Predicting the Dangerousness of a Criminal

No discussion of risk assessment would be complete without at least considering those individuals suffering from a mental disorder. This group tends to suffer a high degree of victimization when it comes to violent crimes, such as assault and murder, as the popular conception of these offenders is that of an "axe wielding mad-man". This depiction often shows a crazed offender wandering the streets looking for victims in seemingly motiveless crimes. It should be emphasized however that the majority of mentally ill people are as lawful as any other group within the population.

An analysis of prime time American television dramas found that 17% involved depictions of mentally ill people, 73% of which were portrayed as violent [Montgomery]. This is compared to only 40% of "normal" television characters. If we were to use these popular conceptions as a yardstick, we would expect to see an increase in the numbers of mentally ill people committing crimes, though this is simply not the case.

The Australian Institute of Criminology (AIC) has undertaken a National Homicide Monitoring Program in which they record the details of homicides on some 77 variables. They are particularly interested in whether or not the offenders are diagnosed as mentally ill at the time of their offence. The results have found that in the period from 1st July 1989 to 30th June 1998 of 3314 offenders, only 4.4% (147 offenders in total) were so diagnosed at the time of the offence. A comparison of the overall population reveals that homicides committed by the mentally disordered occur significantly less frequently than among the general population (that is, those not diagnosed with a mental disorder). These figures would indicate that the stereotype of these offenders is highly in error.

While the number of mentally disordered people in the homicide figures appears to be relatively low, other figures indicate they may however be responsible for a certain type of crime at a certain stage in their illness. The following has been found about mental disorder and illness [Ward]:

• The prevalence of violence is more than five times higher among people diagnosed with a serious mental disorder
• The rates of violence among people with certain disorders (depression, schizophrenia etc) is very similar
• The prevalence of violence among people with a diagnosis of alcoholism is 12 times greater than for those who are not diagnosed as alcoholic, while those diagnosed as drug abusers will be 16 times more likely to be violent than those not meeting the diagnosis

What is important to note is that the presence of mental disorder is an accurate predictor at the time, but that it is not accurate when predicting future violence [Montgomery]. This may be owing to the fact that psychotic symptoms fluctuate and are not necessarily stable over time. Therefore, they may be worse at some times than others.

The characteristics of offences that mentally disordered individuals are committing also differ markedly from other offenders. The following is a summary of some of these differences (adapted from the AIC data):

• Location is more likely to be a private residence in the case of mentally disordered individuals (74.8% versus 61.6%)
• Females suffering form mental illness commit more homicides than their counterparts (29.5% versus 11.4%)
• The most likely victim is another family member (49% versus 13.8%) with other offenders (i.e. those not suffering a mental illness) killing more strangers (18.5% versus 10.2%).