Last week, I focused on the irrefutable fact that most people who are murdered in the United States are killed by someone using a handgun. Predictably the cranks came out of the woodwork to criticize my "foolish drive" to disarm everyone … including the police.

For those who paid attention, I advocated neither position, particularly the latter. When the facts don’t support your position, you impugn the messenger — hopefully not with a handgun.

Along these same lines, I recently read several articles on the links between mental illness, substance abuse and violence. In 2006, Dr. Jeffery Swanson and his colleagues published a terribly flawed study (the CATIE Study) about the relationship between mental illness and proclivities to violence.

People who want to mislead you about the role of mental illness in violent crime like to cherry pick that research by citing things like, "The CATIE violence study found that patients with schizophrenia were 10 times more likely to engage in violent behavior than the general public (19.1 percent vs. 2 percent in the general population)."

It’s worthy to note that the researchers had baseline violence data on less than half (42 percent) of their subjects. Moreover, 36 percent of the study participants also had a substance abuse issue at the time of their violence.

The substance abuse issue is particularly material to the present discussion because it creates what scientists call a confounding variable. A common definition of a confounding variable is one that creates interference so as to distort the association being studied between two other variables —- because of a strong relationship with both of the other variables. In other words, you can’t parse out how much of the "cause" of the violence was mental illness and how much was substance abuse —- of which alcohol is the most common.

The upshot of the Swanson study was to observe a 1.6 percent difference between valid study subjects and a comparison group from a similar study 30 years earlier. As such, people who want to shift the debate on gun control to restrictions for persons with mental illness grotesquely distort the connection between mental illness and violence. Not that we shouldn’t consider all relevant variables, just that this one wouldn’t have prevented the vast majority of murders.

Understanding that people with mental illness are no more or less likely than the general population to commit violent crime, whither shall we look for the boogeyman? As previously mentioned, alcohol appears to play some role in predicting violence. Here too, though it’s just not that simple.

A highly regarded study by Thor Norström & Hilde Pape published in the journal Addiction demonstrates the complexity of it: "Little is known about the mechanisms underlying the alcohol-violence relationship, but complex interactional processes including multiple causes and contributing factors are likely to be involved."

Along with this they observe two important dynamics. First they state, "changes in heavy episodic drinking were associated with changes in violent behavior…"

Second they find, "The effect of drinking is confined to individuals who tend to withhold their angry feelings — a finding that is particularly relevant to the alcohol disinhibition hypothesis."

Many times as a cop I answered a domestic dispute call only to be greeted with the words, "It’s OK. He only gets like this when he’s drunk."

Perhaps, as Norström and Pape suggest, "he" has pent up anger that is only released when intoxicated — not that this in any way absolves him of the violence!

More importantly, this says something significant about violence in the vast American underclass whose existence largely has been predicated on pent up feelings of injustice, persecution and anger. To clarify, I’m just talking about economic class here, not race — but again, even this wanders into those ol’ confounding variables.

Just so the cranks don’t misunderstand, I’m not advocating prohibition as a way to curb violence — in either the general population or cops. Clearly though, we’re not looking at the whole picture.


Matthew Pate is a former law enforcement executive who holds a doctorate in criminal justice from the University of Albany and who has advised police agencies around the country. He writes from Pine Bluff, Ark. Contact him at