The idea that a psychological profile can predict shootings — a suggestion made by President Donald Trump in the wake of two shootings in El Paso and Dayton, Ohio — is simply not true.
But it’s easy to see why both the public and U.S. politicians overplay the link between mental health and gun violence.
In 2017 and 2018, Americans lived through more than 50 mass attacks in public places, defined by the U.S. Secret Service as incidents in which at least three people were harmed. When the agency examined the circumstances behind the incidents, it found almost the same thing for both years: about two-thirds of the perpetrators had mental health symptoms prior to their attacks.
But here is another fact. Approximately 96 percent of violent crimes — including shootings — would likely still occur even if every suspect with a mental health condition was stopped before they carried out an attack.
Both findings can be true because while perpetrators of gun violence — including mass shooters — do show signs of psychiatric distress, the overwhelming majority of mental health patients will never commit a violent act in their lifetimes.
Four mental health experts who spoke with the PBS NewsHour described President Donald Trump’s conflation on Monday that “mental illness and hatred pull the trigger, not the gun” as “completely false” and “irresponsible.”
“Most of the research shows that people with mental illness are actually less likely than the general population to go on to shoot somebody else or to commit mass violence,” said Dr. Jonathan Metzl, a psychiatrist and director of the Center for Medicine, Health, and Society at Vanderbilt University in Tennessee. “To be honest, it’s quite frustrating as a psychiatrist to have this kind of false narrative be perpetuated because it’s a distraction from the story we should be telling.”
Here’s what we actually do know about the traits shared by mass shooters.
The myth about major mental disorders and crime
Metzl said we should start by banishing the idea that gun violence can be predicted by a psychological profile. It can’t.
Psychological profiles, by definition, are composed after violent actors have committed their crimes. In the realm of forensic psychology, such profiles are conclusions — not predictions of what might trigger the next one.
That’s because the risk factors connected to mass attacks are too non-specific. Metzl said even the characteristics most suited to building a mass shooter profile — “white male, angry, slightly paranoid, disaffected, isolated” — would match hundreds of thousands of people, the bulk of which will never go on to shoot others. It would be like looking for a sharp knife in a mountain of dull knives.
Not only would creating a dragnet based on mental health be inefficient, it would reinforce stigmas against the mental health community. Those biases would breed isolation and make those who are predisposed to violence more likely to commit an attack.
The general claim that psychiatric disorders are tied to gun violence is “a gross oversimplification,” said Jeffrey Swanson, a psychiatrist and behavioral scientist at Duke University. He and other scientists have known this for nearly three decades.
Swanson has explored the psychopathology of violent offenders since the late 1980s. (When I pinged my usual sources to ask who I should contact for this story, every single one suggested Swanson.) His research, along with that of colleagues in the field, is the basis for our modern discourse on guns and mental health.
In 1990, Swanson co-authored the first large epidemiologic study to measure the prevalence of minor or serious violent behavior among people with and without psychiatric disorders.
They wanted to know if violent acts — from shoving to shooting — correlated with diagnosable mental conditions. Part of their focus centered on so-called “serious mental illness” — such as depression, schizophrenia and bipolar disorder — which are often assumed to be connected with violence.
The researchers assessed about 10,000 people across three major cities (Baltimore, St. Louis and Los Angeles). They found 15 percent of people without mental illness had committed violence in their lifetimes versus 33 percent of those with serious mental illnesses, as defined above. On the surface, the takeaway might appear to be that psychiatric disorders equal more violence — but that’s only if you don’t consider the size of the two groups.
When you take these population sizes into account, according to Swanson’s research, persons without mental illness are three times more likely to commit violence in a given year than those with a mental health disorder.
“If we cured mental illness … tomorrow, which would be wonderful, our violence problem would go down by about 4 percent and the rest of it would still be with us,” Swanson said. These trends have been replicated over and over again since 1990 — in the U.S., Denmark, Finland and Australia.
Despite the body of evidence, politicians and news organizations continue to spread misinformation about the connection between mental health and violence.
Swanson said that when the public and media stories focus on “how ‘we need to fix the mental health system to fix violence,’ it reinforces this idea in the public mind that people with mental illnesses are dangerous, violent and to be feared.”
You can see this bias in public polling, whereby two-thirds of Americans said they would be “unwilling to have a person with serious mental illness as a neighbor.” Half of those surveyed viewed people with serious mental illness as “more dangerous than the general population.”
“People end up really disadvantaged and sometimes internalize these public attitudes,” Swanson said. “It makes recovery much more difficult.”
Gun violence absolutely has a psychological component, Swanson said, but mental illness is not the place you’d start to solve the problem.
In fact, you should abandon the idea that you can start with any one thing.
The mental (and social) traits that do correlate with gun violence
Metzl said there are many other factors that are strongly associated with shootings, including access to guns, a state’s gun laws, an attacker’s past history of violence, substance abuse, misogyny and racism, to name a few.
And while most mass shooters have a history of showing symptoms — emphasis on “symptoms” — of a mental illness, only about a quarter actually have a diagnosis of a mental illness. This rule also applies to those with personality disorders, said Michelle Galietta, a forensic psychologist at the John Jay College of Criminal Justice in New York City.
Galietta said when researchers looked at 1,100 patients recently discharged from psychiatric hospitals, they found factors like a history of violence, substance use and childhood trauma were more likely to predict a future gun attack than a past diagnosis of a mental illness.
“A person with a delusional disorder who stalks someone or thinks that they’re married to someone has much lower risk for violence than the domestic violence perpetrator,” Galietta said.
Galietta said only a few symptoms of mental illness — such as anger, impulsivity, emotional feelings of isolation — increase a person’s risk for carrying out violence.
And those basic hallmarks feature in an array of mental health conditions, from bipolar to anti-social disorder — or even in normal displays of emotion.
“You can be impulsive and angry — and those traits can be separate from having any disorder,” Galietta said. Even though most mass attackers have prior criminal histories, including ones of domestic violence, few would fit a state’s criteria for being involuntarily committed — something else Trump suggested this week.
The proposal that “mental health expertise can predict which one of the tens of thousands of people in mental health facilities is going to commit mass shootings — that’s a fantasy,” Metzel said.
And often, he said, the mental health narrative after mass shootings is tinged with racism.
“We tell the mental illness story often when the shooter is white. We want to look for an answer in an individual brain. But when the shooter is black or Muslim or Latino, then it’s a disorder of a culture,” Metzel said. Such disparities in perceptions can be witnessed in the absence of news coverage of Chicago, where shootings killed seven and injured 59 last weekend.
But Galietta, Metzl and Swanson all highlighted one unifying contributor to mass shootings — one that could be easily allayed with different policies.
“To say mental illness is the cause of gun violence or mass shootings is a dodge to avoid talking about guns,” Swanson said.
The common thread is guns
Swanson, Galietta and Metzl did agree with one of the president’s statements from Monday: “We must make sure that those judged to pose a grave risk to public safety do not have access to firearms.”
We’re always going to live in a world where people are inclined to harm one another. Most mammals, especially primates, intentionally injure other members of their species. While most people will never consider using a gun against another living being, violence is not aberrant behavior — it’s ordinary.
Swanson said if you look at America’s violent crime rates versus Australia, Canada and nations in Western Europe, our country is actually average or below average. That goes for serious assault and violence against women.
“We don’t have an exceptional violent crime problem in the United States compared to those countries,” Swanson said. But, “we do have an exceptional homicide problem because our homicide rate is several times higher than these other countries.
The one big difference is our relationship to firearms, Swanson added.
America has more guns than people, according to the 2017 Small Arms Survey, a project from the Graduate Institute of International and Development Studies in Geneva, Switzerland. There are 120 guns per 100 American civilians — a rate that’s double the No. 2 country on the list (the Falkland Islands) and quadruple that of westernized countries like Canada. By extension, a homicide in the U.S. is about three times more likely to involve a firearm than in those other westernized countries.
In the wake of the El Paso and Dayton shootings, many have called for an expansion of red flag laws — otherwise known as “extreme risk protection orders.” These regulations, currently enacted in 17 states and the District of Columbia, allow law enforcement or members of a household to obtain a temporary injunction that blocks a person’s access to firearms.
That’s partially because red flag laws have obvious gaps, especially around domestic violence. Domestic violence is one of the strongest predictors of future gun violence, but victims can only apply for a red flag petition if they’ve completed the process of obtaining a restraining order. Nearly half a million nonfatal domestic violence incidents go unreported every year. Moreover, only three states and the district allow individuals other than family to apply for an injunction.
Swanson said there is a failure in our natural surveillance systems when the people closest to would-be shooters don’t notice, downplay or ignore signs.
One sweeping characterization Swanson can make of mass shooters: “They’re marinating in hate towards other people.”
Editor’s note: This story was updated to clarify statements from Jeffrey Swanson about homicide rates.