01/05/2014 06:38 pm ET Updated Mar 07, 2014

Media Coverage of Mental Illness and Violence

Are those with a serious mental illness more likely to be violent than those who are not mentally ill?

That depends on whether or not the seriously mentally ill are in treatment, whether or not the definition of violence includes being the victim of a crime, how the questions are asked by interviewers and how they are interpreted by subjects who may be in a psychotic state.

On Friday, Jan. 3, I spoke by telephone with Professor Jeffrey Swanson, a leading epidemiological researcher on mental illness at Duke University Medical School. Professor Swanson's 1990 study, which was cited in a front-page Dec. 22 article in the New York Times by reporters Michael Luo and Mike McIntire, indicated that 33% of those with a serious mental illness but no substance abuse problems reported violent behavior in the past. By contrast, only 15% of those with no diagnosable disorder reported past violent behavior.

Even if we leave aside the fact that the statistics were obtained from interviews in the 1980s and might need an update, there are a few caveats to the numbers mentioned in the Times' article. As Professor Swanson told me over the phone, the 33% figure represents acts of violence "over the course of a lifetime." When the subjects of the study were asked whether or not they had engaged in violent behavior in the past year, only 7% of those with a serious mental illness indicated that they had been violent, as opposed to 2% to 3% of those with no diagnosis.

In addition, the study was randomized and as a result included a "mix of people in the community," said Professor Swanson. What that means is that some of the subjects who met the criteria for serious mental illness were in treatment and "some were not," as Swanson said.

When we spoke over the phone, Professor Swanson agreed that the seriously mentally ill who are in treatment tend to be no more of a threat to others than those who have no diagnosable disorder. Swanson mentioned other "compounding" factors that make someone with a serious mental illness more likely to be violent: a history of past violence, substance abuse and the incidence of violence in the neighborhood in which the subject lives.

When I mentioned to Swanson that those with a serious mental illness are more likely to be the victims than the perpetrators of violent crime, he did not dispute that and noted that his 1990 study did not break down whether or not the subjects had been victimized.

I can say from personal experience that it also matters how an interviewer asks a question. When I checked into the UCLA psych ward in January 1999, my late psychiatrist, Dr. Michael McGrail, about whom I wrote a glowing tribute in the L.A. Times after he passed away, and his colleagues asked me if I was a threat to myself. I said yes.

Then they asked me if I was a threat to anyone else. I, who had been off my medication for roughly a week and who was in the throes of psychosis, told them that I thought I might have to defend myself given that I feared that I might be assassinated.

Based on my response, they unfortunately wrote down that I was a threat to others, and they held me involuntarily for 72 hours, even though I had never been violent in my life and I had walked more than 20 miles to get to UCLA. No one had to force me to enter the psych ward. I did so of my own volition.

I mention all of this because it shows that those with a serious mental illness, particularly a psychotic disorder, are not only more likely than others to misread a situation and possibly get into a violent confrontation; they are also more likely to be misread by those interviewing them.

This leads me to conclude that the New York Times' recent, front-page, lead article on mental illness and guns, followed by its editorial a week later, could have benefited from a more nuanced and deeper understanding of the issues that have an impact on the studies involving those with a serious mental illness.

For the record, I agree with President Obama's recent executive actions on gun control, and I agree with the overall tenor of the Times' coverage, although I cannot agree that those who have been mandated to receive inpatient or outpatient care should be characterized as "obviously risky citizens," as a Times' Jan. 4 editorial, "Stopping Mentally Ill Gun Buyers," proclaimed. I was never an obviously risky citizen even though I was involuntarily hospitalized.

While I have argued in this space numerous times that those with a serious mental illness should be barred from owning firearms, the reason is not because the seriously mentally ill are necessarily more of a threat to commit a homicide. It is because those with a serious mental disorder, in particular men, are much more likely to take their own lives if they have a gun.