Figure 32.1

Risk profile assessed using 20 VRS dynamic factors. The sample of 918 is a male adult offender normative sample for the VRS (see Wong and Gordon [33]). The psychopathic sample was identified from an offender sample using PCL = R cutoff of 30.

This figure was adapted from Wong and Gordon [33].

Recent research has shown that changes of the VRS dynamic factors assessed within a treatment program were associated with a subsequent reduction in violent reoffending post-release in the community among male high‐risk PCL-R–assessed psychopathic offenders with no active psychotic symptoms [34,35]. Analogous results were obtained for psychopathic sexual offenders assessed using the Violence Risk Scale – Sexual Offender version (VRS-SO) [36] designed for sexual offenders [15,37]. The results suggest that the dynamic factors of the VRS and VRS-SO are modifiable and satisfy the criteria for causative dynamic factors [38]. Once the treatment targets are identified, risk reduction treatment can proceed.

The 2-C model is consistent with the generic and specific factors set forth by Livesley [39–41] for the treatment of personality disorders. The generic factor entails establishing therapeutic and supportive engagements between therapists and clients, vis-à-vis, the interpersonal C1 component, whereas the specific factor includes interventions that target the individual’s specific problem areas, vis‐à-vis, the criminogenic C2 component. The risk/need/ responsivity (RNR) principles are widely accepted as important principles to guide risk reduction treatment of offenders [42,43]. Higher‐risk offenders should receive more intensive treatment (the Risk principle); treatment should be directed toward the person’s criminogenic needs, that is, the causes or closely linked attributes of the criminal behaviors (the Need principle), and, treatment delivery should be tailored to the person’s learning and response style such as the level of motivation, engagement, and intellectual abilities (the Responsivity principle). Risk and Need closely map onto C2, whereas Responsivity maps onto C1.

The 2-C model is also consistent with the National Institute of Health and Clinical Excellence (NICE, UK) guidelines for the treatment of antisocial personality disorder including psychopathy [44]. The guidelines assert that persons with antisocial personality (including psychopathy) should not be excluded from any health or social care service because of their disorder or offending behaviors (p. 7). For reducing reoffending, the guidelines recommend the following: (1) using CBT group-based approaches, (2) adapting treatment to suit the individual, (3) monitoring treatment progress, and 4) providing appropriate staff training and support (pp. 16–18). Pharmacological interventions, however, should not be routinely used for the treatment of antisocial personality disorder or associated behaviors of aggression, anger, and impulsivity (p. 16).

Treatment programs with design and delivery similar to the 2-C model have produced positive outcome results. The Violence Reduction Program [45] and the Clearwater Sex Offender Programme [46–48] are two examples. During such treatment, offenders’ criminogenic needs linked to sexual and nonsexual violence (F2), such as criminal attitudes and beliefs, sexually deviant interests, interpersonal aggression/hostility, substance use, etc., are assessed and identified as possible treatment targets using the VRS/VRS-SO and clinical evaluations. Cognitive-behavioral group and/or individual interventions are used, if appropriate, in a structured but flexible manner to modify antisocial thoughts, feelings, and behaviors. Practice and generalization of socially appropriate behaviors to day-to-day living are very much encouraged and supported with ongoing close monitoring guided by what we referred to as Offence Analogue and Offence Reduction Behaviors (OAB and ORB, respectively) protocols [49]. OABs are the proxies of offending behaviors that manifest within an institutional context, and ORBs are the prosocial counterparts to replace the OABs in day-to-day functioning. Each VRS-identified treatment target should have corresponding OABs and ORBs. The here-and-now OABs are behaviors that treatment staff can focus on, and, using appropriate interventions, can assist offenders to learn to replace them with ORBs. To address F1-related issues, motivational and engagement work are emphasized throughout the program using, for example, motivational interviewing principles [50]. Intensive staff training to manage treatment-interfering behaviors and appropriate staff supervision and support are also important program components. The programs, about 8–9 months in duration, are suitable for both offenders with a significant history of nonsexual and sexual violence as well as for psychopathic offenders. The close integration of risk assessment and risk reduction treatment is essential in the program’s implementation [51].

For MDOs with histories of violence, the 2-C model also can be used to guide risk reduction treatment once acute psychiatric symptoms are well managed, controlled, and carefully monitored and the person has regained a sufficient level of daily functioning to attend to risk reduction treatment requirements (see the next section).

Summary

The 2-C model is developed based on integrating the psychopathy assessment, risk assessment, and offender rehabilitation literatures to guide violence reduction treatment of high-risk and/or psychopathic offenders. Treatment should target the person’s modifiable criminogenic features, analogous to F2 characteristics, which are identified using an appropriate dynamic risk assessment tool. Offenders can then learn, practice, and generalize offense-reducing thoughts, feelings, and behaviors to replace offense-producing behaviors in day-to-day functioning. Staff must closely monitor and manage treatment-interfering behaviors (linked to F1 features) to maintain treatment engagement and integrity. Treatment targeting F1 features, though intuitively appealing as they appear to target the most salient and obvious psychopathic personality traits, will unlikely reduce violence recidivism even if changes were successfully made, as these traits are not linked to future violence. Outcome evaluations of programs similar to the 2-C model have shown some positive results [15,16,35].

Psychopathy, Mental Disorder, and Violence

The majority of mentally ill persons are not violent. Among major mental disorders, psychosis has the closest link to violence. In a meta-analysis using 166 independent data sets, psychosis was associated with a 49%–68% increase in the odds of violence [52]. Again, most persons with psychosis are not violent. A recent systematic review and meta-analysis based on 110 eligible studies by Witt et al. [53] investigated static and dynamic predictors for aggression and violence among MDOs formally diagnosed with psychosis, the majority with schizophrenia (total n = 45,533 adults; 87.8% schizophrenia, 0.4% bipolar disorder, and 11.8% other psychoses). The sample base rate of violence was 18.5%. The strongest predictor for all aggression or serious violence was criminal history – a static predictor. The dynamic predictors were hostile behaviors, poor impulsive control, recent drug/alcohol misuse, lack of insight, and noncompliance with psychological therapies and medication; the predictors were essentially the same for aggression vs. severe violence as well as for inpatient vs. community or mixed settings, although the strengths of association varied. In Figure 32.1, the dynamic predictors identified for the MDOs with psychosis are marked with a double asterisk (**). The static criminal history predictors should have a number of likely underlying dynamic counterparts, such as violent (criminal) lifestyle, criminal attitude, criminal peers, violence (criminal) cycle, and so forth that are a part of the VRS dynamic factors marked in Figure 32.1 with a single asterisk (*). (For a more detailed discussion of this point, see Wong and Gordon [33].) The overlaps of dynamic violence predictors for the three groups are considerable (Figure 32.1), although the data were collected using very different methodologies. These findings are consistent with two meta-analyses, both showing criminological, rather than clinical, variables to be better predictors for violent and general recidivism for MDOs [54,55]. A recent study with MDOs and non-MDOs on parole also obtained very similar results [56]. Given the similarities in the risk factors for the three groups, it is possible that they share similar etiological pathways.

Developmental Trajectory of Schizophrenia

In the past two decades, the extant literature, including large longitudinal cohort studies, has identified three different types of MDOs with schizophrenia (MDO-S) with different developmental trajectories (types I, II, and III; see Hodgins [57] for a review). Type I or MDO-S early starters are those whose conduct problems start before their illnesses, with an onset around late adolescence or early adulthood. Their significant childhood conduct problems persist into adolescence and adulthood, often resulting in a record of quite diverse criminal behaviors. These Type I MDO-Ss share many similarities with life-course persistent antisocial offenders without mental illness [58]. The Type II MDO-S presents with no history of antisocial or aggressive behavior prior to illness onset (late onset), after which they repeatedly engage in many externalizing aggressive behaviors. Given their late onset, they generally accumulate fewer criminal convictions compared to the Type I. Of importance, a larger proportion of the Type II MDOs had been convicted of homicide than the Type I [59]. Type III MDOs with schizophrenia are likely men in their late thirties with no history of antisocial or aggressive behaviors who kill or try to kill someone who is likely their care provider. Many of the MDO-S cases in Witt et al.’s [53] study also had a significant criminal history, substance abuse problems, hostility, and impulsivity that were predictive of future violence – characteristics similar to the Type I MDO-S cases.

A separate study in Sweden investigated all men who underwent pretrial psychiatric assessments and were later convicted of violent offenses in a six-year period; 202 men were diagnosed with schizophrenia (the MDO-S cases), and 78 met PCL-R criteria for psychopathy without mental disorder [59]. Twenty-nine percent of the MDO-S obtained high scores on the PCL-R and they appear to be similar to non-mentally ill men with psychopathy. The high ratings of psychopathy are associated with earlier ages of first conviction for a criminal offense and more convictions among the men with schizophrenia, just as among men with no mental illness [59]. It is not unexpected that both MDO-S and non-MDOs who met PCL-R criteria would share similar criminological features, since high PCL-R ratings as well as the presence of antisocial personality disorder [60] would signal an early-onset and persistence of conduct problems, substance abuse, juvenile delinquency, criminal versatility, and so forth, essentially PCL-R F2 features.

Among MDOs with schizophrenia, those with higher PCL-R scores are more likely to be found among Type I early starters than Type II or Type III. We hypothesize that among MDO-S, the presence of high PCL-R scores is probably a proxy indication of life-course persistent antisocial behaviors, that is, a preponderance of PCL-R F2 features more so than F1 core psychopathic personality traits. In fact, it was noted that in the non-offender population, few MDOs with schizophrenia have PCL-R ratings that satisfy the criteria for psychopathy, and characteristics such as glibness, superficial charm, promiscuity, and many short-term relationships (PCL-R items) are rarely observed among them [61]. It is also possible that the ratings of some PCL-R F1 items, such as shallow affect, lack of guilt or remorse, callous/lack of empathy, could be confounded by the presence of negative symptoms of schizophrenia, thus artificially inflating PCL-R scores. It remains to be seen what PCL-R composite and factor scores Type I MDOs would obtain should the ratings be made based only on their personality characteristics assessed prior to the onset of their illnesses.

If our hypothesis was correct and the relatively high ratings on the PCL-R among Type 1 are mainly due to the preponderance of F2 rather than F1 features, it would follow that risk reduction treatment of these MDOs should address their violence risk predictors (proxy of F2 features), not unlike the treatment of non-mentally ill offenders with psychopathy. A comprehensive risk assessment using an appropriate dynamic risk assessment tool should inform what risk factors are present that can be used as treatment targets; treatment delivery can be similarly guided by the proposed 2-C model. Assessing the possible presence and extent of F1 features would inform us of how best to manage the person to reduce the impact of treatment-interfering behaviors such as disruption of treatment group, staff splitting, etc.

Conclusion

Recent advances in the assessment of psychopathy, risk assessment, and offender rehabilitation have enabled the integration of these literatures to inform risk reduction treatment of psychopathy as illustrated by the recently developed two-component (2-C) treatment model for violence-prone psychopathic offenders. Parallel advances in the study of MDOs, in particular those with schizophrenia, have also shed light on their characteristics and possible etiology. This article reviewed the literature and extends the 2-C treatment model to mentally disordered offenders with schizophrenia, violence, and psychopathy with supporting evidence.

Disclosures

The authors do not have anything to disclose.

References

1.Felthous AR. The “untreatability” of psychopathy and hospital commitment in the USA. Int. J. Law Psychiatry. 2011; 34(6): 400–405.

2.Cleckley H. The Mask of Sanity, 5th edn. St. Louis, MO: Mosby; 1976.

3.Hare RD. The Hare Psychopathy Checklist–Revised, 2nd edn. Toronto, ON: Multi-Health Systems, Inc.; 2003.

4.Skeem JL, Polaschek D, Patrick C, Lilienfeld SO. Psychopathic personality: bridging the gap between scientific evidence and public policy. Psychological Science in the Public Interest. 2011; 12(3): 95–162.

5.Cooke DJ, Hart SD, Logan C, Michie C. Explicating the construct of psychopathy: development and validation of a conceptual model, the Comprehensive Assessment of Psychopathic Personality (CAPP). International Journal of Forensic Mental Health. 2012; 11(4): 242–252.

Only gold members can continue reading. Log In or Register to continue