What treatment should psychopaths receive? The next two pages present two sides of the debate over how psychopaths should be treated

What treatment should psychopaths receive? The next two pages present two sides of the debate over how psychopaths should be treated

Grant T. Harris

So little well-controlled research exists that no conclusions can be drawn about the efficacy of treatment for psychopathy. But the available work does offer some ideas about what is contra-indicated. Some readers might recall our research on an intensive therapeutic community designed for the treatment of psychopaths in the 1960s and 1970s, which appeared in a 1992 issue of Law and Human Behavior. It emphasized insight-oriented therapy, was emotionally evocative and placed patients in clinical leadership roles. The program was faithful to its operating principles and based on the best available information about psychopathy. Clinicians and outside experts felt the program was effective. Our follow-up research showed, however, that although the program reduced recidivism among non-psychopaths, it increased the violent recidivism of psychopaths (compared to prison).

We think this reveals valuable lessons: Clinicians cannot assume their efforts are beneficial; it is possible to do harm. This has been demonstrated elsewhere–some well-intentioned services increase the likelihood of crime. Treatments to reduce recidivism must follow treated patients after release to assess recidivism, and evaluation must include an untreated comparison group.

Our study showed that psychopaths actually behaved more poorly in the program (compared to other patients), but were as likely, or even more likely, to be trusted by clinical staff. Other researchers have shown that clinicians’ impressions are a poor index of the benefits of therapy, especially concerning psychopaths. Perhaps most important: Psycho-dynamic, insight oriented, emotionally evocative therapy should not be provided.

Researchers have attempted to identify what programs are effective for criminal offenders in general (not necessarily psychopaths). The reviews all arrive at similar conclusions: Insight-oriented, emotion-based therapy should not be provided. More severe punishment and trying to “scare ’em straight” are also ineffective. Effective programs teach something useful–academic, vocational, social or personal management skills. Effective programs are firm but fair. Although offenders might appear depressed, deliberately targeting self-esteem would not decrease risk and might even increase it. Psychopaths generally have high self-esteem, and among offenders, high self-esteem is related to aggression.

We believe that psychopaths do not have deficits in the clinical sense. Natural selection has designed psychopaths to follow a fundamentally different life strategy–one emphasizing deception, aggression and indifference to the welfare of others. What is known about psychopaths is consistent with the view that they do not have a mental disorder as it is usually defined.

Fortunately, not all clinical interventions, even psychopharmacological ones, require a clinical deficit in need of a remedy. Is there a drug that would reduce the risk of psychopaths? There is not much research on drug treatment for psychopathy, but it is a sensible avenue for exploration. Psychopaths have a different physical make up; giving them a drug to make their neurophysiology more similar to ours might make them less dangerous. But careful evaluation is essential; impressions of efficacy cannot be trusted.

The best example of a clinical intervention that does not require there to be a disorder is behaviour modification. There is empirical evidence that this approach has worked with some offender and violent populations (although not psychopaths). We believe the evidence favours a strategy that applies behavioural principles to reducing the harm caused by psychopathy. Where psychopaths have already committed serious offenses are at high risk of future violence, we favour using selective incapacitation in the form of long-term institutionalization with behavioural methods to manage day-to-day psychopathic behaviour.

We also favour the use of a sophisticated token economy. Such a program is explicit and concentrates on reinforcing behaviour incompatible with psychopathic conduct (i.e., delaying gratification, telling the truth, being responsible, being helpful, being cooperative) and penalizing impulsive, dishonest, aggressive, irresponsible and criminal actions. There is no expectation that the program will ever end. Consequences for behaviour are consistently monitored by staff and always based on observed, overt behaviour, never on what inmates report about thoughts, feelings or conduct.

However, conditions would permit the use of this strategy with only a minority of psychopaths (and a very small minority of offenders). For most psychopathic offenders, release to the community in the form of parole or probation will inevitably occur. The greatest prospect for an effective intervention lies in the challenge of applying these wrap-around behavioural principles to psychopaths under conditional release.

There are no data indicating that any treatment targeting psychopathy has been effective in reducing anti-social conduct. The available evidence affords no guarantee that intensive versions of current offender programs will be effective for psychopaths because research shows that psychopaths are fundamentally different from other offenders. Therapists’ impressions of improvement should not be trusted in the evaluation of treatment. Insight oriented, emotionally evocative, relationship-based therapy should not be provided. Therapy should not seek to increase self-esteem. Because there is a real possibility that an intervention could cause harm, all interventions should be accompanied by rigorous evaluation assessing violent and criminal recidivism. No assurances from clinicians should permit them to escape this requirement.



the last word

Howard E. Barbaree

In the accompanying article, Grant Harris and Marnie Rice state that treatment is potentially harmful for psychopaths. The “treatment causes harm” hypothesis has been articulated by leading scholars in the field, in their 1997 chapter in the Handbook of Antisocial Behavior, Steve Hart and Robert Hare suggested it was possible that “group therapy and insight-oriented programs help psychopaths to develop better ways of manipulating, deceiving, and using people but do little to help them to under stand themselves.”

If treatment does cause harm to psychopaths (or to put it more aptly, if treatment causes them to harm others more frequently), some very difficult problems arise. While specific treatment is rarely provided for psychopathy per se, psychopaths are often included in treatment programs targeting other problem behaviours, such as substance abuse, sexual deviance and anger management. Such treatment programs are regular features in psychiatric and correctional settings. Should psychopaths be excluded from them?

The implementation of such a policy in most hospital and correctional settings would be complex and difficult, raising all sorts of questions about a patient/inmate’s right to treatment. If release to the community is contingent on successful treatment, as it often is, exclusion from treatment could impair their access to community release.

The “harm” hypothesis also speaks to the assessment of the psychopath’s suitability for release to the community. Detained inpatients or inmates who are seeking some kind of community release depend on favourable decisions by parole or review boards. Would psychopaths be viewed as less deserving of community release if they had participated in treatment? Of course, in the present climate, the reverse is in effect–the more treatment the offender has successfully completed, the more likely he is to receive a favorable decision. The “harm” hypothesis places psychopaths in a veritable “catch-22.”

The important question to address is whether sufficient empirical evidence exists in support of the “harm” hypothesis to justify, taking the drastic step of excluding psychopaths from treatment.

In a 1992 article in Law and Human Behavior, Harris and Rice, along with co-author Catherine Cormier, describe their evaluation of a therapeutic community intervention as the strongest support for the hypothesis. Inpatients treated in a therapeutic community in a maximum-security psychiatric hospital were compared with untreated subjects sampled from a correctional setting. Comparing the groups overall, there were no differences in recidivism. However, separating the groups into psychopathic and non-psychopathic sub-groups, treated psychopaths were more likely to re-offend violently than untreated psychopaths, whereas treated non-psychopaths were less likely to re-offend violently than untreated non-psychopaths.

In the sex offender treatment literature, Harris and Rice have been the harshest critics of studies that claim a treatment benefit. They have set out standards for what they claim are minimally informative evaluations of treatment efficacy. These standards promote the use of the randomized controlled trial (RCT) over other research designs. According to their 2003 article in the Annals of the New York Academy of Sciences, Rice and Harris state, “To be even minimally useful, the groups need to be comparable on (1) established static predictors of recidivism, (2) jurisdiction and cohort, and (3) volunteering and completing treatment.”

Since I cannot think of any rationale for recommending or accepting different standards for evaluating treatment harm compared with treatment benefit, it might be helpful to apply their standards to their 1992 follow-up study.

First, the study was not a prospective RCT, but a retrospective convenience design. Second, the treated group was detained in a psychiatric hospital having been found “not guilty by reason of insanity” (NGRI). The untreated group were men who had been refitted to psychiatric hospital for assessment, but who were not found NGRI. These two groups were therefore not equivalent in some important aspect of criminal responsibility.

Third, the treated group was released to the community through the mental health system of outpatient follow-up where mental health professionals would have had no legal jurisdiction forcing compliance to treatment directives. In contrast, the comparison group was released to the community through a process of case management in the federal and provincial probation and/or parole services. In this system, the authorities can re-arrest and re-incarcerate for noncompliance with a case manager’s directions. I would argue that these two groups were not comparable in jurisdiction. In Ontario, the mental health system operates quite differently from the correctional system.

Harris and Rice propose a treatment for managing psychopaths in institutional and community settings. Interestingly, that treatment is similar in many respects to how correctional authorities have come to manage psychopaths and other difficult criminals. institutional rules are established and communicated. Infractions to the rules are punished with absolute (as far as possible) consistency. Most psychopaths learn to live without difficulty in an environment with this kind of structure. Contrast the structure in the correctional facilities with the complete absence of structure in the therapeutic community. Perhaps a better way of conceiving of the study result is to think of the correctional structure as the treatment, its absence in hospital as the comparison or control-condition, and the group differences described by Harris and Rice as indicating an effective treatment for psychopaths.

The current empirical support for the “harm” hypothesis is weak and does not justify excluding psychopaths from treatment programs that target specific problem behaviours using cognitive-behavioural therapeutic techniques.


COPYRIGHT 2006 Centre for Addiction and Mental Health

COPYRIGHT 2007 Gale Group