenders.

One trial (DAVIDSON2008) met the eligibility criteria of the GDG in the first systematic search to assess the treatment of antisocial personality disorder.

Two further searches were conducted separately on behaviours and symptoms associated with the antisocial personality disorder construct, and on offending behaviour.

7.2.4. Studies considered11

A total of 22 trials relating to clinical evidence met the eligibility criteria set by the GDG, providing data on 3,237 participants. Of these, two trials were reported in books (JOHNSON1995, PORPORINO1995), two were reports from the US Department of Justice (AUSTIN1997, PULLEN1996), and 18 were published in peer-reviewed journals between 1973 and 2008 (ARMSTRONG2003, DAVIDSON2008, DEMBO2000, DUGAN1998, ELROD1992, GREENWOOD1993, GUERRA1990, KINLOCK2003, LEEMAN1993, LIAU2004, OSTROM1971, ROHDE2004, ROSS1988, SCHLICHTER1981, SHIVRATTAN1988, SPENCE1981, VAN VOORHIS2004, VANNOY2004). In addition, 97 studies were excluded from the analysis. The most common reason for exclusion was lack of a comparison group (further information about both included and excluded studies can be found in Appendix 15).

For the treatment of people with antisocial personality disorder, there was one trial (DAVIDSON2008) that met the eligibility criteria of the review providing information on 39 participants.

For the treatment of people with symptoms or behaviour associated with the antisocial personality disorder construct, there was one trial that investigated the treatment of anger by comparing anger management with control (VANNOY2004).

For the treatment of offending behaviour in adults with substance misuse problems, there were four trials investigating cognitive and behavioural interventions, three of which were group-based interventions (AUSTIN1997; JOHNSON1995; KINLOCK2003) and one which was individually based (DUGAN1998).

For the treatment of offending behaviour in adults, there were five trials comparing group-based cognitive and behavioural interventions with control (ARMSTRONG2003; LIAU2004; PORPORINO1995; ROSS1988; VANVOORHIS2004).

For the treatment of offending behaviour in young people, seven trials compared group-based cognitive and behavioural skills interventions with control (GUERRA1990; LEEMAN1993; OSTROM1971; PULLEN1996; ROHDE2004; SCHLICHTER1981; SPENCE1981); one trial was on individual cognitive and behavioural interventions (SHIVRATTAN1988) and three trials compared multi-component interventions with control (ELROD1992; GREENWOOD1993; DEMBO2000).

7.2.5. Clinical evidence for the treatment of antisocial personality disorder

The search identified one study relating to the treatment of antisocial personality disorder (DAVIDSON2008). The study compared CBT with treatment as usual for people with antisocial personality disorder living in the community (see Table 29 and Table 30). Full study characteristics and forest plots can be found in Appendices 15 and 16 respectively.

Table 29. Study information table on CBT for treatment of antisocial personality disorder.

Table 29

Study information table on CBT for treatment of antisocial personality disorder.

Table 30. GRADE evidence summary for interventions for antisocial personality disorder.

Table 30

GRADE evidence summary for interventions for antisocial personality disorder.

DAVIDSON2008 did not find an effect for CBT on anger or verbal aggression compared with treatment as usual for people with antisocial personality disorder in the community. The trial did find a small, non-significant effect for social functioning and physical aggression compared with treatment as usual.

7.2.6. Clinical evidence summary for the treatment of antisocial personality disorder

The evidence for the treatment of antisocial personality disorder in the community is limited to one trial. The quality of the evidence is low to moderate where further research is likely to have an impact on the effect estimate of CBT in the community for people with antisocial personality disorder. The limited economic evidence from this trial suggests that CBT may not be cost saving in the short term (see below).

7.2.7. Economic evidence for the treatment of antisocial personality disorder

One economic study on psychological treatment of antisocial personality disorder was included in the systematic economic literature review (Davidson et al., 2008).

The study, which was conducted in the UK, was a simple cost analysis of CBT plus treatment as usual versus treatment as usual alone conducted alongside an RCT included in the guideline systematic review of clinical evidence (DAVIDSON2008). The study examined healthcare costs (including psychiatric care, accident and emergency visits and primary care), social work costs and costs borne by the criminal justice system. The time horizon of the analysis was 12 months. Overall, the total cost per person in the CBT group was higher than the respective cost in the treatment as usual group (£38,004 versus £31,097, respectively). The healthcare cost was similar in both groups (£1,295 in the CBT group and £1,133 in the TAU group). The cost of providing CBT was £1,300 per participant. Details on the methods used in the systematic review of the economic literature are described in Chapter 3. Evidence tables for all economic studies included in the guideline economic literature review are in Appendix 14.

7.2.8. Clinical evidence for the treatment of the constructs of antisocial personality disorder

One trial relating to clinical evidence for the treatment of the constructs of antisocial personality disorder met the eligibility criteria set by the GDG, providing data on 31 participants (VANNOY2004). The included study was a trial of group-based anger management versus waitlist in an offender population. This small study reported data only on a continuous measure and was considered to be of low quality. The outcomes of the trial were trait anger (STAXI; SMD −0.64, −1.36 to 0.09) and state anger (STAXI; SMD −0.96, −1.70 to −0.21).

7.2.9. Clinical evidence summary for the treatment of the constructs of antisocial personality disorder

The evidence for the treatment of the constructs of antisocial personality disorder is extremely limited and does not support the development of any recommendations.

7.2.10. Economic evidence for the treatment of the constructs of antisocial personality disorder

No evidence on the cost effectiveness of treatment of the constructs of antisocial personality disorder was identified by the systematic search of the literature. Details on the systematic search of the literature are provided in Chapter 3.

7.2.11. Clinical evidence for the treatment of offending in substance misuse offenders

The review found four trials that investigated cognitive and behavioural interventions for the treatment of offending in substance misuse offenders, three of which were group based interventions (AUSTIN1997; JOHNSON1995; KINLOCK2003) and one was individually based (DUGAN1998). This review provided data on 582 participants (see Table 31 and Table 32). Full study characteristics and forest plots can be found in Appendices 15 and 16 respectively.

Table 32. GRADE evidence summary for the treatment of offending in substance misuse offenders.

Table 32

GRADE evidence summary for the treatment of offending in substance misuse offenders.

For the treatment of offending in substance misuse offenders, the five included studies were identified as cognitive and behavioural interventions. The review found this intervention to have a medium effect on offending and major infractions combined (RR = 0.76; 0.60, 0.97) and a small non-significant effect on mean number of offences (SMD 0.19; −0.18 to 0.55).

7.2.12. Clinical evidence summary for the treatment of offending in substance misuse offenders

There appears to be modest evidence for the effectiveness of cognitive and behavioural interventions, primarily delivered in groups, in reducing offending for adults with substance misuse problems. This effect has been found in a variety of settings including institutional prison-based settings and outpatient and probation settings in the community.

7.2.13. Economic evidence for the treatment of offending in substance misuse offenders

One study met the inclusion criteria for the systematic economic literature review (Alemi et al., 2006). The study, which was conducted in the US, compared the costs over 2.75 years of a combination of probation and substance misuse treatment versus probation alone. Overall, a combination of probation and treatment was $6,300 more expensive than traditional probation per participant annually, mainly because of greater mental hospitalisation and additional treatment costs. The study characteristics and results are presented in the form of evidence tables in Appendix 14. Details on the systematic search of the economic literature are provided in Chapter 3.

7.2.14. Clinical evidence for the treatment of offending behaviour in adults

There were five trials comparing the effects of group-based cognitive and behavioural interventions with control on re-offending for adult offenders treated within the criminal justice system (institutional settings or in the community on probation or parole) (see Table 33 and Table 34). Conviction was considered the most robust measure of re-offending but where this was not reported, other re-offending outcomes were extracted (for further details see Section 7.2.2). Full study characteristics and forest plots can be found in Appendices 15 and 16 respectively.

Table 33. Study information table for group-based cognitive and behavioural interventions for offenders.

Table 33

Study information table for group-based cognitive and behavioural interventions for offenders.

Table 34. GRADE evidence summary for group-based cognitive and behavioural intervention for offenders.

Table 34

GRADE evidence summary for group-based cognitive and behavioural intervention for offenders.

Group-based cognitive and behavioural interventions were found to provide a modest effect on re-offending (RR 0.78; 0.55 to 1.08). The population included in this analysis was predominantly adult male offenders. LIAU2004, which included a small proportion of female offenders, was not included in the meta-analysis because it was not possible to extract intention-to-treat (ITT) data.

7.2.15. Clinical evidence summary for the treatment of offending behaviour in adults

There appears to be modest evidence for the effectiveness of group-based cognitive behavioural skills interventions, delivered in community and institutional settings, in reducing offending for adults in the criminal justice system.

Group-based cognitive behavioural skills interventions for offending behaviour delivered to offenders in criminal justice settings (prison/institutional settings and probation/parole) have a small but positive effect on the rate of re-offending for adult male offenders aged 21 and over. However, the more limited evidence base on young adult offenders aged 18 to 20 indicates that young offenders do not respond to these interventions.

7.2.16. Health economic evidence for the treatment of offending behaviour

Systematic literature review

One US study focusing on interventions targeted at adult offenders was identified by the systematic search of economic literature (Zhang et al., 2006). The study evaluated a state-wide multiple community-based services parole programme in California by comparing programme costs with incarceration costs avoided because of decreases in recidivism. Over 2 years, programme participants had lower recidivism and reincarceration rates than the untreated population, resulting in significant net savings of $21 million.

Economic modelling

Objective

The guideline systematic review and meta-analysis of clinical evidence demonstrated that provision of Reasoning and Rehabilitation, a group-based cognitive behavioural skills intervention (Cann et al., 2003), to adult offenders can potentially reduce the rates of future offending behaviour. Offending behaviour leads to substantial costs to the society, including the criminal justice system and victims of crime. A cost analysis was undertaken to assess whether the costs to the NHS of providing Reasoning and Rehabilitation to adults with offending behaviour are offset by future cost savings resulting from reduction in re-offending behaviour in this population.

Methods

Intervention examined Reasoning and Rehabilitation programmes are offered to people with offending behaviour in institutional and community correctional settings. They typically consist of 38 curriculum-based sessions of 2 hours’ duration each over approximately 8 to 12 weeks. Programmes are delivered to small groups of eight to ten participants (T3 Associates, 2003).

Costs considered in the analysis A simple economic model was developed to estimate the net total costs (or cost savings) associated with provision of Reasoning and Rehabilitation to adult offenders. Published evidence on the costs incurred by adults with offending behaviour is limited. One study conducted in the UK that assessed the effectiveness of CBT in adults with antisocial personality disorder reported 12-month service costs incurred by this population, including healthcare, social work and criminal justice system costs (Davidson et al., 2008). The total costs per adult with antisocial personality disorder receiving CBT over 12 months were £38,000. Of these, only 7% were healthcare costs (including provision of CBT); the vast majority of costs were associated with social work and use of criminal justice system services.

NICE recommends that economic analyses of healthcare interventions adopt a NHS and PSS perspective (NICE, 2006a). However, the criminal justice system and social and other public services are likely to bear the majority of costs incurred by adults with offending behaviour and only a small proportion of costs fall on the NHS and PSS. For this reason, the economic analysis adopted a broader perspective than that of the NHS and PPS, including any costs to public services for which appropriate information was available.

Existing clinical evidence suggests that provision of Reasoning and Rehabilitation to adults with offending behaviour may reduce rates of re-offending, and therefore costs relating to crime. It is unknown whether participation of adult offenders in such programmes has an effect on other costs, such as costs to health and social care services, although it is likely that reducing offending behaviour may result in a decrease in other costs too. Because of lack of appropriate relevant data that could inform the economic model, the analysis has considered only intervention costs (that is, costs of providing Reasoning and Rehabilitation) and costs related to crime/adult offending behaviour. All other categories of public sector costs, such as health and social care costs, were conservatively assumed to be the same for adult offenders participating in Reasoning and Rehabilitation programmes and for those not receiving the intervention, and were subsequently omitted from the analysis. This is acknowledged as a limitation of the economic analysis. However, costs relating to crime are likely to constitute the most substantial part of the costs incurred by adult offenders; therefore, the economic analysis is likely to have considered the majority of costs associated with providing Reasoning and Rehabilitation to adults with offending behaviour.

Model input parameters

Clinical efficacy of Reasoning and Rehabilitation and baseline re-offending rate in adult offenders

Clinical data on re-offending rates associated with Reasoning and Rehabilitation were taken from three studies (PORPORINO1995; ROSS1988; VAN VOORHIS2004). Meta-analysis of these data undertaken for the guideline showed that the intervention reduced the rate of re-offending in adult offenders compared with no treatment, but results were non-statistically significant at the 0.05 level (mean RR of re-offending for Reasoning and Rehabilitation versus control: 0.78; 95% CI: 0.55 to 1.08). These results were characterised by considerably high heterogeneity caused by inclusion of ROSS1988 in the meta-analysis. When this study was removed, there was no heterogeneity in the results but the effect of the intervention was reduced (mean RR of re-offending for Reasoning and Rehabilitation versus control: 0.88; 95% CI: 0.75 to 1.03). Details of the clinical studies considered in the economic analysis are available in Appendix 15. The forest plots of the respective meta-analyses are provided in Appendix 16.

The baseline re-offending rate for adults with previous offending behaviour was taken from a national report containing 12-month data on re-offending for adults released from custody or commencing a court order (sentences under probation supervision excluding fines) in England and Wales in 2006 (Ministry of Justice, 2008a). According to this document, the re-offending rate in this population was 39% over 12 months. This rate was determined by the number of offenders in the cohort offending at least once during the 12-month follow-up period, where the offence resulted in a court conviction or an out-of-court disposal. The 12-month rate of adult re-offending following provision of Reasoning and Rehabilitation in the economic analysis was calculated by multiplying the estimated RR of re-offending of the intervention versus control by the baseline re-offending rate.

Intervention costs (costs of providing the Reasoning and Rehabilitation programme)

In order to estimate total intervention costs, relevant resource use was estimated and combined with respective unit costs. Resource use estimates associated with provision of a Reasoning and Rehabilitation programme were adopted from T3 Associates (2003) and were consistent with resource use described in studies providing the efficacy data for this analysis. According to these estimates, the evaluated intervention consisted of 38 sessions lasting 2 hours each, delivered to groups of eight adults with offending behaviour.

The unit cost of therapists providing Reasoning and Rehabilitation was assumed to equal that of clinical psychologists (Band 7) due to lack of more relevant unit cost estimates. However, it is recognised that therapists providing Reasoning and Rehabilitation may be on a lower salary scale, and therefore the total intervention cost may have been overestimated. The national unit cost of clinical psychologists has been estimated at £67 per hour of client contact in 2006/07 prices (Curtis, 2007). This estimate was based on the mid-point of Agenda for Change salaries Band 7 of the April 2006 pay scale according to the National Profile for Clinical Psychologists, Counsellors and Psychotherapists (NHS, 2006). It includes salary, salary on costs, overheads and capital overheads but does not take into account qualification costs as the latter are not available for clinical psychologists.

Based on the above resource use estimates and the unit cost of clinical psychologists, the cost of providing the Reasoning and Rehabilitation programme was estimated at £637 per adult with offending behaviour at 2006/7 prices.

Costs of adult offending behaviour/cost savings from reducing adult re-offending rates

In order to estimate the annual cost resulting from re-offending behaviour by adult offenders, three types of data are needed:

  • proportion of different types of offences committed by adult re-offenders
  • costs associated with each type of offence
  • number of offences per adult re-offender per year.

Data on the proportion of different types of offences committed by adult re-offenders in England and Wales were derived from a national report published by the Ministry of Justice (2008a). The same document reported that the number of offences per adult re-offender were 3.742 over 12 months.

Costs associated with each type of offence committed by adult offenders were taken from a variety of sources, as reported in Chapter 5, Section 5.4.14. Costs were uplifted to 2007 prices using the Retail Prices Index (Office for National Statistics, 2008). The cost per offence committed by adult re-offenders was estimated as the mean cost of all offences weighted by the proportion of offences committed on average by an adult re-offender. Table 35 shows the percentage of offences committed by adult re-offenders, the cost of each type of offence as estimated in the literature, and the weighted average cost per offence committed by adult re-offenders.

Table 35. Percentage and costs of offences committed by adult re-offenders.

Table 35

Percentage and costs of offences committed by adult re-offenders.

The average cost per offence committed by adult re-offenders was estimated at £2,706. Since this population commits 3.742 offences over 12 months (Ministry of Justice, 2008a), the 12-month cost associated with offending behaviour is £10,127 per adult re-offender.

Time horizon of the analysis

The three studies included in the guideline meta-analysis of clinical data on Reasoning and Rehabilitation had time horizons ranging between 4 and 9 months. It is not known whether the beneficial effect of Reasoning and Rehabilitation lasts beyond 9 months. Therefore, for the base-case analysis, a 1-year time horizon was chosen; alternative time horizons up to 5 years were tested in sensitivity analysis, to explore the magnitude of potential savings that could be gained if the intervention has a longer lasting effect.

Discounting

Costs incurred beyond 12 months were discounted at an annual rate of 3.5%, as recommended by NICE (NICE, 2006a). Table 36 provides all input parameters utilised in the base-case analysis of the economic model of Reasoning and Rehabilitation for adults with offending behaviour.

Table 36. Input parameters utilised in the economic model assessing the net costs (or savings) resulting from provision of Reasoning and Rehabilitation to adults with offending behaviour.

Table 36

Input parameters utilised in the economic model assessing the net costs (or savings) resulting from provision of Reasoning and Rehabilitation to adults with offending behaviour.

Sensitivity analysis

One- and two-way sensitivity analyses were undertaken to explore the robustness of the results under the uncertainty characterising some model input parameters. The following scenarios were tested in sensitivity analysis:

  • Use of the 95% CIs of the RR of re-offending of Reasoning and Rehabilitation versus control.
  • Exclusion of data from ROSS1988, which introduced heterogeneity in the meta-analysis (resulting in a mean RR of re-offending of Reasoning and Rehabilitation versus control: 0.88 with 95% CIs 0.75 to 1.03).
  • Reduction in the baseline re-offending rate for adult offenders; an annual rate of 30% was tested.
  • Extension of the time horizon of the analysis beyond 1 year; although currently there is no evidence to suggest that Reasoning and Rehabilitation programmes have a clinical effect lasting longer than 1 year, consecutive time horizons of 2 to 5 years were tested in sensitivity analysis to explore the magnitude of potential cost savings achieved by provision of the intervention to adult offenders, if the intervention has a longer lasting effect.
  • Potential net savings accrued over 2 to 5 years were also estimated assuming that the effect of the intervention was reduced over time; in this scenario the RR of Reasoning and Rehabilitation versus control was multiplied by a factor of 1.15 for every year after the first year following initiation of the intervention, to capture this assumed decline in the clinical effect over time, until Reasoning and Rehabilitation had no beneficial effect over control.
  • Combination of alternative time horizons between 1 and 5 years with the rest of the hypotheses described above.

In addition, threshold analyses identified the values of specific input parameters where the results of the analysis were reversed. The parameters tested were the relative effect of Reasoning and Rehabilitation versus control (expressed in RR), the average cost of offence committed by adult re-offenders, and the baseline re-offending rate of adults with offending behaviour over 12 months.

Results

Base-case analysis The reduction in the re-offending rates achieved by provision of Reasoning and Rehabilitation to adult offenders yielded cost savings equalling £869 per adult with offending behaviour over 1 year. Because the provision of Reasoning and Rehabilitation programmes costs £637 per adult offender, the intervention results in an overall net saving of £232 per adult with offending behaviour over 1 year. Full results of the base-case analysis are reported in Table 37.

Table 37. Results of economic analysis assessing the net costs (or savings) resulting from provision of Reasoning and Rehabilitation to adults with offending behaviour.

Table 37

Results of economic analysis assessing the net costs (or savings) resulting from provision of Reasoning and Rehabilitation to adults with offending behaviour.

Sensitivity analysis Results of the cost analysis were sensitive to the different scenarios tested in sensitivity analysis. The results of meta-analysis (both including and excluding ROSS1988) were not statistically significant at the 0.05 level and therefore using the upper 95% CI of the RR of the intervention versus control did not lead to any savings because offending behaviour was in these cases increased following provision of Reasoning and Rehabilitation to adult offenders. In all other scenarios Reasoning and Rehabilitation resulted in net savings within the first year. Although no long-term studies could demonstrate whether the beneficial effect of the programme in reducing offending behaviour lasts beyond 1 year, sensitivity analysis showed that, if such a longer effect exists, then the intervention could save on average £3,424 per adult offender over 5 years (or £1,578, when ROSS1988 was excluded from analysis).

Threshold analysis showed that the intervention became cost neutral over 1 year when the cost per offence committed by adult offenders fell at £1,980, when the baseline rate of re-offending was reduced at 29% over 12 months, and when the RR of the intervention versus control was a maximum of 0.84.

Full results of one- and two-way sensitivity analyses are presented in Table 38.

Table 38. Results of sensitivity cost analysis of providing Reasoning and Rehabilitation to adults with offending behaviour.

Table 38

Results of sensitivity cost analysis of providing Reasoning and Rehabilitation to adults with offending behaviour.

In addition, threshold analyses identified the values of specific input parameters where the results of the analysis were reversed. The parameters tested were the relative effect of Reasoning and Rehabilitation versus control (expressed in RR), the average cost of offence committed by adult re-offenders, and the baseline re-offending rate of adults with offending behaviour over 12 months.

Discussion – limitations of the analysis

The results of the economic analysis indicate that Reasoning and Rehabilitation programmes for adults with offending behaviour might be cost saving from a wide economic perspective in the UK. The substantial intervention costs, resulting from the high intensity of such programmes, could be offset by savings from a reduction in the rates of re-offending in adults. However, economic results were characterised by uncertainty, as revealed in sensitivity analysis. This uncertainty was caused by the statistical insignificance characterising the clinical data utilised in the economic model.

Although adult offenders incur a wide variety of costs, such as health and social service costs and costs to the criminal justice system (Davidson et al., 2008), the economic analysis considered only intervention costs and costs relating to offending behaviour, owing to lack of evidence for a difference in other costs between Reasoning and Rehabilitation and no treatment. Cost data on offending behaviour were derived from several published sources reporting UK data and included, in most cases, a wide range of costs, such as costs incurred in anticipation of offending behaviour, for example security expenditure, costs directly resulting from offending, such as costs of stolen or damaged property, emotional and physical impact on victims, costs of offering health and other services to victims, as well as costs to the criminal justice system. Although it is acknowledged that omission of other health and social care costs constitutes a limitation of the analysis, existing evidence indicates that costs of offending behaviour are probably the most significant costs incurred by adult offenders. Moreover, the intervention reduces offending behaviour and this can potentially lead to a reduction in other costs such as healthcare costs and social benefit payments. If this is the case, then the economic analysis has only underestimated the net savings gained from Reasoning and Rehabilitation programmes. Moreover, some cost data utilised in the economic analysis consisted exclusively of costs to the criminal justice system. Other costs, such as healthcare costs and emotional distress of victims, the financial and economic burden to the families of both victims and offenders, and the feelings of fear and insecurity at anticipation of crime were not considered in most documents reporting cost data on offending behaviour. Consideration of these factors might increase the reported figures on cost-savings resulting from reduction in offending behaviour achieved by offering Reasoning and Rehabilitation programmes to adult offenders.

Rates of re-offending are higher in adults with more severe offending behaviour, as expressed by the number of previous offences they committed in the past. Moreover, this population commits a higher numbers of offences per year (Ministry of Justice, 2008a). Therefore, providing Reasoning and Rehabilitation to adults with more serious history of offending behaviour is likely to lead to higher cost savings from reduction in offending behaviour.

Reasoning and Rehabilitation programmes are intensive interventions and are therefore characterised by high costs. It is possible that savings resulting from reduction in re-offending do not outweigh intervention costs. However, even if the intervention resulted in a modest net cost per person, considering the further potential benefits to participants and their families from implementation of the programme (such as increase in employment rates, reduction in drug and alcohol misuse and other healthcare costs), this cost may be justified.

The time horizon of the economic analysis was 1 year, as available evidence came from relatively short-term studies, with a maximum time horizon of 9 months. There is currently no evidence that Reasoning and Rehabilitation has a beneficial effect in reducing offending behaviour extending beyond this time. Nevertheless, in order to explore the potential magnitude of cost savings resulting from implementing the intervention, the economic analysis considered multiple consecutive time periods of 1 and up to 5 years to ascertain if the beneficial effect is retained in adult offenders. Further research is needed to explore whether the effect of the intervention lasts in the long term, as this would have substantial financial and emotional (positive) implications for society.

Conclusion

Group-based cognitive behavioural interventions delivered as Reasoning and Rehabilitation programmes are potentially cost effective in the UK setting. Besides the clinical benefits to adults with offending behaviour, they may produce net cost savings to society, resulting from reduction in offending behaviour.

7.2.17. Evidence to recommendations

There is relatively robust clinical evidence indicating that cognitive and behavioural interventions are moderately effective for offenders. The economic analysis showed that such interventions are potentially cost saving, as the intervention costs may be offset by savings associated with a reduction in re-offending; however, the results of economic analysis were characterised by great uncertainty. The finding of a reduction in re-offending is supported by evidence from cognitive and behavioural interventions for offenders with substance misuse problems, which also have a significant impact on reducing offending in a population with a high incidence of antisocial personality disorder.

The GDG judged that it would be reasonable to conclude that such interventions were likely to be effective for people with antisocial personality disorder. As was noted in the Section 7.2.1, these interventions were developed and provided almost exclusively within the criminal justice system. However, in addressing offending behaviour the interventions also attempt to focus on problems with impulsivity, aggression and rule-breaking. Such problems are also experienced by people with antisocial personality disorder without criminal records. In light of this the GDG felt it reasonable to extrapolate from this dataset of offenders and support the use of group-based cognitive and behavioural interventions for non-offending populations with antisocial personality disorder in the community.

In addition, the GDG considered that it would be possible to extrapolate these findings to people who meet criteria for DSPD and therefore concluded that cognitive and behavioural interventions would likely be moderately effective in this population. However, it was also felt that the intervention would need to be adapted in order to be beneficial for people with DSPD. The GDG also noted the recommendation in the borderline personality disorder guideline (NICE, 2009) supporting use of multi-modal treatments, for example the combination of individual and group treatments. Given that a proportion of people who meet criteria for DSPD may have comorbid personality disorders, including borderline personality disorder, the GDG considered this recommendation when formulating recommendations for antisocial personality disorder. Such modifications would include extending the nature and duration of the intervention and providing close monitoring and supervision of staff.

7.2.18. Recommendations for offending behaviour in adults

7.2.18.1.

For people with antisocial personality disorder, including those with substance misuse problems, in community and mental health services, consider offering group-based cognitive and behavioural interventions, in order to address problems such as impulsivity, interpersonal difficulties and antisocial behaviour.

7.2.18.2.

For people with antisocial personality disorder with a history of offending behaviour who are in community and institutional care, consider offering group-based cognitive and behavioural interventions (for example, programmes such as ‘Reasoning and Rehabilitation’) focused on reducing offending and other antisocial behaviour.

7.2.18.3.

When providing cognitive and behavioural interventions:

  • assess the level of risk and adjust the duration and intensity of the programme accordingly (participants at all levels of risk may benefit from these interventions)
  • provide support and encouragement to help participants to attend and complete programmes, including people who are legally mandated to do so.
7.2.18.4.

For people in community and institutional settings who meet criteria for psychopathy or DSPD, consider cognitive and behavioural interventions (for example, programmes such as ‘Reasoning and Rehabilitation’) focused on reducing offending and other antisocial behaviour. These interventions should be adapted for this group by extending the nature (for example, concurrent individual and group sessions) and duration of the intervention, and by providing booster sessions, continued follow-up and close monitoring.

7.2.19. Clinical evidence for the treatment of offending behaviour in young people

In addition to looking at adult offenders, the review also included young offenders up to the age of 17 years. Eleven trials on cognitive behavioural interventions met the inclusion criteria of the review where all but two trials were of interventions delivered in prison; OSTROM1971 and PULLLEN1996 were interventions delivered in a probation setting. Eight trials were of cognitive and behavioural interventions (GUERRA1990, LEEMAN1993, OSTROM1971, PULLEN1996, ROHDE2004, SCHLICHTER1981, SHIVRATTAN1988, SPENCE1981) and three were multi-component interventions (DEMBO2000, ELROD1992, GREENWOOD1993).

Summary study information and evidence from the included trials are shown in Table 39, Table 40 and Table 41. Full study characteristics and forest plots can be found in Appendices 15 and 16 respectively.

Table 39. Study information table for trials of interventions targeted at adolescents in the criminal justice system.

Table 39

Study information table for trials of interventions targeted at adolescents in the criminal justice system.

Table 40. GRADE evidence summary for cognitive behavioural interventions for adolescents in the criminal justice system.

Table 40

GRADE evidence summary for cognitive behavioural interventions for adolescents in the criminal justice system.

Table 41. GRADE evidence summary for multi-component interventions versus control for adolescent offenders.

Table 41

GRADE evidence summary for multi-component interventions versus control for adolescent offenders.

The evidence suggests that cognitive behavioural interventions delivered primarily in groups in institutional settings are more effective than control for reducing offending for both intent to treat data (RR = 0.62; 0.39 to 0.98) and completer only data (RR = 0.65, 0.45 to 0.95). All studies included only males, except for GUERRA1990, which had both male and female participants.

Three trials on multi-component interventions for adolescent offenders were included in the review. Two (DEMBO2000; ELROD1992) tested the efficacy of interventions delivered in the community and one (GREENWOOD1993) examined interventions provided in prison, which included an aftercare component in the community. Multi-component interventions comprised family therapy, parenting skills and cognitive problem-solving skills training (DEMBO2000); group-based cognitive and behavioural interventions and parent training (ELROD1992); and group-based cognitive and behavioural interventions and family therapy (GREENWOOD1993). These studies found the intervention to have a modest but non-significant effect on re-offending (RR 0.87; 0.65 to 1.16). ELROD1992 and DEMBO2000 involved both male and female participants while GREENWOOD1993 included only male participants. ELROD1992 was the least effective trial—in addition to group-based cognitive and behavioural interventions and parent training, there was a wilderness experience programme.

7.2.20. Clinical evidence summary for the treatment of offending behaviour in young people

There appears to be modest but statistically significant evidence for the effectiveness of group-based cognitive and behavioural interventions delivered in institutional settings in reducing offending for adolescents involved in the criminal justice system.

Multi-component interventions were less effective than the more focused group-based cognitive and behavioural interventions. This is consistent with the evidence found for multisystemic therapy. There is evidence from studies of implementation of multisystemic therapy, and other complex multimodal interventions, that maintaining fidelity to the model is strongly associated with a positive outcome. It could be that the diminished effectiveness of the multi-component interventions for offending behaviour reflected a lack of overall fidelity to or integration of the intervention.

7.2.21. Health economic evidence for the treatment of offending behaviour in young people

Four US-based studies were identified in the systematic evidence search that presented economic evaluations of interventions for young offenders (Caldwell et al., 2006; Robertson et al., 2001; Myers et al., 2000; Dembo et al., 2000a). Details on the characteristics and results reported in the studies are provided in the form of evidence tables in Appendix 14. Details on the methods used for the systematic review of the economic evidence are provided in Chapter 3.

Caldwell and colleagues (2006) compared an intensive juvenile corrective service treatment programme with usual juvenile corrective service treatment in a secured juvenile facility. The initial costs of the intensive programme were offset by improved treatment progress and lowered violent recidivism. The intensive treatment programme dominated usual treatment, resulting in lower net costs per offender and better outcomes in terms of a reduction in felony and violent offences.

Robertson and colleagues (2001) performed a cost-benefit analysis, examining local justice system expenditures associated with intensive supervision and monitoring or CBT in comparison with regular probation. They demonstrated that, relative to those on probation, the CBT programme resulted in a net saving in expenditure of $1,435 per offender during the 18-month investigation. No significant difference in justice system expenditures were demonstrated by the intensive supervision and monitoring group.

The study by Myers and colleagues (2000) was a simple cost comparison study of a multi-component intervention programme for early-career juvenile offenders. The initial costs of the programme, total costs and differences in crime rates were compared with respective costs and outcomes of an untreated community control group. Over 12 months, the programme resulted in net savings of $1,800 per young person due to lower crime rates compared with the untreated group.

Dembo and colleagues (2000a) compared the criminal justice costs of a family empowerment intervention programme versus an extended services intervention programme for juvenile offenders and their families. Over 2 years, the family empowerment intervention programme resulted in significant net savings mainly as a result of lower arrest rates.

7.2.22. From evidence to recommendations

There was consistent evidence that cognitive and behavioural interventions were effective for the treatment of offending behaviour in young people. In addition, these interventions may be cost effective, according to evidence derived from US settings. The use of such interventions for young people with offending behaviour is supported.

7.2.23. Recommendations

7.2.23.1.

For young offenders aged 17 years or younger with a history of offending behaviour who are in institutional care, offer group-based cognitive and behavioural interventions aimed at young offenders and that are focused on reducing offending and other antisocial behaviour.

7.3. TREATMENT OF COMORBID DISORDERS IN PEOPLE WITH ANTISOCIAL PERSONALITY DISORDER

7.3.1. Introduction

As highlighted in Chapter 2, people with antisocial personality disorder commonly present with comorbid mental disorders including significant drug and alcohol problems, other personality disorders and a range of common mental health problems, including depression and anxiety. The presence of these comorbidities will increase the burden of illness and may directly contribute to the exacerbation of the problems associated with the antisocial personality disorder. Unfortunately people with antisocial personality disorder often reject treatment (Tyrer et al., 2003), and even where they seek treatment for their comorbid disorders may find themselves unable to assess treatment.

Current practice

The current treatment of comorbid mental health problems falls under three broad categories: that provided by general mental health services in primary and secondary care, that provided or funded by specialist mental health services in secondary and tertiary care, and that provided within the criminal justice system.

The extent of treatment for comorbid disorders for common mental health problems such as anxiety and depression in primary and secondary mental health services is not well known. It is likely, given what is known about the epidemiology of antisocial personality disorder (for example, Robins et al., 1991; Swanson et al., 1994) that a significant number of people do seek help but their comorbid problem may not be recognised, or if they are offered treatment they may be more likely to drop out of it or not adhere to it (ESMHCG, 2005). The position with regard to the treatment of drug and alcohol problems is somewhat different, with a significant proportion of people with drug or alcohol misuse disorders receiving treatment from specialist substance misuse services provided by or funded by the NHS. This is important because alcohol misuse is associated with increased violence in people with antisocial personality disorder (Yang & Coid, 2007). An important issue is whether sufficient adaptation of drug and alcohol treatment programmes is undertaken to engage and retain people with antisocial personality disorder.

Within specialist mental health services, a small but growing number of units offer treatment specifically for personality disorder (Crawford & Rutter, 2007). In principle these units have a remit to treat antisocial personality disorder (Department of Health, 2003), but in practice few do (Crawford et al., 2007), with a much greater focus on the treatment of borderline personality disorder.

Tertiary or forensic mental health services do treat people with antisocial personality disorder and their associated comorbidities, but as noted in Chapter 4 the percentage of people with antisocial personality disorder in the care of forensic services is approximately 50% (Singleton et al., 1998).

Within the criminal justice system, there is considerable treatment of comorbid mental disorders, primarily in prison settings, which comprises two aspects. First, inmates’ general mental health is managed through prison-based mental health teams (often linked to local mental health services). These services have seen significant investment in recent years in recognition of the historically poor mental healthcare of prisoners (ESMHCG, 2005), but it is likely that for many services the concentration is on psychosis and other severe mental disorders. The second major area of activity in addressing comorbid mental health problems in prison is the treatment of drug and alcohol misuse, with many prisons now having specialist drug treatment services (usually provided by the NHS or tertiary sector services).

Definition and aim of intervention

This review was limited to the following comorbid mental health problems:

  1. drug and alcohol misuse in people with antisocial personality disorder
  2. common mental disorders in people with antisocial personality disorder.

As there was limited evidence from individual trials or systematic reviews of reasonable quality for other personality disorders in people with antisocial personality disorder, this review makes no further comment on this. The review acknowledges that the presence of comorbid personality disorder in people with antisocial personality disorder may have an overall consequence on treatment and would need to be taken into consideration when formulating a treatment plan.

Psychotic disorders were excluded from the review in large part because where antisocial personality disorder and a psychotic disorder co-exist, the primary focus of treatment would be on the psychotic disorder.

Interventions were broadly defined to include all interventions for common mental health disorders covered by the current NICE guidelines for those disorders (for example, NCCMH, 2005a). For drug and alcohol misuse interventions NICE guidelines were also used (NCCMH, 2007a, 2007b) along with other authoritative guidance (for example, Department of Health, 2007b).

7.3.2. Databases searched and inclusion/exclusion criteria

Information about the databases searched and the inclusion/exclusion criteria used for this section of the guideline can be found in Table 42.

Table 42. Databases searched and inclusion/exclusion criteria for clinical evidence.

Table 42

Databases searched and inclusion/exclusion criteria for clinical evidence.

7.3.3. Studies considered

The review team conducted a new systematic search that assessed the efficacy of the treatment of comorbid disorders for people with antisocial personality disorder.

Only one psychosocial trial reporting data relating to the treatment of comorbid substance misuse in antisocial personality disorder met the eligibility criteria set by the GDG, providing data on 108 participants with dependence on cocaine (Messina et al., 2003). This trial compared contingency management, cognitive behavioural therapy with one another and treatment as usual. In addition, there were four RCTs that assessed in post hoc analyses the impact of antisocial personality disorder (compared with absence of an antisocial personality disorder diagnosis) on the outcomes of psychosocial interventions. Two studies looked at these effects on participants with drug misuse disorders (Woody, 1983; McKay et al., 2000) and a further two trials on alcohol dependence (Wölwer et al., 2001; Hesselbrock, 1991). Five studies were excluded from the analysis. The most common reason for exclusion was either treatment or control group did not have antisocial personality disorder.

7.3.4. Clinical evidence for psychological interventions for the treatment of comorbid substance misuse

Messina and colleagues (2003) reported on a subgroup analysis of people with antisocial personality disorder receiving either contingency management, CBT, a combination of CBT and contingency management, or control. In addition, all participants were receiving methadone maintenance treatment. Contingency management was particularly effective for the treatment of drug misuse (RR 4.40; 1.20 to 16.17) in the antisocial personality disorder population. These results were largely consistent with those found in a systematic review on psychosocial interventions for drug misuse (see NCCMH, 2007a).

Brooner and colleagues (1998) compared contingency management with control in opioid dependent people with antisocial personality disorder. Contingency management included contingent increases in methadone dose, scheduling of methadone, therapy sessions that were more convenient for the participant, and so on. There appeared to be a reduction in drug use for the treatment group compared with control but this was not statistically significant. This study had a number of limitations: first, urinalysis data were reported in a manner that could not rule out double counting of individuals therefore it was difficult to interpret the results; second, this study used a very different method of reinforcement (vouchers which could be exchanged for goods and services) in comparison with Messina and colleagues (2003), which may have contributed to the lack of positive effect.

Woody (1983) compared supportive-expressive psychotherapy with cognitive behavioural psychotherapy for the treatment of opioid dependence. Woody reported that participants with antisocial personality disorder had worse outcomes, whereas participants with depression and no antisocial personality disorder generally showed better outcomes. Participants with antisocial personality disorder and depression generally fell in between the two groups on a broad range of drug misuse outcomes. McKay and colleagues (2000) compared group therapy with individualised relapse prevention for cocaine dependence and found no significant differences between cocaine users with and without antisocial personality disorder for any substance misuse outcome (including cocaine and alcohol).

Wölwer and colleagues (2001) compared CBT with coping skills training and treatment as usual for alcohol dependence and found no significant differences between subgroups of patients with or without antisocial personality disorder, as measured by abstinence at 3 or 6 months after detoxification. In contrast, Hesselbrock (1991) in a study of inpatient alcoholism treatment reported worse outcomes (as measured by mean daily alcohol consumption and alcohol-related problems at 1 year) for participants with antisocial personality disorder.

7.3.5. Clinical evidence summary for psychological interventions for the treatment of comorbid substance misuse

The evidence on psychological interventions for drug misuse indicates that people with antisocial personality disorder can benefit from treatment. There was a particularly large effect found when using contingency management to treat drug misuse in people with antisocial personality disorder. Although there was some inconsistency, in that another trial did not show such positive effects, this appears to be partly explained by the method of contingency management used in the latter trial and is consistent with a review of the drug misuse literature that suggests that contingency management has the strongest evidence for effectiveness (see NCCMH, 2007a, 2007b). While the other studies reviewed above do not report such positive effects, the picture of generally poor outcomes for people with antisocial personality disorder, which is commonly assumed to be the case, was not confirmed. People with antisocial personality disorder may be able to benefit as much from these interventions as people without antisocial personality disorder.

7.3.6. Health economic evidence on the treatment of comorbid substance misuse and alcohol dependence

No evidence on the cost effectiveness of treatments of comorbid substance misuse and alcohol dependence was identified by the systematic search of the literature. Details on the systematic search of the economic literature are provided in Chapter 3.

7.3.7. From evidence to recommendations

The limited evidence reviewed above suggests that people with antisocial personality disorder can benefit from treatments for drug and alcohol misuse and that this benefit could be of the same order as those without a personality disorder. The encouraging results for contingency management are in line with the expectation that people with antisocial personality disorder may respond well to positive reinforcement. It was also the judgement of the GDG that such findings could generalise to people who meet criteria for DSPD.

7.3.8. Recommendations

7.3.8.1.

For people with antisocial personality disorder who misuse drugs, in particular opioids or stimulants, offer psychological interventions (in particular, contingency management programmes) in line with recommendations in the relevant NICE clinical guideline [NICE, 2007].

7.3.8.2.

For people with antisocial personality disorder who misuse or are dependent on alcohol, offer psychological and pharmacological interventions in line with existing national guidance for the treatment and management of alcohol disorders.

7.3.8.3.

For people who meet criteria for psychopathy or DSPD, offer treatment for any comorbid disorders in line with existing NICE guidance. This should happen regardless of whether the person is receiving treatment for psychopathy or DSPD because effective treatment of comorbid disorders may reduce the risk associated with psychopathy or DSPD.

7.3.9. The psychological treatment of comorbid depression and anxiety disorders

There is considerable evidence that a personality disorder may have a negative impact on the course of a common mental disorder (for example, Massion et al., 2002) and that a common mental disorder may be associated with a poorer outcome in personality disorder (for example, Yang & Coid, 2007). It is also the case that adults with antisocial personality disorder often have multiple comorbidities. For example, those with comorbid anxiety and antisocial personality disorder also had significantly higher levels of comorbid depression, alcohol dependence and substance dependence and higher rates of suicide attempts compared with adults with antisocial personality disorder or anxiety disorders alone (Goodwin & Hamilton, 2003). This suggests that effective treatment for common mental disorders in antisocial personality disorder may be challenging but potentially important.

A systematic search identified no high-quality trials focused on the treatment of depression or anxiety disorders comorbid with antisocial personality disorder. Therefore the GDG and review team searched for high-quality systematic reviews that addressed the question of the treatment of comorbid depression and anxiety disorders. The GDG took the view that as the initial search for systematic reviews had failed to identify a significant numbers of reviews focused solely on the issue of comorbidity with antisocial personality disorder that they should consider (1) reviews of a broad range of personality disorders and their impact on the treatment of depression and anxiety and, (2) reviews of personality variables (such as trait anxiety, impulsivity and aggression) that might have an impact on treatment outcomes. The GDG also agreed to review the existing NICE guidelines for common mental disorders to determine if any recommendations had been made about comorbid common mental health problems and antisocial personality disorder or indeed any other personality disorder.

A number of systematic reviews were identified and quality assessed. The following reviews were considered (Dreessen & Arntz, 1998; Mulder et al., 2003; Newton-Howes et al., 2006). In addition, the following NICE guidelines were also reviewed (NCCMH, 2005a, 2005b, 2006, 2009a).

From these reviews a number of common themes emerged. First, there is no consistent evidence that demonstrates people with antisocial personality disorder do not benefit from evidence-based psychological interventions for common mental health problems or that they may be harmed by such interventions (see for example the reviews by Mulder and colleagues [2003] on personality disorder and depression). (It should be noted there is some evidence to suggest that brief interventions may have little benefit for borderline personality disorder; NCCMH, 2009b.) Second, there is evidence from post hoc analyses of individual trials that the presence of a personality disorder, or developmental or social factors that are commonly associated with a personality disorder, may lead to a diminution of effectiveness. This was commonly addressed in the treatment trials by extending the duration of treatment (for example, Fournier et al., 2008). There was also some evidence that more experienced therapists were better able to deal with Axis II comorbidity (Hollon, personal communication, 2008). Nemeroff and colleagues (2003), in a post hoc analysis of the Keller and colleagues’ (2000) trial of nefazodone and a cognitive behavioural-analysis system of psychotherapy for chronic depression, found that patients with a significant history of childhood trauma obtained better outcomes with psychological treatment, while those with no history of abuse obtained better outcomes with pharmacological treatments.

7.3.10. Clinical evidence summary for the psychological treatment of comorbid depression and auxiety disorders

People with antisocial personality disorder have high levels of comorbid common mental health problems, which are associated with poorer long-term outcomes. Evidence from clinical trials relating directly to this issue is lacking, but post hoc analysis of data drawn from individual trials and from systematic reviews across a range of personality disorders suggest that effective treatment of common mental health disorders is possible, but may require the extension of the duration of the treatment and/or high levels of clinical skill and experience.

7.3.11. Health economic evidence on the treatment of comorbid depression and anxiety disorders

No evidence on the cost effectiveness of treatments of comorbid depression and anxiety disorders was identified by the systematic search of the literature. Details on the methods adopted in the systematic search of the economic literature are provided in Chapter 3.

7.3.12. From evidence to recommendations

The evidence reviewed suggested that the treatment of common mental disorders in antisocial personality disorder is possible, but that caution is required in developing any recommendations because the evidence base is drawn from trials involving a wider range of personality disorders than just antisocial personality disorder. There is a clear indication in the evidence reviewed that consideration should be given to extending the duration of treatment. In addition, staff should be mindful of the need to take steps to address the increased likelihood that people with antisocial personality disorder will drop out of treatment.

7.3.13. Recommendations

7.3.13.1.

People with antisocial personality disorder should be offered treatment for any comorbid disorders in line with recommendations in the relevant NICE clinical guideline, where available. This should happen regardless of whether the person is receiving treatment for antisocial personality disorder.

7.3.13.2.

When providing psychological interventions for comorbid disorders to people with antisocial personality disorder, consider lengthening their duration or increasing their intensity.

7.4. THERAPEUTIC COMMUNITY INTERVENTIONS FOR PEOPLE WITH ANTISOCIAL PERSONALITY DISORDER AND ASSOCIATED SYMPTOMS AND BEHAVIOURS

7.4.1. Introduction

In the history of psychological treatments for personality disorder the therapeutic community has played an important role (Rapoport, 1960). The therapeutic community movement had a significant impact on mental healthcare in the mid to late 20th century (Lees et al., 2003) with developments in the prison service (Snell, 1962), drug services and for other personality disorders (Lees et al., 2003). However, in healthcare there has been a recent move away from therapeutic communities, in part influenced by high costs in the absence of convincing evidence for efficacy (Lees et al., 2003).

Therapeutic communities differ from other treatment approaches in the use of the residential ‘community’ as the key agent for change. Peer influence is used to help individuals acquire social skills and learn social norms, and so take on an increased level of personal and social responsibility within the unit (Smith et al., 2006). In addition to therapeutic communities based on social learning theory, there are rehabilitation centres that emphasise more behavioural, hierarchical principles that positively and negatively reinforce a range of behaviours. Residential therapeutic communities involve therapeutic group work, one-to-one keyworking, the development of practical skills and interests, education and training. The intensive nature of their approach means that such programmes tend to be longer in duration (6 to 12 months) (Greenwood et al., 2001). In the UK, the Community of Communities project (Keenan & Paget, 2006) has developed standards of good practice for therapeutic communities.

Current practice

Therapeutic communities are found within health, education and social care and prison settings in the UK and often work with people with symptoms and behaviours associated with the antisocial personality disorder construct.

There are a number of therapeutic communities specialising in the treatment of substance misuse, with over half of residential services in the National Treatment Agency for Substance Misuse online directory12 describing themselves as therapeutic communities. In addition, of the 56 therapeutic communities surveyed by the Community of Communities, 15 were in prison settings (Royal College of Psychiatrists, 2008b).

7.4.2. Definition and aim of review

The review assessed therapeutic communities for people with antisocial personality disorder, people with symptoms and behaviours associated with this diagnostic construct, and people with comorbid substance misuse.

7.4.3. Databases searched and inclusion/exclusion criteria

Information about the databases searched and the inclusion/exclusion criteria used for this section of the guideline can be found in Table 43.

Table 43. Databases searched and inclusion/exclusion criteria for clinical evidence.

Table 43

Databases searched and inclusion/exclusion criteria for clinical evidence.

7.4.4. Studies considered13

The review team conducted a new systematic search for RCTs that assessed the efficacy of therapeutic communities for people with antisocial personality disorder or symptoms and behaviours associated with antisocial personality disorder. A systematic search for non-RCTs that assessed the efficacy of therapeutic communities for offenders was also conducted.

There were no trials of therapeutic communities for people with antisocial personality disorder that met the eligibility criteria of the GDG. However, three trials that assessed therapeutic communities for offenders who misused drugs (NIELSEN1996; SACKS2004; WEXLER1999) met the eligibility criteria set by the GDG, providing data on 1,682 participants. All were published in peer-reviewed journals.

As there was only one RCT for therapeutic communities for offenders without substance misuse problems (Lamb & Goentzel, 1974), the review team conducted a systematic search for non-RCTs that assessed the efficacy of therapeutic communities in this population; two non-RCTs (Marshall, 1997; Robertson & Gunn, 1987) were identified.

In addition, seven studies were excluded from the analysis. The most common reason for exclusion was the lack of relevant outcomes (further information about both included and excluded studies can be found in Appendix 15).

7.4.5. Clinical evidence on therapeutic communities for offenders with substance misuse problems

Three RCTs have been conducted in institutional settings evaluating the evidence for therapeutic communities in substance misuse offenders. In two trials the intervention included treatment within prison followed by release to a residential community of 6 months’ duration (SACKS2004; WEXLER1999). The third trial (NIELSEN1996) assessed a work-release therapeutic community programme.

Summary study information and evidence from the included trials are shown in Table 44 and Table 45. Full evidence profiles and forest plots can be found in Appendices 15 and 16.

Table 44. Study information table for trials of therapeutic communities for offenders with substance misuse problems.

Table 44

Study information table for trials of therapeutic communities for offenders with substance misuse problems.

Table 45. GRADE evidence summary for therapeutic communities for offenders with substance misuse problems.

Table 45

GRADE evidence summary for therapeutic communities for offenders with substance misuse problems.

Therapeutic community prison and aftercare programmes for offenders who misused drugs (many of whom had antisocial personality disorder) were associated with relatively large reductions in offending (RR = 0.62; 0.49 to 0.78). At 5-year follow-up the difference was still statistically significant (RR = 0.93; 0.87 to 0.99).

7.4.6. Health economic evidence on therapeutic communities for offenders with substance misuse problems

Four US-based studies that reported on the cost-effectiveness of therapeutic communities for offenders with substance abuse problems were identified by the systematic search of economic literature (McCollister et al., 2003, 2003a and 2004; Griffith et al., 1999). One study by McCollister and colleagues (2003 McCollister and colleagues (2004) evaluated the short and long-term cost-effectiveness (12 months and 5 years respectively) of a Californian in-prison therapeutic community and aftercare programme for male prisoners with history of substance misuse. Cost data included in-prison and aftercare treatments, hospital inpatient and outpatient episodes, methadone treatments and other self-help programmes. The measure of effectiveness was the number of incarceration days avoided during follow-up. In the comparison with no treatment, the ICER was $80 per avoided incarceration day after 12 months, which came down to $65 after 5 years. A similar study based in Delaware evaluated a work-release therapeutic community and aftercare programme for male prisoners with history of substance misuse (McCollister et al., 2003a). Only costs of the in-prison and after-care treatment programmes were included and follow-up was 18 months. The ICER versus no treatment was $65 per incarceration day avoided.

Another study evaluated the cost effectiveness of in-prison therapeutic community treatment for male offenders with a history of substance misuse in Texas (Griffith et al., 1999). A retrospective analysis was undertaken over 3 years, comparing treated offenders’ parole and aftercare costs, as well as re-incarceration rates, with respective costs and outcomes in an untreated comparison group. Separate analyses were conducted for low-risk and high-risk offenders. For the low-risk group analysis, the ICER was $494 per 1% reduction in re-incarceration; for the high-risk group analysis, the ICER fell to $165. This was largely explained by the higher re-incarceration rates experienced by the high-risk untreated group compared with the treated group, while for the low-risk group analysis, re-incarceration rates were similar in the two cohorts (treated and untreated).

Details on the study characteristics and results are provided in the form of evidence tables in Appendix 14. The methods adopted for the systematic review of economic literature are discussed in Chapter 3.

7.4.7. Clinical evidence on therapeutic communities for adult offenders

There were three trials that investigated the efficacy of therapeutic communities for general offenders in institutional and community settings. Of these, one was an RCT (Lamb & Goentzel, 1974) and two were non-RCTs (Marshall, 1997; Robertson & Gunn, 1987). The RCT investigated a community alternative to prison in the US and the two non-RCTs investigated the effects of therapeutic communities for prisoners treated in HMP Grendon in the UK. For general offenders a meta-analysis was not conducted as these studies differed in study design; instead these studies were narratively reviewed.

Lamb & Goentzel (1974) randomised participants to regular prison services or to a therapeutic community as an alternative to prison in a community setting. The therapeutic community comprised three phrases. In phases one and two, the participants were given more responsibility and privileges within each phase. Phase three continued while the participant was on probation. The participant returned to the therapeutic community to visit their assigned probation officer and to participate in social activities. The study found the therapeutic community to have a harmful effect on re-offending at 1-year follow-up for 31 participants in the treatment group in comparison with 31 participants in the control group (RR 1.22; 0.59, 2.53).

Robertson & Gunn (1987) conducted a 10-year prospective cohort study of participants released from HMP Grendon in comparison with a matched control; there were some differences between the two groups, such as the treated group having more desire for psychiatric help compared with the control group. The study found no significant differences between participants treated in a therapeutic community compared with regular prison services (93% and 85% respectively, x2 = 1.37, d.f. 1, NS).

Marshall (1997) conducted a retrospective cohort study of participants who stayed in HMP Grendon (N = 702) from 1984 to 1989. These participants were compared with participants who were selected for Grendon in the same period but who did not actually go there (N = 142). The retrospective study found no effect on the therapeutic community for participants who attended Grendon versus the comparison group who did not (RR 0.92; 0.82 −1.03).

7.4.8. Health economic evidence on therapeutic communities for adult offenders

No economic evidence on therapeutic communities for adult offenders was identified in the literature.

7.4.9. Clinical evidence summary on therapeutic communities

The majority of RCT evidence available on therapeutic communities was on people who misuse drugs in the criminal justice system. These samples had a fair proportion of people diagnosed with antisocial personality disorder (between 39 and 51%) in addition to all participants reporting behaviour or symptoms associated with the antisocial personality disorder diagnostic construct. There was found to be a relatively large reduction in offending. The economic evidence suggests that in-prison therapeutic communities for offenders with history of substance abuse may be cost effective in terms of reducing future re-offending.

In contrast the evidence for therapeutic communities for general offenders is limited and based on weaker study design. There is no evidence to suggest that therapeutic communities are effective for general offenders.

7.4.10. From evidence to recommendations

The GDG concluded that therapeutic communities appeared to be effective for people in prison or on probation who misuse drugs, many of whom were diagnosed with antisocial personality disorder. Therefore their judgement was that therapeutic communities targeted specifically at drug misuse are likely to be effective in people with antisocial personality disorder who misuse drugs. However, the GDG concluded there was insufficient evidence to apply these findings to therapeutic communities targeting general offenders.

7.4.11. Recommendations

7.4.11.1.

For people with antisocial personality disorder who are in institutional care and who misuse or are dependent on drugs or alcohol, consider referral to a specialist therapeutic community focused on the treatment of drug and alcohol problems.

7.5. PHARMACOLOGICAL INTERVENTIONS FOR ANTISOCIAL PERSONALITY DISORDER

7.5.1. Introduction

A rationale for pharmacological approaches in antisocial personality disorder is that many of the behavioural traits of personality disorder may have a biological basis and are associated with neurochemical abnormalities of the central nervous system (Coccaro et al., 1996a, 1996b; Hollander et al., 1994). However, a major problem in studying the effects of medication is that it is difficult to map drug action on the personality disorders as they are listed in DSM. The reason for this is that they are so heterogeneous that it may be more fruitful to focus on behavioural clusters (Markovitz, 2001). Soloff (1998) has been influential by introducing a symptom-orientated approach. Ignoring the specific DSM Axis II disorders, he grouped personality psychopathology into the following symptom domains: cognitive-perceptual, affective, impulse-behavioural and anxious-fearful. Affective symptoms in turn were subdivided into dysregulation of (a) mood and (b) anxiety. He suggested that since these domains were mediated by the same neurotransmitter systems as Axis I disorders, albeit in an attenuated form, this approach could lead to more rational prescribing.

Applying this approach, Soloff found evidence that conventional antipsychotic drugs in low doses were effective in reducing the cognitive perceptual abnormalities (Soloff et al., 1986a; Goldberg et al., 1986). For dysregulation of mood, there was some evidence for the use of SSRIs (Cornelius et al., 1990; Markovitz et al., 1991), tricyclic antidepressants (Soloff et al., 1986b), venlafaxine (Markovitz & Wagner, 1995) and monoamine oxidase inhibitors (MAOIs) (Parsons et al., 1989). For impulsive behavioural dyscontrol, most attention had been focused on the SSRIs (Cornelius et al., 1990; Kavouissi et al., 1994), but lithium (Tupin et al., 1973; Links, 1990) and anticonvulsants such as carbamazepine (Cowdry & Gardner, 1989), valproate (Stein et al., 1995) and divalproex sodium (Wilcox, 1995) had also shown some positive outcomes.

Various features of antisocial personality disorder might be targets for a pharmacological intervention. Paranoia, for instance, emerges from factor analysis and hence might be a target of low-dose antipsychotic medication. Similarly, impulsive dyscontrol and aggressive behaviour are important features of antisocial personality disorder and might usefully be targeted with SSRIs or mood stabilisers. This section therefore reviews the evidence on the use of drugs for those with antisocial personality disorder.

As with assessing the effectiveness of psychological interventions, there are three difficulties that need to be considered. First, antisocial personality disorder is often comorbid with Axis I conditions and, as it may often be the presence of the latter that causes the individual to present for treatment, it is not always clear whether it is the Axis I or Axis II condition that is being targeted when medication is used. Second, use of alcohol and other illicit substances, which is common in people with antisocial personality disorder, may diminish response rates to pharmacotherapy (Markovitz, 2001). Third, with complex conditions such as antisocial personality disorder, it is likely that multiple neurotransmitter systems are at play in producing, for example, the affective dysregulation (Soloff, 1998). This again makes drug selection difficult.

Current practice

The state of current practice in relation to the use of pharmacological interventions to treat antisocial personality disorder is unclear, but it is likely that pharmacological interventions are used in this population to treat symptoms rather than as an intervention for the disorder itself. The reported level of prescription in the prison population does not suggest that pharmacological interventions are used at a generally high level in offender populations (Christina Rowlands, presentation to the GDG, December 2007).

7.5.2. Databases searched and inclusion/exclusion criteria

Information about the databases searched and the inclusion/exclusion criteria used for this section of the guideline can be found in Table 46.

Table 46. Databases searched and inclusion/exclusion criteria for clinical evidence.

Table 46

Databases searched and inclusion/exclusion criteria for clinical evidence.

7.5.3. Studies considered14

Ten trials relating to clinical evidence met the eligibility criteria set by the GDG, providing data on 749 participants (BARRATT1997, COCCARO1997A, GOTTSCHALK1973, HOLLANDER2003, LEAL1994, MATTES2005, MATTES 2008, NICKEL2005B, POWELL1995, SHEARD1976, STANFORD2005). Of these, all were published in peer-reviewed journals between 1973 and 2008. In addition, 16 studies were excluded from the analysis. The most common reasons for exclusion were non-random allocation of participants to treatment and control and populations that would not meet the GDG’s inclusion criteria, for example, participants with schizophrenia (further information about both included and excluded studies can be found in Appendix 15).

There was one trial providing evidence for pharmacological interventions for antisocial personality disorder (BARRATT 1997). The purpose of the study was to look at the effects of anticonvulsants on aggression among offenders in prison, however all participants at baseline met DSM-III-R criteria for antisocial personality disorder.

Two trials were found that investigated pharmacological interventions for a sub-population of antisocial personality disorder with comorbid substance misuse. (LEAL1994, POWELL1995) One trial compared amantadine and desipramine with placebo for participants with cocaine dependence (LEAL1994) and one trial compared nortriptyline and bromocroptine with placebo for participants with alcohol dependence (POWELL1995).

For the review of pharmacological evidence for antisocial personality disorder and associated symptoms or behaviour, eight trials were included (COCCARO1997A, GOTTSCHALK1973, HOLLANDER2003, MATTES2005, MATTES2008, NICKEL2005, SHEARD1976, STANFORD2005). Six trials compared anticonvulsants with placebo (GOTTSCHALK1973, HOLLANDER2003, MATTES2005, MATTES2008, NICKEL2005, STANFORD2005), one compared antidepressants with placebo (COCCARO1997A) and one compared lithium with placebo (SHEARD1976). The population in all the trials had an elevated level of impulsive aggression and/or anger while two trials looked specifically at offenders (SHEARD1976, GOTTSHALK1993). The age range for the trials were 19 to 67 years.

7.5.4. Clinical evidence for antisocial personality disorder

There was one trial (see Table 47 and Table 48) that looked at the effects of anticonvulsants on aggression among prison inmates who all met DSM-III-R criteria for antisocial personality disorder (BARRATT1997). Using the modification of the Overt Aggression Scale (OAS), the study found the anticonvulsant phenytoin to have a small but non-significant effect on aggression compared with placebo (SMD −0.26; −0.61, 0.09). Full study characteristics and forest plots can be found in Appendices 15 and 16 respectively.

Table 47. Study information on pharmacological interventions for antisocial personality disorder.

Table 47

Study information on pharmacological interventions for antisocial personality disorder.

Table 48. GRADE evidence summary table on pharmacological interventions for antisocial personality disorder.

Table 48

GRADE evidence summary table on pharmacological interventions for antisocial personality disorder.

7.5.5. Clinical evidence for antisocial personality disorder and comorbid substance misuse

Two trials (LEAL1994, POWELL1995) studied the effects of antidepressants in people with antisocial personality disorder and comorbid substance misuse compared with placebo (see Table 49). It was not possible to meta-analyse the data so there is no GRADE evidence summary table. Summary study information and evidence from the included trials are shown in Table 43 and Table 44. Full study characteristics can be found in Appendix 15.

Table 49. Study information for pharmacological interventions for antisocial personality disorder with comorbid substance misuse.

Table 49

Study information for pharmacological interventions for antisocial personality disorder with comorbid substance misuse.

For antidepressants versus placebo there was a small effect for leaving the study early (RR 0.90; 0.52, 1.55) for participants with cocaine dependence (LEAL1994) and alcohol dependence (POWELL1995), and a moderate effect on abstinence (RR 0.72; 0.53–0.97) for participants with alcohol dependence. However, the effect on abstinence was small and based on only one study (POWELL1995).

The two trials also looked at the effects of dopaminergic drugs versus placebo (LEAL1994, POWELL1995). No significant differences were found between drop out for both treatment and placebo groups (RR 1.18; 0.72, 1.94) and a small but non-significant difference was found in abstinence for participants with alcohol dependence (RR 0.91; 0.75, 1.10). This effect was small and based on sparse data.

7.5.6. Clinical evidence for antisocial personality disorder and associated symptoms or behaviour

Table 50 summarises the study information for trials concerned with pharmacological interventions for aggression in people with antisocial personality disorder. The GRADE evidence summaries can be found in Table 51.

Table 50. Study information for the trials of pharmacological interventions for aggression.

Table 50

Study information for the trials of pharmacological interventions for aggression.

Table 51. GRADE evidence summary for pharmacological interventions for aggression.

Table 51

GRADE evidence summary for pharmacological interventions for aggression.

Full study characteristics and forest plots can be found in Appendices 15 and 16 respectively.

Anticonvulsants versus placebo

Six trials investigated the effects of a number of anticonvulsants on impulsive aggression and found a small and non-significant effect on aggression at end of treatment (SMD −0.13; −0.35 to 0.09). The quality of evidence was very low with high heterogeneity (I2 = 74.4%).

SSRIs versus placebo

One trial compared the SSRI fluoxetine with placebo for reducing aggression in a population with elevated aggression and found the effects of treatment to be medium to large (SMD −0.73; −1.41 to −0.04). However, this is based on one study of low quality.

Lithium versus placebo

There was only one trial that investigated lithium versus placebo in a population with elevated levels of the antisocial personality disorder construct aggression that met the eligibility criteria. The trial showed a medium effect for treatment, which was, however, non-significant and of low quality (SMD −0.60; −1.23, 0.03).

7.5.7. Clinical evidence summary for antisocial personality disorder and associated symptoms or behaviour

There was no consistent evidence, including from uncontrolled studies, that supported the use of any pharmacological intervention to treat antisocial personality disorder or to treat the behaviour and symptoms that underline the specific diagnostic criteria for antisocial personality disorder.

7.5.8. Health economic evidence on pharmacological interventions for antisocial personality disorder

No evidence on the cost effectiveness of pharmacological interventions for antisocial personality disorder with or without comorbid substance misuse and associated symptoms of behaviour was identified by the systematic search of the literature. Details on the methods adopted for the systematic review of economic literature are provided in Chapter 3.

7.5.9. From evidence to recommendations

The evidence did not support the generation of recommendations for the routine use of pharmacological interventions for the treatment of people with antisocial personality disorder.

7.5.10. Recommendations for pharmacological interventions

7.5.10.1.

Pharmacological interventions should not be routinely used for the treatment of antisocial personality disorder or associated behaviours of aggression, anger and impulsivity.

7.5.10.2.

Pharmacological interventions for comorbid mental disorders, in particular depression and anxiety, should be in line with recommendations in the relevant NICE clinical guideline. When starting and reviewing medication for comorbid mental disorders, pay particular attention to issues of adherence and the risks of misuse or overdose.

7.5.11. Recommendations on general issues in the treatment of adults with antisocial personality disorder

7.5.11.1.

When providing psychological or pharmacological interventions for antisocial personality disorder, offending behaviour or comorbid disorders to people with antisocial personality disorder, be aware of the potential for and possible impact of:

  • poor concordance
  • high attrition
  • misuse of prescribed medication
  • drug interactions (including with alcohol and illicit drugs).

7.6. RESEARCH RECOMMENDATIONS

Through identifying research limitations from the evidence based reviews, the guideline development group has formulated the following research recommendations.

7.6.1.1. Severity as a potential moderator of effect in group-based cognitive and behavioural interventions

Does the pre-treatment level of the severity of disorder/problem have an impact on the outcome of group-based cognitive and behavioural interventions for offending behaviour? A meta-analysis of individual participant data should be conducted to determine whether the level of severity assessed at the beginning of the intervention moderates the effect of the intervention. The study (for which there are large data sets that include over 10,000 participants) could inform the design of a large-scale RCT (including potential modifications of cognitive and behavioural interventions) to test the impact of severity on the outcome of cognitive and behavioural interventions.

Why this is important

Research has established the efficacy of cognitive and behavioural interventions in reducing re-offending. However, the effects of these interventions in a range of offending populations are modest. The impact of severity on the outcome of these interventions has not been systematically investigated, and post hoc analyses and meta-regression of risk as a moderating factor have been inconclusive. Expert opinion suggests that severe or high-risk individuals may not benefit from cognitive and behavioural interventions, but if they were to be of benefit then the cost savings could be considerable.

7.6.1.2. Group-based cognitive and behavioural interventions for populations outside criminal justice settings

Are group-based cognitive and behavioural interventions effective in reducing the behaviours associated with antisocial personality disorder (such as impulsivity, rule-breaking, deceitfulness, irritability, aggressiveness and disregard for the safety of self or others)? This should be tested in an RCT that examines medium-term outcomes (including cost effectiveness) over a period of at least 18 months. It should pay particular attention to the modification and development of the interventions to ensure the focus is not just on offending behaviour, but on all aspects of the challenging behaviours associated with antisocial personality disorder.

Why this is important

Not all people with antisocial personality disorder are offenders but they exhibit a wide range of antisocial behaviours. However, the evidence for the treatment of these behaviours outside the criminal justice system is extremely limited. Following publication of the Department of Health’s policy guidance, Personality Disorder: No Longer a Diagnosis of Exclusion (2003), it is likely that there will be an increased requirement in the NHS to offer treatments for antisocial personality disorder.

7.6.1.3. Treatment of comorbid anxiety disorders in antisocial personality disorder

Does the effective treatment of anxiety disorders in antisocial personality disorder improve the long-term outcome for antisocial personality disorder? An RCT of people with antisocial personality disorder and comorbid anxiety disorders that compares a sequenced treatment programme for the anxiety disorder with usual care should be conducted. It should examine, over a period of at least 18 months, the medium-term outcomes for key symptoms and behaviours associated with antisocial personality disorder (including offending behaviour, deceitfulness, irritability and aggressiveness, and disregard for the safety of self or others), as well as drug and alcohol misuse, and anxiety. The study should also be designed to explore the moderators and mediators of treatment effect which could help determine the role of anxiety in the course of antisocial personality disorder.

Why this is important

Comorbidity with Axis I disorders is common in antisocial personality disorder, and chronic anxiety has been identified as a particular disorder that may exacerbate the problems associated with antisocial personality disorder. There are effective treatments (psychological and pharmacological) for anxiety disorders but they are often not offered to people with antisocial personality disorder. Current treatment guidelines set out clear pathways for the stepped or sequenced care of people with anxiety disorders. An RCT to test the benefit of this approach in the treatment of anxiety would potentially lead to a significant reduction in illness burden but a reduction in antisocial behaviour would have wider societal benefits. The study should provide important information on the challenges of delivering these interventions for a population that has typically both rejected and been refused treatment.

7.6.1.4. Using selective serotonin reuptake inhibitors to increase cooperative behaviour in people with antisocial personality disorder in a prison setting

Although there is evidence that selective serotonin reuptake inhibitors (SSRIs), such as paroxetine, increase cooperative behaviour in normal people and do so independently of the level of sub-syndromal depression, this has yet to be tested in other settings. Given that people with antisocial personality disorder are likely to have difficulties cooperating with one another (because of a host of personality traits that include persistent rule-breaking for personal advantage, suspiciousness, grandiosity, and so on), an RCT should be conducted to find out whether these reported changes of behaviour with an SSRI in normal people generalises to clinical populations in different settings.

Why this is important

There is little evidence in the literature on the pharmacotherapy of antisocial personality disorder to justify the use of any particular medication. However, multiple drugs in various combinations are used in this group either to control aberrant behaviour or in the hope that something might work. Current interventions lack a clear rationale. This recommendation has the potential to advance the field in that (a) it is linked to a clear hypothesis (that cooperative behaviour is linked to a dysregulation of the serotonin receptors – for which there is substantial evidence) and (b) that it is feasible to obtain an answer to this question, given that there are a large number of individuals detained in prison settings who would meet antisocial personality disorder criteria. Constructing an experimental task that requires cooperative activity would not be difficult in such a setting, since all of those who might be willing to participate are already detained. The successful execution of this research would be important in that it (a) would establish the feasibility of conducting such a trial in a prison setting with this group, and (b) provide a clear and sensible outcome measure of antisocial behaviour that might be generalised to other settings.

7.6.1.5. A therapeutic community approach for antisocial personality disorder in a prison setting

Is a therapeutic community approach in a prison setting more clinically and cost effective in the treatment and management of antisocial personality disorder than routine prison care? There should be a large-scale RCT comparing the clinical and cost effectiveness of the therapeutic community approach for adults with antisocial personality disorder with routine care. It should examine the medium-term outcomes (for example, offending behaviour, mental state and vocational outcomes) over a period of at least 18 months following release from prison. The study should also be designed to explore the moderators and mediators of treatment effect, which could help to determine the factors associated with benefits or harms of the therapeutic community approach.

Why this is important

There is evidence from RCTs that the therapeutic community approach is of value with drug and alcohol misusers in a prison setting at reducing the incidence of offending behaviour on release. However, there are no equivalent studies of a programme in the prison system on antisocial personality disorder populations that do not have significant drug or alcohol problems. Data that do exist are from non-UK settings. Answering this question is of importance because outcomes for adults with antisocial personality disorder are poor and there are already considerable resources devoted to a therapeutic community approach in the UK prison system (for example, HMP Grendon Underwood). The study could inform policy and resources decisions about the management of antisocial personality disorder in the criminal justice system.

Footnotes

11

Here and elsewhere in the guideline, each study considered for review is referred to by a study ID in capital letters (primary author and date of study publication, except where a study is in press or only submitted for publication, then a date is not used). The references for studies can be found in Appendix 15.

12
13

Here and elsewhere in the guideline, each study considered for review is referred to by a study ID in capital letters (primary author and date of study publication, except where a study is in press or only submitted for publication, then a date is not used). The references for studies can be found in Appendix 15.

14

Here and elsewhere in the guideline, each study considered for review is referred to by a study ID in captial letters (primary author and date of study population, except where a study is in press or only submitted for publication, then a date is not used). The references for studies can be found in Appendix 15.

Copyright © 2010, The British Psychological Society & The Royal College of Psychiatrists.

All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Enquiries in this regard should be directed to the British Psychological Society.

Cover of Antisocial Personality Disorder
Antisocial Personality Disorder: Treatment, Management and Prevention.
NICE Clinical Guidelines, No. 77.
National Collaborating Centre for Mental Health (UK).
Leicester (UK): British Psychological Society; 2010.

NICE (National Institute for Health and Care Excellence)

PubMed Health Blog...

read all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...