Book Reaches Sweeping Conclusions About Forensic Services

Dr. Grant Harris

One of my favourite movies is What About Bob? Richard Dreyfus plays a pompous and uptight psychiatrist tormented by his crazy patient (played by Bill Murray) who turns out to be saner than the psychiatrist. At their first meeting, the patient, Bob, exhibits a hilariously wide range of the most severe neurotic symptoms imaginable. Dr. Leo Marvin merely uses this as an opportunity to give Bob a copy of his "ground breaking new book, Baby Steps." As soon as Bob leaves, Dr. Marvin sends him a bill for the session and for the book.

Recently, members of the Research Department celebrated their own new book - Violent Offenders: Appraising and Managing Risk by Vern Quinsey (former Research Director), Grant Harris, Marnie Rice, and Catherine Cormier; published by the American Psychological Association, Washington, DC (1998, 356 pp). Over the last 25 years, members of our research Team have published more than 150 scientific or professional articles and book chapters on the general topic of crime and violence. We wanted to integrate the results of all this work in one place.

The book begins with a review of efforts to assess, treat and manage violent offenders. It shows that progress has been slow over the last two centuries with a curious repetition. Reforms to public policy are made with great enthusiasm but without systematic approaches. When scant resources are devoted to reforms and effectiveness not measured, they soon become reforms in name only.

Public disenchantment and calls to abolish the reforms quickly follow, and then the cycle repeats itself. That is, public policy is characterized by cycles of change followed by retrenchment, but little progress. In the realm of violent offenders, we attribute this cycle to the failure to apply scientific methods to assessment and treatment.

The book reviews what is known about how mental health (including forensic) clinicians assess and treat violent offenders. When they make decisions based on unaided clinical judgment, they perform poorly on many tasks.

They show low agreement with each other. Their judgments lack stability due to random fluctuations. Despite assertions to the contrary, they do not use specialized assessments or technical information. Their performance is little affected by experience and training so that they perform as well as lay people as assessors and therapists. None of this is meant to say there exists no useful knowledge about assessment and treatment. It does mean, however, that many clinicians do not use that knowledge.

The book also summarizes follow up studies of released offenders. This research paints a clear picture -- certain objectively measured personal characteristics are related to violent recidivism in all offenders. These include, for example, age, criminal history, substance abuse, history of aggression as a child, and antisocial or psychopathic personality.

The consistency of these findings makes it possible to develop a means to evaluate risk with wide application among violent offenders.

The book illustrates the validity of this conclusion as applied to fire setters and sex offenders (and describes our research with these groups). We then describe our work developing the actuarial Violence Risk Appraisal Guide (and Sex Offender Risk Appraisal Guide, the version used for sex offenders). Included are summaries of several later tests, by ourselves and independent researchers, establishing the accuracy of these instruments in new samples.

Based on our own work and more research over several decades, we recommend a change to the way the risk of violent offenders is appraised. We advise that it not be a matter of clinical opinion. Based on the available science, we recommend that clinical opinion be replaced by actuarial tools such as the Violence Risk Appraisal Guide.

We argue against merely incorporating actuarial scores into clinical opinion because this will lead to less than optimal improvements in performance. Many clinicians dislike these recommendations and resist them.

In a later section of the book we tackle the most cogent arguments against our seemingly radical recommendations. We try to show how these arguments are based on ignorance and misunderstanding of the science or law, or the intrusion of clinicians' self interest.

The longest chapter is about therapy for violent offenders. Sadly, very few drug and psychological treatments have been properly evaluated. The treatments we evaluated ourselves did not work. We do know that some treatments cause harm by increasing violence. We know that others have no effect at all.

But there is knowledge about what has the best chance of working and we describe how interventions should be organized. We devote a little space to what all our research has taught us about the causes of crime and violence, and how our thinking has changed over the years. And there is a long Appendix to teach the interested reader how to use the Violence Risk Appraisal and Sex Offender Risk Appraisal Guides.

One of our primary goals in writing this book was to summarize all our research in a way that would be useful to staff working in forensic settings.

We believe that our work is directly applicable to forensic clinicians' daily tasks and we hope to see it applied. So far, we are pleased with our book's reception; it is a bestseller for the publisher, the American Psychological Association, and is in its second printing. There are copies in the Library and we hope staff will give it a look. And I recommend What About Bob?, available on video.