• info@criticascience.org

Controlling Gun Violence

What Might Work and What Probably Won’t

The horror of mass shootings, like the most recent ones in El Paso, Texas and Dayton, Ohio, predictably galvanize immediate calls for some sort of gun control action, followed almost as quickly by retreats from taking any actual action. But after a few days of reflection, objections to the various proposals for gun violence control seem to overcome the enthusiasm to act.

One of the problems we see in all of this is the proliferation of myths and misunderstandings about firearm-related violence and therefore about what might work to reduce it and what probably will have little to no effect.

The first thing we need to get straight is that most deaths from gunshots are suicides, not homicides.

In the United States there are about 40,000 firearm-related deaths every year, about two-thirds of which are suicides. Therefore, if we are really interested in saving the most lives possible, we need to put at least some of our focus on preventing suicides and not just the more dramatic but far less common mass shootings.

The horrific nature of mass shootings also obscures the fact that they account for a very small proportion of gun-related homicides. For instance, although every death from gun violence is horrible, there have been 273 gun-related deaths from mass shootings so far in 2019, while on average every month 52 women are shot to death by a domestic partner, or about 600 annually. Access to a gun makes a murder in a domestic setting five times more likely. Once again, then, we need to consider a wider range of gun-related violence when we think about enacting gun control regulations and laws.

Mass shootings rightly horrify and us and lead to calls for gun violence control laws. But they represent a very small fraction of gun-related deaths

(source: Shutterstock)

Three ideas for gun violence control have once again circulated prominently since the Dayton and El Paso shootings. They are:

  • Improved and increased background checks

  • Assault rifle bans

  • Red flag laws

We will consider each of these methods of gun violence control in turn from the vantage point of what the existing data tell us concerning three questions: would any reduce mass shootings? Would any decrease gun-related homicides? Would any prevent firearm-related intimate partner violence? We will consider inner city gun violence, for which we could find fewer studies, in a future commentary.

Do Background Checks Work?

Federal law requires that licensed gun dealers complete an automated background check before selling anyone a gun. That background check identifies people with a criminal record, fugitives from justice, people with final domestic violence court orders against them, and people who have been involuntarily committed to a hospital because of psychiatric illness or adjudicated as being mentally incompetent by a court.

It seems reasonable to disallow people with histories of committing violent crimes or violent domestic abuse to have guns. The mental health question, on the other hand, is more complicated. Because only a very small number of people who have even been diagnosed with a psychiatric illness get committed involuntarily to a hospital ward, few mentally ill people are actually identified by federal background checks. What if, however, we could identify every person who has ever received a diagnosis of anxiety, depression, schizophrenia, and so forth and keep them from owning a gun? It is conceivable that this would reduce the suicide rate because even though most people with psychiatric diagnoses do not kill themselves, about 90% of those who do have a diagnosable mental illness. We know that availability of guns is associated with increased risk for suicide, so restricting access to them should reduce the suicide rate.

But all of this is fanciful thinking because there is no way we will ever be able to screen for every mental health diagnoses as part of gun background checks. Fewer than half of people with a mental illness ever receive any treatment, meaning that many never receive a formal diagnosis of illnesses that might predispose to suicide that is recorded in any database. Even if we wanted all psychiatric diagnoses ever made to be available to a federal database, most people with mental illness would never be identified. So background checks are not going to be particularly helpful in reducing suicides.

Nor will they help to decrease mass shootings. Many studies have shown that people with mental illness do not commit the vast majority of gun homicides, including mass shootings. “Treating gun violence as primarily a mental health problem is a path to nowhere,” writes attorney Philip Rotner.  In fact, none of the proposed background check laws would have prevented most of the mass shootings committed in the U.S. 

There is evidence, however, that universal background checks before gun purchase can decrease the risk of gun-related domestic violence homicides. According to a review of the literature by the Rand group, about 10% of women who are murdered by domestic partners have already been the victims of domestic violence in the preceding month. The review concludes that “recent research evidence suggests that [background check] laws targeting domestic violence offenders may reduce homicide rates.”

In order to be effective, however, it is critical that all instances of domestic violence reported to law enforcement or adjudicated in courts throughout the country be reported to the federal database used for background checks. Furthermore, federal law requiring background checks be conducted by licensed dealers would need to be extended to include private dealers and gun shows as well in order for there to be a meaningful impact on intimate partner gun violence.

We conclude that strictly enforced and expanded background checks would have little effect on reducing rates of suicide or mass shootings but would decrease the number of firearm-related intimate partner murders.

Assault Rifle Bans

What exactly is an assault rifle or weapon? Here’s what Merriam-Webster has to say:

Definition of assault rifle: any of various intermediate-range, magazine-fed military rifles (such as the AK-47) that can be set for automatic or semiautomatic fire

also: a rifle that resembles a military assault rifle but is designed to allow only semiautomatic fire.

Between 1994 and 2004 there was a federal ban on the purchase of such guns and gun control advocates are insisting that the now lapsed ban be reinstated. Using a mass shooting database, a law professor and a law student from Stanford University found that the federal assault weapon ban “was associated with a 25 percent drop in gun massacres (from eight to six) and a 40 percent drop in fatalities (from 81 to 49)”. However, “in the decade after the ban [ended], there was a 347 percent increase in fatalities in gun massacres, even as overall violent crime continued downward.” A group of researchers from New York University came to the same conclusion using three different open source databases.     

Assault rifles are capable of killing multiple people in seconds and were banned from the U.S. between 1994 and 2004

(source: Shutterstock)


The two Stanford lawyers also point out that the number of mentally ill people in the U.S. did not drop between 1994 and 2004 and then suddenly surge in the decade thereafter, further reinforcing the point that mental illness is not responsible for most gun massacres. It seems fairly clear, therefore, that mass shootings could be reduced by reinstating the ban on assault rifles and the ammunition magazines used in them.

Extreme Risk Protection Orders

Fifteen states and Washington D.C. now have Extreme Risk Protection Order (ERPO) laws, popularly known as Red Flag Laws. Although the specifics vary from state to state, these generally allow family members and law enforcement officers to petition a court to order confiscation of firearms from someone believed to be at imminent risk of using them for suicide or homicide. The length of time that weapons can be withheld depend on individual circumstances and state regulations.

Studies of red flag laws in Connecticut and Indiana showed they decreased suicide rates. More recently, a well-publicized study published in the Annals of Internal Medicine attempted to see if they can prevent mass shootings as well. Because most mass shooters make public threats before carrying out homicides, it seems reasonable to hope that red flag laws might prevent at least some of them. Researchers from the UC Davis School of Medicine reported on a series of 21 cases in which California’s red flag law was used in an effort to prevent mass shootings. They found that “No mass shootings, other homicides, or suicides by persons subject to [red flag laws] were identified.”

The authors of the study clearly acknowledge that “it is impossible to know whether violence would have occurred had [ERPOs] not been issued, and we make no claim of a causal relationships.” They do point out, however, that “these cases suggest that this urgent, individualized intervention can play a role in efforts to prevent mass shootings.” Given the very sorry state of firearms research to date, it is definitely admirable that investigators have tried to conduct empirical studies of the effectiveness of red flag laws. We can conclude on the basis of what has been published so far that there is indeed evidence suggesting, although not proving, that red flag laws can reduce the risk for both firearm-related suicides and mass shootings. It does not seem to be known yet whether they are effective against more common forms of gun homicides, but both more red flag laws and red flag research seem warranted on the basis of the information we now have.

Gun advocates will correctly point out that all of the research studies we have cited have in common that they are observational and often involve very small sample sizes. No one is going to randomize large groups of people to have or not have assault rifles and see if there is a statistically significant difference in the rates of suicides and murders between the groups. We could, however, certainly have bigger and better research about gun violence were it not for the 1996 Congressional Dickey Amendment, which stated that "none of the funds made available for injury prevention and control at the Centers for Disease Control and Prevention (CDC) may be used to advocate or promote gun control.” This led to a virtual cessation of all federally funded gun research. As the editors of Scientific American recently noted, there is “a crying need to evaluate violence-prevention policies and programs based on data about individuals who participate in large randomized controlled trials". According to a Brookings Institute report, there is “strong public support for a range of measures to regulate the sale and possession of firearms”.

We think that current evidence supports continued pressure to enact universal background checks, assault weapon bans, and red flag laws. None of these, however, will stop all gun violence. To do that, we will need to fund more and better science to guide us to the best policies.

17 views0 comments