Predicting Extreme Fatal Violence

The idea of "dangerousness" has been a paramount issue in the forensic psychiatric arena for many years, yet establishing an empirical body of data from which to make accurate predictions has been difficult. According to John Monahan and his colleagues, who looked at the over-prediction of threat during the 1980s that resulted in a high percentage of unnecessary commitments, such research must meet specific criteria. Notably, risk factors must be segregated into component parts, the potential for harm should be assessed only in terms of probability, and the research must be done on representative samples.

Monahan claimed that these criteria were met in a study undertaken by the MacArthur Foundation, wherein researchers examined the relationship between mental disorder and violent behavior. (Others complain that it was not representative.) They devised a comprehensive list of risk factors across four domains: the person's disposition, his or her history of violence, the context for potential violence, and any past or current clinical issues. Of the four domains, only contextual and clinical were deemed relevant to risk management, because these factors could be changed.

But among those factors that were most significantly linked with violence were the following: being male, having a prior record of violence or aggression, physical abuse in childhood, having a parent who was a substance abuser or criminal, living in a disadvantaged neighborhood, having a diagnosis of an adjustment disorder or substance abuse, evidence of psychopathy (the strongest factor), a suspicious attitude toward others, an experience of an auditory hallucination that commanded a violent act, and thinking or fantasizing about harming others.

The latest update on the MacArthur Violence Risk Assessment Study indicates that, compared to other instruments that assess those already hospitalized, its results appear to be highly accurate. However, it is a complex and time-consuming approach, so professionals have now devised computer software to assist the process. It's become clear that clinical appraisal is too subjective to be effective, and replacing that with statistical data is both more scientific and more reliable. The study observes, "our data are most consistent with the view that the propensity for violence is the result of an accumulation of risk factors, no one of which is either sufficient or necessary for a person to behave aggressively toward others." In other words, there is no single path in life that results in violence.

The researchers found that among the best predictors of future violence was the Psychopathy Checklist-Revised (PCL-R), created by Dr. Robert Hare and his colleagues. However, not all potential offenders have been assessed with this instrument, so its utility is limited to those who have.