Diagnostic criteria

In DSM-5, antisocial personality disorder is classified under Cluster B personality disorders, together with borderline, histrionic and narcissistic personality disorders (American Psychiatric Association 2013). Central features include irresponsible and antisocial behaviour, impulsivity, aggressiveness and a tendency to disregard rights and boundaries of others. ICD-10 (World Health Organization 1992) defines a conceptually similar disorder to antisocial personality disorder in dissocial personality disorder, although there is more emphasis on the interpersonal deficits of the disorder than in the DSM-5 criteria and a history of conduct disorder is not necessary to make the diagnosis.

Relationship with psychopathy

Confusion exists regarding definitions of psychopathy and its relationship with antisocial personality disorder. The term psychopathy continues to be loosely applied within mental health, and is excluded from both the current ICD and DSM classifications of dissocial and antisocial personality disorder respectively. In the UK, until recently, a person could be detained in hospital against their will under the legal category of ‘psychopathic disorder’ within the Mental Health Act 1983, for ‘a persistent disorder or disability of mind (whether or not including significant impairment of intelligence) which results in abnormally aggressive or seriously irresponsible conduct’. It was generally accepted that the term psychopathic disorder did not represent a single clinical disorder, but was a legal category describing a number of severe personality disorders, which contributed to the person committing antisocial acts. However, the reform of the Mental Health Act in 2007 has replaced all previous separate definitions of disorder, including the term psychopathic disorder, with a single category of ‘mental disorder’ applied throughout the Act.

In recent years, use of the term psychopathy increasingly refers to antisocial and behavioural personality traits as measured by the Hare Psychopathy Checklist – Revised (PCL-R; Reference HareHare 2003). Hare empirically validated Cleckley’s seminal clinical description of a ‘psychopath’ (Reference CleckleyCleckley 1941), to produce a reliable and validated operationalised risk assessment instrument consisting of a 20-item questionnaire that is now widely used in the assessment and management of violent patients in forensic psychiatric services, and to a lesser extent in the management of violent offenders within the criminal justice system. High scores are empirically associated with a considerable increase in the likelihood of violent recidivism (Reference HareHare 1991). Hare psychopathy is not synonymous with a behavioural history of criminality or a categorical diagnosis of antisocial personality disorder, although it is often a correlate of both in severe cases.

Ethical considerations

The DSM criteria for antisocial personality disorder have been criticised for their focus on antisocial behaviour and criminality rather than on underlying personality structure and interpersonal deficits (Reference Widiger and CorbittWidiger 1993). Defining a syndrome largely by behaviours considered unacceptable by society poses ethical challenges for medical diagnosis. There has been considerable debate historically and currently about whether antisocial personality disorder, psychopathic disorder or psychopathy constitute medical conditions at all when the main symptoms are social rule-breaking and the ‘patient’ does not seem to suffer (Reference McGauley, Adshead, Sarkar, Felthouse and SassMcGauley 2007). Tensions between legal and medical approaches persist over whether such individuals should be punished for their behaviours rather than treated, and whether ‘treatment’ is really a disguise for moral re-education, particularly for people who are not voluntarily seeking help themselves.

Epidemiology

Antisocial personality disorder is a common condition, with the general population prevalence in men ranging from 1% (Reference Torgerson, Kringlen and CramerTorgerson 2001; Reference Coid, Yang and TyrerCoid 2006) to 6% (Reference Robins, Tipp, Przybeck, Robins and RegierRobins 1991). Studies in the UK report a prevalence of between 2% and 3% in the general population (Reference MoranMoran 1999; Reference Coid, Yang and TyrerCoid 2006). Prevalence rates are consistently higher in men compared with women (Reference Dolan and VollmDolan 2009) and the disorder becomes increasingly common in mental health services, the judicial system and prison settings, representing up to 80% of the prison population (Reference Singleton, Melzer and GatwardSingleton 1998; Reference MoranMoran 1999; Reference OgloffOgloff 2006).

Comorbidity

Antisocial personality disorder is associated with considerable and complex comorbidity with other mental disorders (Reference Swanson, Bland and NewmanSwanson 1994), particularly substance misuse (Reference Robins, Tipp, Przybeck, Robins and RegierRobins 1991; Reference Compton, Conway and StinsonCompton 2005). At least half of individuals with antisocial personality disorder have a co-occurring anxiety disorder (Reference Goodwin and HamiltonGoodwin 2003) and a quarter have a depressive disorder (Reference Lenzenweger, Lane and LorangerLenzenweger 2007). It is important to diagnose comorbid conditions, as the presence of antisocial personality disorder acts as a negative moderator of treatment response when these conditions are treated by conventional approaches.

Costs of antisocial personality disorder

Antisocial personality disorder is also associated with increased mortality, particularly at a young age, largely because of reckless behaviour (Reference Black, Baumgard and BellBlack 1996). As well as causing considerable harm to the individual, people with the disorder have a broad impact on family, relationships and social functioning, and make heavy demands on the judicial system, Social Services and mental healthcare. The costs of those individuals who engage in criminal behaviour may include direct emotional and physical injury to victims, damage to property, police time, involvement with the criminal justice system, increased use of healthcare facilities, lost employment opportunities, family disruption, relationship breakdown, child-care proceedings, gambling, and problems related to alcohol and substance misuse (Reference Myers, Stewart and BrownMyers 1998; Department of Health 1999; National Research Council 1999; National Institute for Health and Clinical Excellence 2009).

Yet, despite recognition of its high prevalence and level of psychiatric morbidity, there is evidence that people with antisocial personality disorder continue to be rejected from services because there remains a fundamental reluctance among psychiatrists to conceptualise the disorder as an illness requiring treatment. Reference Crawford, Sahib and BrattonCrawford et al (2009) found that 50% of a sample of patients with antisocial personality disorder presented to emergency medical services and 21% were admitted to a mental health in-patient unit. Despite this, only 21% of the overall sample were provided with follow-up care from mental health services. However, levels of alcohol and drug misuse were significantly lower among those that were given follow-up, suggesting that general psychiatric support can be useful, although this may have been a self-selected group. Perhaps not unreasonably, given the scarcity of evidence for effective interventions for the disorder as well as the difficulties in engaging individuals, mental health services find it difficult to see a role in the management and treatment of this condition.

Historical treatment studies

Early attempts to treat antisocial personality disorder and psychopathic disorder in the UK took place largely in high secure hospitals. As with the treatment of personality disorder more generally, psychoanalytic approaches to treatment were most prevalent and led to the development of therapeutic communities (Reference JonesJones 1952) in which the environmental setting itself became an essential therapeutic tool.

Current treatment approaches

More recently, in the absence of convincing evidence to justify the high costs of running such highly specialised residential treatments, the therapeutic community approach has waned in popularity and has been largely replaced by specific cognitive–behavioural therapy (CBT) programmes for offenders with personality disorder delivered within criminal justice settings. Much of the rationale for the use of these cognitive–behavioural approaches is based on the criminological literature on interventions that reduce offending behaviour. Programmes found to have the largest effect size adhere to the Risk–Need–Responsivity (RNR) model (Reference Andrews and McGuireAndrews 1995; Reference SimonSimon 1998), focusing on targeting patients at greatest risk of reoffending (risk), empirically established dynamic criminogenic risk factors such as criminal attitudes, substance misuse and impulsivity (need), and delivering interventions in a manner that maximises offender engagement in the treatment process (responsivity). However, it is not clear from the criminological literature how many participants in the studies considered would fulfil the diagnosis of antisocial personality disorder, and outcome criteria are usually restricted to the presence or absence of reoffending (National Institute for Health and Clinical Excellence 2009).

Evidence base for psychological therapies

Despite the proliferation of treatment programmes for offenders, only a few high-quality trials of specific psychological therapies have been conducted among people with antisocial personality disorder. A number of meta-analyses have highlighted the limited evidence base for effective treatments for this patient group (Reference Warren, McGauley and NortonWarren 2003; Reference Duggan, Adams and FerriterDuggan 2007; National Institute for Health and Clinical Excellence 2009; Reference Gibbon, Duggan and StoffersGibbon 2010) and reflect not only the difficulties in engaging patients in treatment studies, but also diagnostic inconsistencies and the failure of many outcome studies to distinguish between patients with and without psychopathy (Reference De Brito, Hodgins, McMurran and HowardDe Brito 2009). A comparative evaluation of available studies is further hampered by differences in defining and measuring outcome and a focus on treating incarcerated patients rather than those in the community.

These methodological limitations were highlighted in a Cochrane review of all prospective randomised controlled trials for individuals with antisocial personality disorder (Reference Gibbon, Duggan and StoffersGibbon 2010) (Box 1). However, because of the relatively robust clinical evidence indicating that group-based cognitive and behavioural interventions are moderately effective for offenders, the NICE guidelines concluded that because this population has such a high incidence of antisocial personality disorder, it would be reasonable to conclude that such interventions were likely to be effective for people with antisocial personality disorder, and therefore recommend their use (National Institute for Health and Clinical Excellence 2009).

Pharmacological treatment

Evidence for pharmacological treatment of antisocial personality disorder is equally limited. Two large meta-analyses (National Institute for Health and Clinical Excellence 2009; Reference Khalifa, Duggan and StoffersKhalifa 2010) concluded that there was no consistent evidence that supported the use of any pharmacological intervention to treat the disorder. Pharmacological interventions should not be routinely used for treating the primary traits of antisocial personality disorder or associated behaviours of aggression, anger and impulsivity, but should only be used for the treatment of comorbid mental disorders (National Institute for Health and Clinical Excellence 2009). Issues of treatment adherence and the risks of misuse or overdose are emphasised.

Biological and genetic factors

As with most psychiatric conditions, the aetiology of antisocial personality disorder is considered to have both a biological and a psychosocial foundation. In recent years there has been much interest in the genetic contribution to antisocial behaviour, particularly in children showing callous and unemotional traits, thought to be the precursors of psychopathy in adulthood. Aggressive and antisocial behaviour in such children shows a strong heritability compared with children with antisocial behaviour but an absence of callous and unemotional traits (Reference Viding, Blair and MoffittViding 2005), suggesting that such behaviour is ‘hard-wired’ into the brain from an early age. However, sophisticated adoption studies demonstrate significant gene–environment interactions. They show that children who are genetically vulnerable to behaving in an antisocial manner are likely to also experience harsh and inconsistent parenting, resulting in more pathology than would be expected from either genetic or adverse environmental influence acting alone or in combination (Reference Cadoret, Yates and TroughtonCadoret 1995).

Many studies have shown differences in physiological responses and brain structure in people who are aggressive and violent compared with those who are not, including different electro encephalogram activity, reduced levels of brain serotonin (Reference Coccaro, Kavoussi and ShelineCocarro 1996a, Reference Coccaro, Berman and Kavoussi1996b; Reference Dolan, Anderson and DeakinDolan 2001), and dysfunction in frontocortical and limbic brain areas (Reference Raine, Lencz and BihrleRaine 2000; Reference Blair, Peschardt and BudhaniBlair 2006). However, although it is important to understand the biology of aggression, the data are limited in that the studies have mostly been carried out on individuals with aggressive behaviour rather than those with antisocial personality disorder, and also fail to control for confounding factors such as comorbidity and substance misuse, as well as focusing on simple neurophysiological processes such as motor impulsivity or recognition of basic emotions rather than on more complex behaviour (National Institute for Health and Clinical Excellence 2009).

Antisocial personality disorder and psychopathy

The DSM-5 antisocial personality disorder category covers a heterogeneous population and does not distinguish between individuals with mild, moderate or severe psychopathy, as measured on the PCL-R. A substantial body of research has shown that, at most, only one out of three patients with antisocial personality disorder has severe psychopathy, and this latter group has a significantly poorer treatment prognosis than do patients with mild to moderately psychopathic antisocial personality disorder (Reference HareHare 1991, Reference Hare2003) (Box 2). There is evidence to show that these different subgroups have different aetiological and developmental pathways and show different responses to treatment.

Reference De Brito, Hodgins, McMurran and HowardDe Brito & Hodgins (2009) emphasise the importance of co-occuring anxiety in identifying subgroups within antisocial personality disorder. On the basis of studies of children and adults, De Brito & Hodgins propose that around half of the antisocial personality disorder population are characterised by anxiety as well as persistent antisocial behaviour, have low levels of callous–unemotional traits as children and low levels of psychopathic traits as adults. This group is more likely to have experienced physical abuse as children and resort to violence as a compensatory response to underlying emotional conflict and distress. The other half of the population have normal to low levels of anxiety and varying levels of psychopathy, but include a subgroup with high levels of psychopathy. This group shows high levels of callous–unemotional traits as children, low levels of anxiety, more predatory violence and are less amenable to treatment. Thus, the measurement of psychopathy is important to delineate subgroups of individuals with antisocial personality disorder who have lower psychopathy scores and higher levels of anxiety, and who may show a better response to treatment (Reference HodginsHodgins 2007, Reference Hodgins, De Brito and Chhabra2010; Reference Ulrich and CoidUlrich 2010).

Other environmental aetiological factors implicated in antisocial personality disorder include malnutrition, smoking in pregnancy, and harsh parenting practices including parental abuse (Reference Caspi, McClay and MoffittCaspi 2002). These development risk factors for adult antisocial personality disorder are also risk factors for early attachment difficulties, which may represent a common pathway for some forms of the disorder.

Attachment studies

Empirical evidence to support the hypothesis that antisocial personality disorder is a developmental disorder rooted in attachment comes from studies showing abnormal attachment patterns in forensic patients and prisoners diagnosed with the disorder. Several studies have shown higher levels of insecure attachment in violent offenders than in the normal population (Reference van IJzendoorn, Feldbrugge and Derksvan IJzendoorn 1997; Reference Frodi, Dernevik and SepaFrodi 2001; Reference Adshead, Pfafflin and AdsheadAdshead 2004; Reference Levinson and FonagyLevinson 2004; Reference Bogaerts, Vanheule and DeclerqBogaerts 2005). This evidence is supported by studies which show that offenders in prison or secure forensic institutions are more likely to report having experienced separations, abuse and neglect from their early caregivers than the general population (Reference CoidCoid 1992; Reference Heads, Taylor and LeeseHeads 1997; Reference Weeks and WidomWeeks 1998).

Although these studies do not prove a causal link between early attachment difficulties and the later development of antisocial personality disorder, the primacy of genetic and constitutional factors in the development of psychopathy is being challenged by emerging evidence that links callous and unemotional traits and psychopathic tendencies in children with disorganised attachment relationships (Reference Pasalich, Dadds and HawesPasalich 2012), maladaptive interactive patterns in families (Reference Dadds, Allen and OliverDadds 2012) and severe institutional deprivation (Reference Sonuga-Barke, Schlotz and KreppnerSonuga-Barke 2010).

Mentalisation deficits

Reference Bateman and FonagyBateman & Fonagy (2008) propose that a substantial group of patients who fulfil diagnostic criteria for antisocial personality disorder have experienced significant trauma and disruptions to their attachment system in childhood, which has interfered with their neurobiological development and the development of psychological defences. This compromises their capacity to mentalise and lowers their threshold for emotional reactivity.

Mentalisation is the capacity to reflect and to think about mental states, including thoughts, beliefs, desires and affects, to be able to distinguish one’s own mental states from others, and to be able to interpret the actions and behaviour of oneself and others as meaningful and based on intentional mental states (Reference Allen, Fonagy and BatemanAllen 2008). Secure attachment in childhood is associated with the development of enhanced mentalisation processes, which in turn contribute to the homeostatic regulation of negative affect and arousal regulation (Reference SchoreSchore 1996, Reference Schore2001; Reference Sarkar and AdsheadSarkar 2006). Children who experience significant childhood abuse, neglect or domestic violence, however, will be unable to feel safe about what others think of them, which may lead to a poorly developed capacity to mentalise, deficits in empathy, and a difficulty in distinguishing one’s own mental states from others’.

Violence and mentalisation

Violent offenders may be divided into a high-affect, impulsive and hyperaroused group, and a low-affect, unemotional and psychopathic group (Reference MeloyMeloy 2006; Reference Adshead, Pfafflin and AdsheadAdshead 2013). The first group are more likely to become violent in response to perceived threats to their sense of self. Reference Bateman and FonagyBateman & Fonagy (2008) propose that such reactive or affective violence arises where there is a failure or inhibition of the capacity for mentalisation. A person with a limited capacity to mentalise is unable to tolerate negative emotions such as anger, hatred and a wish to hurt others, but instead may become highly aroused and overwhelmed with negative affects. For the person with antisocial personality disorder who is often particularly sensitive to threats to their self-worth or ‘respect’, shame and humiliation are particularly difficult emotional states to bear, and may be experienced as threatening their very psychic survival (Reference GilliganGilligan 2002). These unbearable feelings cannot be managed by representational means within their mind, but are experienced very concretely as feelings that need to be expelled through violence. The expression of aggression is further potentiated by the reduced capacity of the individual to mentalise – if they are unable to see others as having mental states different from theirs, this will reduce the inhibition of their aggression and violence towards others as they are unable to empathise or appreciate another person’s suffering. Mentalisation therefore protects against violence, and individuals whose capacity for mentalisation is reduced are more likely to be violent.

Assessment

Treatment planning starts with a careful clinical assessment, including identification of comorbid treatable conditions such as depression or substance misuse, and assessment of key personality traits which indicate a positive response to treatment and prognosis, particularly some capacity for attachment and ability to form relationships, experience of anxiety and willingness to seek help (Box 3). Careful risk assessment, with particular attention to the patient’s history of violence, should be conducted, ideally with the use of validated instruments such as the Historical Clinical Risk-20 (HCR-20; Reference Webster, Douglas and EavesWebster 1997) or the Violence Risk Appraisal Guide (VRAG; Reference Harris, Rice and QuinseyHarris 1993). The severity of psychopathy is determined with the PCL-R. Caution should be administered before considering treatment for individuals scoring highly (above 30) on the PCL-R. Reference MeloyMeloy (1988, Reference Meloy, Yakeley, Clarkin, Fonagy and Gabbard2010) recommends that therapy will be of no benefit and should not be offered to patients with psychopathy who manifest any of the following features:

1. • sadistic, aggressive behaviour resulting in serious injury

2. • complete absence of remorse or justification for such behaviour

3. • very superior or mildly intellectually disabled intelligence

4. • a historical absence of capacity to form emotional attachments

5. • unexpected ‘atavistic’ fear felt by the experienced clinician in the patient’s presence.

Creating a safe setting

An essential component of management is the setting in which treatment is delivered, wherein containment, risk, boundaries and disclosure of information are paramount (Reference Meloy, Yakeley, Clarkin, Fonagy and GabbardMeloy 2010). The treatment setting, whether it be a high-, medium-or low-security in-patient setting, prison or the community, must be secure enough to ensure the safety of both patients and staff before treatment planning can begin.

An attachment perspective highlights the relational aspects of the treatment setting and how the patient’s early history of attachment difficulties will influence their response to care. Individuals with antisocial personality disorder are less likely to engage in help and seek treatment, and when treatment is offered they often reject it. This dismissing stance towards relationships and attacking and/or neglectful relationship to care (Reference RuszczynskiRuszczynski 2010) may be conceptualised as reenactments of earlier experiences of abusive or neglectful caregivers and consequent disruptions of childhood attachment. Such challenging behaviour, however, may provoke powerful negative emotional responses in clinicians and other professionals. If unrecognised, these may lead to problematic reactions and behaviours in staff such as high staff turnover and ‘burnout’, avoidance of patient contact, boundary violations, pathological team and institutional dynamics, and passive-aggressive or punitive attitudes towards patients such as denying leave or overestimation of a patient’s risk.

Risk can only be managed safely if the anxieties of patients, individual staff and the institutions in which they work are adequately contained (Reference YakeleyYakeley 2010). Underlying anxiety that may trigger aggressive and antisocial acts in patients is often unrecognised, but equally, the anxieties of the staff induced by managing such patients are also frequently unacknowledged or denied. Managing negative affects in staff is a critical component of relational security, which includes recognition and analysis of the powerful countertransference reactions that these patients may provoke in those involved in their care (Box 4). Supervision and reflective practice are essential in helping staff think about how they are affected, both consciously and unconsciously, by the patients with whom they work.

Mode of treatment

The mode in which therapy is delivered can be viewed in itself as a vehicle for psychic change. The therapeutic mode can be individual, group or family therapy, or the therapeutic community itself, using its distinctive culture as a therapeutic tool. Group therapy has particular advantages for individuals with poor impulse and affect control because it is less arousing than individual therapy, and also offers more opportunities to understand other peoples’ minds (Reference Bateman and FonagyBateman 2004). Other group processes are also important, such as modelling of more appropriate behaviour and interpersonal interactions.

Engagement

Engaging and maintaining the person in treatment is a key challenge. The majority of individuals do not accept that they have mental health difficulties and will not willingly present for help. Treatment non-completion is also a significant problem, with up to 75% of offenders with personality disorder who initially engage in treatment failing to complete the course (Reference McCarthy and DugganMcCarthy 2010). Treatment non-completers show poorer clinical outcomes than do treatment completers, as well as having significant financial cost implications for services and lowering staff morale (Reference McMurranMcMurran 2012).

Referrals for treatment may be prompted by professionals who are more concerned about reducing risk of future violent or antisocial behaviours than effecting improvements in the person’s mental health or intrapsychic change. Although many programmes for offenders in secure mental health institutions or prison require the person to volunteer for treatment, motivation is often overestimated. Offenders who appear compliant and motivated in residential programmes may revert to impulsive and risky behaviours once released into the community. Moreover, it is questionable as to whether medical services should be providing an intervention whose primary aim is to reduce reoffending. Therapy should be aimed at addressing mental health and psychological factors, such as anxiety or paranoia, that contribute to the violent behaviour, rather than the latter being the sole focus of treatment.

Therapists treating patients in the community should anticipate numerous challenges, including lateness or failure to attend appointments and therapy sessions, frequent crises, boundaries violations, and drug and alcohol misuse. To minimise attrition and non-adherence to treatment, careful pre-treatment preparation is recommended. A variety of strategies may be useful, including psychoeducation about personality disorder, goal-based motivational interviews, instilling beliefs about capability for change, and implementing strategies to modulate known factors likely to reduce adherence, such as substance misuse, self-harm or a chaotic life-style (Reference Dowsett and CraissatiDowsett 2008; Reference McMurranMcMurran 2012). Proactive follow-up of patients, such as calling to remind them of appointments and positively reinforcing their engagement in therapeutic tasks, will also increase adherence.

Many individuals may need an extended period of engagement or motivational work, which may occur in unconventional settings (such as the individual’s hostel) and in the absence of a comprehensive assessment. Even if such psychological management does not lead to engagement in formal treatment, it may be therapeutically effective in itself.

Building a therapeutic rapport and collaborative relationship with the patient is essential. The treatment relationship should be characterised by caring, fairness and trust, and an authoritative, but not authoritarian, style. Interventions that are punitive or that focus on control and surveillance may increase risk of recidivism if not combined with rehabilitative efforts (Reference Skeem, Polaschek, Manchak, Skeem, Douglas and LilienfeldSkeem 2009).

Establishing a group process

Whatever the treatment model, the initial task in group therapy is to establish a group process for patients whose long-standing interpersonal difficulties, suspicion and lack of trust will inevitably enter into their relationships with others in the group. This calls for active intervention on the part of the therapists in fostering links between the discourses of individual patients, who initially tend to show little interest in each other. A common tendency is for group members to drift into discussions about the problems of society where people will always take advantage of others. Although this may represent an attempt to form a common bond within a paranoid view of the world, it also marks a shift into a mode of thinking which may defend against talking about more difficult emotive topics.

Linking actions to affects and internal mental states

An important component of many psychological therapies for personality disorder is to link the patient’s damaging behaviours to their internal mental states and affective experiences. However, individuals often find it very difficult to reflect on their mental states or to link their violent and antisocial behaviour with their thoughts, feelings and emotions. Asking them an apparently simple question such as ‘What were you feeling before you attacked him?’ can make them feel stupid and humiliated. Such individuals display a narcissistic fragility and hypersensitivity to being criticised, corrected or even interrupted, and anger may be easily activated when a patient is attempting to describe an emotive topic. Such threats to the patient’s self-esteem (being ‘disrespected’) are important to identify, as these may expose unbearable feelings of shame that are frequently the trigger for violent behaviour. If the patient becomes too aroused, attention may be temporarily deflected away from them until they calm down by asking the other patients what they think, the therapists reflecting themselves on the situation or, if the patient appears very angry or distressed, by changing the subject altogether until the group’s reflective capacity can be regained.

Because individuals with antisocial personality disorder find it very difficult to empathise with other people’s affective states, interventions aimed at consideration of their effects on others, or ‘victim empathy’, are likely to be counterproductive and should be avoided in the early stages of treatment. Instead, interventions focusing on improving self-regard and promoting social and interpersonal success may contribute to creating a positive and hopeful atmosphere in the group (Reference Bateman, Fonagy, Bateman and FonagyBateman 2011).

Boundaries

Individuals with antisocial personality disorder experience relationships in terms of power and control, and issues of dominance and hierarchy will inevitably pervade the therapy. Rules and boundaries, which are important in any treatment, are a central feature in group therapy. Based on their own early attachment experiences of maltreatment they have a distrust of parental figures and authority and will rebel against whatever rules are imposed, including those of the group, but may have their own strict code of conduct. Boundary violations are therefore to be anticipated.

Therapists must be prepared to tolerate some expression of anti-authoritarian attitudes, acting as parental figures against which the patients can rebel, until such attitudes can be safely explored and understood within the group. A culture of trust, openness and honesty may be nurtured in the context of attachment relationships with the other group members by first helping the patients explore their own code of conduct and interactions with others within and outside of the group, and identifying the affects and states of mind that lead to violence. Developing a sense of responsibility and awareness of appropriate boundaries in relation to others is an important task of treatment, although this may be difficult in patients who have grown up in a criminal subculture.

Breaks in therapy

Breaks in treatment are particularly difficult times for these patients and are associated with an increased risk of ‘acting out’. The potential impact of the therapist’s absence may be underestimated due to the erratic attendance of the patients themselves and their denial of attachment needs. It is important not to collude with the patient’s conscious rejection of their need for treatment, and be alert to patients becoming more disturbed during breaks in treatment, which they may find difficult to acknowledge. Similarly, difficult feelings regarding the ending of therapy should be anticipated and discussed if possible to avert premature drop-out.

Gender

We propose that male and female patients should be treated separately, and that specific services should be developed for women. This is because women present with a somewhat different diagnostic and needs profile to men, showing a higher incidence of comorbid borderline personality disorder and associated self-harm, as well as complex issues regarding pregnancy and care of their children, with the frequent involvement of Social Services and the family courts. Furthermore, our experience is that mixed groups of men and women may expose negative male attitudes towards women, which can be disruptive and contribute to poor engagement and outcomes among female patients.