I was distressed to wake up this morning to coverage of another shooting, this time in a mall in Omaha. A teenager named Robert Hawkins went into the mall and shot 12 people, killing 8 thus far, and then shot himself. The scene resembled the Virginia Tech shooting in several regards, particularly because there is some indication that the shooter in both cases was mentally ill.
Robert Hawkins, who killed eight people then himself on a gun rampage through a Nebraska mall, had "lots of emotional problems", says the woman who took him in after he left home.
Debora Maruca Kovac, said he was withdrawn - "like a lost pound puppy that nobody wanted".
She said Hawkins, 19, was not on any medication for mental illness but was treated for depression in the past.
He had also recently broken up with his girlfriend, and then lost his job.
The Omaha World Herald newspaper reported that he had been arrested on Friday for being a minor in possession of alcohol.
The newspaper also said he was sent to jail for seven days in 2005 on charges of disorderly conduct.
My deepest sympathies go out to the victims of this brutal and arbitrary crime.
Clearly, there is a great deal to say about guns and violence in our society, but I want to address the issue of violence specifically in relation to the mentally ill. After the Virginia Tech shooting there was a great deal of talk about restricting access to firearms to the mentally ill as means of preventing violence of this nature.
I want to make clear the evidence shows that mental illness, and in particular depression, is at best a weak predictor of violent behavior.
Let me just clarify. I am not suggesting increasing or decreasing access to guns as a blanket principle. (To reveal my own bias, this is actually the subject about which I am the least libertarian. Guns strike me as by their nature in the business of violating the rights of others, and therefore I don't lose sleep about regulating them. However, I think that only blanket policies of gun restriction are likely to be effective, if anything is likely to be effective.)
The purpose of this post is not to get into an argument about whether access to guns should be restricted. The purpose of this post is to show that we should not stigmatize the mentally ill by suggesting that they are violence prone.
Let's look at the evidence.
Corrigan and Watson looked at data from the National Comorbidity Survey to determine the rates of violence in people with mental illness. They did indeed find that having anxiety or a mood disorder such as depression increased the likelihood of reporting violent behavior during the last 12 months by 3 to 4 times. Likewise people with drug abuse or bipolar disorder were about 8 times more likely to report violent behavior than the general population. Co-morbid drug or alcohol abuse increased the likelihood further.
However, when we compare the likelihood increases due to mental illness to the increases associated with the main predictors of violence -- male gender, ethinicity, and age -- mental illness is only a weak predictor.
These data are summarized below (Table 4 of Corrigan and Watson):
The left column shows the prevalence of violent behavior in various groups by self-report. The middle column shows the overall prevalence of those groups in our society. If we were to multiply these two percentages by a hypothetical population of 1 million, we number of people in the hypothetical population that we would expect to be violent based on which group they are in. This is show in the right column.
Now if we compare the number of people in the population who we expect to be violent because of mental disorders in aggregate we get about 19,000 or 1.9% (the authors point out that this is inflated because many individuals have more than one mental disorder). Then you can compare this to the number of people in the population we would expect to be violent because they are male (28,000 or 2.8%) or young (14,000 or 1.4%). What you see is that any particular mental disorder is much less of a predictor than being male or young. You have to combine all the mental disorders to get even close to the predictive capacity of age or gender.
Basically, this chart shows that while mental illness is a predictor of violence, when you consider that mental illness is rare it becomes only a weak predictor of violence in our society.
But I got more...
We have evidence that violent behavior follows a trajectory in the sense that violent acts are rarely perpetrated by individuals who have never been violent before. Previous violent behavior in childhood is a good predictor of violent behavior in adolescence and adulthood. Thus, a discussion about the predictors of violence in childhood is fair to add. (For more about risk assessment of violent acts in young people read this.)
With respect to risk of violent behavior in childhood, Blitstein et al. performed a longitudinal survey with inner city 7th graders from Minnesota. They compiled a list of odds ratios that reflect what demographic and personal characteristics make an individual more likely to commit violent acts. Below are reproduced Tables 2 and 3 from Blitstein et al:
If you look at the middle of Table 3, you find that depressive symptoms do confer an increased risk of violent behavior, but the increase is small and barely significant. We can compare that increase in risk to that associated with recent use of alcohol or weekly smoking and see that the effects of these are much larger.
Take home points: The evidence suggests that individuals with mental illness are more prone to violent behavior than other people. However, the size of that increased risk is relatively small, particularly when the size of the mentally ill population is taken into account. Furthermore, age, gender, and ethnicity are much better predictors than mental illness.
I posted this data two make two points.
1) We should not stigmatize the mentally ill with the tinge of violence because of the behavior of a few individuals. The vast majority of mentally ill people are just very sick. They are not naturally violent.
2) This information should inform our discussions of violence abatement via gun control. Some individuals have suggested that one means to limit gun violence would be to exclude the mentally ill from obtaining firearms. Assuming that it is possible to do so, I remain skeptical that such measures will be effective at reducing gun violence. Gun violence is endemic in our culture, and you have an all-or-nothing problem in dealing with it. You either limit guns for everyone or not at all because limiting them for particular groups is not likely to be effective.
Because mental illness is a weak predictor of violence, efforts to limit gun access to the mentally ill are not likely to be effective as gun violence abatement measures.
- Log in to post comments
We're going to see more and more of this bizarre behavior from our youth because most of them (and adults too) are so poorly nourished in this country that the brain is sersiously suffering in many ways. We're the riches in the world but we're the sickest. Dr. Weston A. Price tried to warn about this "breakdown" in our society clear back in the 1930's in his book "Nutrition and Physical Degeneration". The book is a national treasure and should be required reading of every MD and phychologist in the country! The world for that mattter. The science is simple, yet profound. We are a degenerating species. Thank God for the efforts of the Weston Price Foundation! www.westonaprice.org. Nothing to sell you, just answers to you and your child's future!
Mall shootings and school shootings are wrong, they destroy the peace and freedom we seek to create. Your opinion and statistical effort, Jake Young, say that we should not stigmatize the mentally ill. Your effort assumes there is and will be mentally ill people, mall and school shootings, and you attempt to change people's thinking about the people behind the bullets. Fine, good, think any thought you want while a loved one dies, okay? I've no argument with what any person thinks. However, allow me to point out to you; if you would put the same effort toward statistically comparing such shootings with the taking of psychotropic drugs, I think your conclusion might be helpful. Telling me what to think about the "mentally ill" doesn't help my life a bit! While understanding the causes of such disaster could be helpful. Have a nice day.
Mental illness is a bad predictor for violent behavior in general, but is it also a bad predictor for shooting sprees?
I think part of the reason that people feel mental illness may be a predictor of violence is the way the shootings are portrayed. I'm DEFINITELY not saying that they should not be covered/are not tragic, etc. but how many people are killed by guns wielded by those who do not have a mental illness? Just because it is not a "shooting spree" does not mean that the families of these people are not affected just as much as the families of VT or Omaha. The sprees just get more coverage because they do not happen everyday (though unfortunately they seem to be becoming more and more prevalent).
So yes, a blanket law restricting gun access would seem to be more beneficial than one targeting the mentally ill. And maybe something to combat the "semi-automatic weapons are OK even though you can buy a kit to make it fully automatic right after you purchase it" thing here in TX? Because I KNOW that you aren't using it for hunting. Except maybe people hunting.
School shootings and mall shootings are so rare that any attempt to pre-identify such perpetrators is going to have a very large false-positive result. Incarceration of such individuals may well turn them into mass murderers, rather than preventing it. Just as the war in Iraq has made more terrorists than it has destroyed.
A much better predictor of violence is childhood maltreatment, known as the "cycle of violence". People exposed to violence in utero, as infants, as children, as adolescents, as young adults and as adults tend to become more violent themselves. Is this a surprise? Isn't it well known that the best defense a good offense? Should we expect something different?
http://archpedi.ama-assn.org/cgi/content/full/156/8/824
Why then do so many demand "tough" leaders who are "tough" on crime, and who treat children with "tough love", and treat criminals harshly? Do we somehow expect that treating people harshly will somehow turn them into gentle creatures?
Isn't the definition of insanity doing the same thing over and over again and expecting different results? Who here is mentally ill? I think it is those who expect harsh treatment to "cure" or "prevent" violence. It never has, and there is no evidence that it ever will. What it does do is maintain the status quo, with tough leaders protecting people by being tough on potential perpetrators which only serves to make perpetrators faster than they can be locked up in prison and increasing the need for tougher and tougher leaders.
I agree. In my work as a psychiatrist, I have never noticed a strong correlation between depression/anxiety/even psychosis and violence. If you consider anti-social personality disorder, sure, but labeling the mentally ill as more likely to violence and/or disregarding the so called 'healthy' people with potential for violence might not be a good idea. The only incidence of violent behavior I've personally witnessed in a psychiatric facility was committed by a person who had all the other risk factors as well mental illness. Question..do this statistics include violence against self?
These shootings are not all connected to children who come from violent homes or have a history of violence. The one thing that seems to be a common thread is the fact that they have all been treated with a mind altering medication that has been proven to cause permanent brain shrinkage and damage. When will we learn that we have to stop pumping so many children up with pills?