Limitations of Study on Violent Crime and Mental Illness

To the Editor: I read with interest the Swedish study by Seena Fazel, M.B.Ch.B., M.R.C.Psych., M.D. and Martin Grann, C.Psych., Ph.D. that reported 6.6 % of individuals with severe mental illness were convicted of violent crimes, compared with 1.8% of the general population (1) . The authors discussed several limitations of their study, but did not mention factors that selectively contribute to higher rates of criminal convictions among the mentally ill.

At every phase of the criminal-justice process, there are selective factors that determine who are apprehended, arrested, and convicted of criminal offenses. For example, individuals with mental illnesses may not be as adept in eluding apprehension as those without mental disorders. I recall a patient with a severe mental illness who was convicted of bank robbery after pretending to have a weapon and demanding money from a teller. Afterwards, he sat on a bench outside the bank until the police arrived.

Teplin reported that attitudes of arresting officers toward the mentally ill can result in arrest rates that are significantly higher than those for non-mentally ill offenders (2 , 3) . Following arrest, selective sampling continues regarding who is convicted of a crime. Individuals with mental illnesses may not present themselves well in court and may not have access to the best defenses.

The previous comments are based on the U.S. criminal justice system. I assume similar factors affect cohort selection in Sweden and other countries.

Drs. Fazel and Grann also hypothesize that violent crime in their study would have been reduced by 5.2% if the mentally ill who committed these crimes had been “institutionalized indefinitely” ( 1 , p.1400). However, violent behavior does not necessarily stop when individuals are hospitalized or incarcerated. Institutionalization just limits the victims of violent crimes to members of hospital and prison communities.

The study by Drs. Fazel and Grann provides useful information regarding mental illness and violence. However, because of the many selective factors affecting the mentally ill throughout the criminal process, I do not believe any conclusions can be drawn from it regarding a causative relationship between mental illness and violent crime.