Contrary to popular belief and previous studies, people who have a serious mental illness and experience delusions are no more prone to violence than seriously mentally ill people who do not experience delusions.

A study in the April American Journal of Psychiatry, found no significant differences in the number of violent acts committed by these two groups of psychiatric patients.

Violence was defined as physical assaults resulting in injuries with or without a weapon, sexual assaults, and threats of violence made with a weapon, according to the article.

"This finding is significant because the public tends to perceive people with serious mental illnesses as more dangerous than the rest of the population, particularly if they have delusions. This was exemplified by the recently pulled ABC show ‘Wonderland,’ " said lead author and APA Vice President Paul Appelbaum, M.D., in an interview.

However, he cautioned that the results should not be interpreted to mean that individuals with delusional disorders are never violent. "We only have to remember Russell Weston, who shot and killed several people on Capitol Hill, and John Hinckley, who shot former president Ronald Reagan," said Appelbaum.

The other authors are Pamela Robbins of Policy Research Associates in Delmar, N.Y., and John Monahan, Ph.D., of the University of Virginia School of Law in Charlottesville.

The data used in the analysis were drawn from the MacArthur Violence Risk Assessment Study in which 1,136 patients were interviewed five times during a one-year period after discharge from a psychiatric unit or hospital between 1992 and 1996. The patients had a diagnosis of schizophrenia, schizophreniform disorder, schizoaffective disorder, depression, mania, brief reactive psychosis, delusional disorder, alcohol or drug abuse/dependence, or a personality disorder.

The authors found that delusional patients who took their medication according to their psychiatrists’ instructions were significantly less likely to commit acts of violence than delusional patients who were not compliant, said Appelbaum.

Although the authors did not study the reasons for that difference, Appelbaum hypothesized "that medication makes them more functional and less dangerous."

The authors also found no significant difference in the rate of violence between the delusional and nondelusional groups of subjects during the one-year follow-up. When the researchers compared the nondelusional subjects with those found to have persecutory and mind/body control subtypes of delusions, they were surprised to find that the subtype group had significantly lower rates of violence at the first 10-week evaluation, according to Appelbaum. However, the rates were similar at the following evaluations.

The mind/body control subtype is defined as a person’s belief that outside forces control his or her mind or body. The persecutory subtype is the belief that someone is trying to harm him or her, said Appelbaum.

Previous studies had shown that subjects with these two subtypes of delusions committed significantly more violent acts than the nondelusional subjects.

Appelbaum attributed the difference to two main factors: "Our study was prospective rather than retrospective, and our interviewers did not rely solely on patients’ self-reports of delusions"

Instead, they probed further to determine the accuracy of patients’ delusions using a questionnaire based primarily on delusion items from the Diagnostic Interview Schedule.

Appelbaum mentioned that one woman who had reported having mind/body control delusions was in fact angry at her husband for not paying child support on time. A man who believed he met the criteria for paranoia lived in a rough neighborhood and had seen drug dealers who he thought were watching him come and go.

"We found that patients most at risk for committing violence were angry, impulsive, or antisocial or had substance abuse problems," said Appelbaum.—C.L.