It’s hard to explain (and fix) evil

Mass shootings and mental illness

In the aftermath of mass violence and the furor of the 24-hour news cycle, the words “mental illness” are often used to try to explain a perpetrator’s actions. In many cases, the assumption that mental illness had to be involved in these incidents drives the public debate about solutions.

This kind of coverage might raise awareness of how important mental health care is, but it is ultimately problematic because it too often it spreads inaccurate definitions of “mental illness,” it imposes further stigma on those who suffer from mental health conditions, and the limited scope solutions proposed will not achieve the intended results—to reduce mass shootings.

What is mental illness?

Mental illnesses are discrete and treatable health conditions involving distress or functional impairment related to thinking, emotion, or behavior. Examples include anxiety, depression, bipolar disorder, post-traumatic stress disorder, and schizophrenia.

There are also negative, antisocial thoughts, feelings, and actions associated with typical human functioning, such as anger, aggression, hate, envy, grievance, impulsivity, and reactivity. These human experiences, while in many cases undesirable and problematic for individuals, families, and communities, are unfortunately typical in a human life. While a range of strategies can be employed to counteract these problematic patterns – parenting, education, faith practices, some types of psychotherapy, policing, and the justice system – these thoughts, feelings, and behaviors are not illnesses, or even necessarily indicative of an illness. Antisocial behavior, for example, is not on its own sufficient to warrant diagnosis of antisocial personality disorder, which involves an ingrained pattern of predatory behavior and lack of conscience that is very difficult in most cases to treat successfully, but which does constitute an illness.

Mental illness and mass violence

Although incidents of mass murder are devastating and bring long-term individual- and community-level trauma, mass shootings account for <1% of firearm injuries in the U.S. Thus, it is difficult to use statistical methods to confidently and precisely know how often mental illness is a primary driver of mass murder.

Some research has shown that if we line up mass shooters in the U.S., 15-20% had a diagnosed mental illness before the event. Studies using a broader definition of mental illness, like statements by law enforcement or family after the incident, found rates of mental illness range from 30-60%. But there are two problems with these latter statistics:

1. This information is gathered after the fact and, as Skeem and Mulvey eloquently said, involves circular thinking: “‘Why did this man do this terrible thing?’ Because he is mentally ill. ‘And how do you know he is mentally ill?’ Because he did this terrible thing.”

2. Among the general public, nearly 50% of people will experience symptoms of a mental illness at some point in their life, and among those, only 50% will ever be diagnosed. So we don’t know whether these rates simply reflect rates in the general population. Correlation does not equal causation.

Another factor, often not discussed, is that many people who commit mass murder spend considerable time planning their violent rampages. While their motives are deeply disturbed, their cognitive capacities to plan and logically order their actions are more intact than is usually the case in people with serious mental illnesses.

Mental illness and violence in general

If we cast a wider net and examine violence more broadly, evidence suggests mental illness does not cause violence. Large epidemiological studies have shown that rates of violence among people with mild-to-moderate mental illnesses range from 2%-4%, compared to 1%-3% in the general population. One of the strongest longitudinal studies, called the MacArthur Violence Risk Assessment Study, found that only 1% of patients discharged from psychiatric facilities committed an act of violence against a stranger with a gun.

Stronger associations emerge between having a mental illness and victimization. One study found individuals with mental illness were three times more likely to be a victim than a perpetrator of violence. Those with mood disorders, such as major depression, are not more likely to hurt others, though they are more likely to harm themselves. Firearms are used in 50% of suicides and account for 3/5 of all gun deaths in the U.S. People with anxiety disorders are less likely to harm others.

Some studies have shown severe mental illness predicts violence, even after accounting for substance use/abuse. Specifically, people with psychosis—very severe symptoms such as hallucinations (hearing or seeing things that are not there) and delusions (false and sometimes bizarre beliefs)— are 15 times more likely to commit homicide. However, it’s uncommon for symptoms of psychosis to immediately precede violent acts. This also doesn’t explain all mass shootings, as psychosis only plays a major role in 11% of mass shootings.

Other factors?

So not all mass shooters have mental illnesses. And the overwhelming majority of people with mental illness will not commit mass murder (or violence in general). This suggests that other factors are more important in predicting mass violence. Experts believe the following factors, especially in combination, are more predictive:

1. Particular motivations, such as revenge or envy. A type of mass murderer identified by some experts is the “pseudocommando,” who “kills in public during the daytime, plans [the] offense well in advance, . . . comes prepared with a powerful arsenal of weapons,” expects to die during the massacre, and is driven by intense anger, resentment, and revenge.

2. Adoption of extremist beliefs that promote the use of violence to attain one’s goals. There is a growing concern in the U.S. about the rise of extremist groups and influences and their relationship to mass killings.

3. Social isolation. When people with these and other characteristics that put them at risk for violence become socially isolated, the combination can place them at even greater risk.

Whether one terms these characteristics dysfunction or evil, such behaviors do not constitute a diagnosable and treatable health condition.

A research group studying mass shootings for decades (called The Violence Project) concluded that mass shootings are largely the results of a constellation of behaviors involving a buildup of childhood trauma, an identifiable crisis point (separate from psychosis), the need to blame someone, and the opportunity to conduct a mass shooting (i.e. access to firearms). Blaming mental illness entirely “conceals it more than it reveals it.”

Among mass school shootings, in particular, the U.S. Secret Service found a similar theme. While they reported that most teen perpetrators had symptoms of mental illness, few had a psychotic illness and nearly all had histories of severe bullying, social isolation, school discipline, and adverse childhood events, like abuse, substance use in the home, parental incarceration, or parental mental health problems.

Misguided solutions

Mass shootings consistently prompt calls to reform the process for assessing and treating people with mental illnesses include changes to legal rights (e.g., access to firearms) as well as building more psychiatric hospitals and committing greater numbers of people with mental illnesses to them. There is a belief that a properly structured mental health system could successfully identify and intercept people who commit mass murder before they are able to act.

We really need to address mental health and our mental health system in the U.S. Improving it will certainly play a role in saving lives in general and reducing deaths of despair. It will also help with the direct aftermath of mass shootings, including community-level trauma. But, this approach has limited utility in directly preventing mass shootings. Mass murderers are so rare that identifying them before they act is like trying to find “a needle in a haystack,” even once we improve our mental health system. Characteristics of mass murderers are shared by millions more who never commit such tragic acts. We need to be immensely clear, data-driven, and evidence-based in how we change mental health in the U.S. and for what intended purpose.

Mental illness, and even a broken mental health system, is not unique to the U.S. But mass shootings are. We need to have more difficult conversations about complex social problems that involve behavioral, cultural, legal, and political aspects as well as psychological factors. The life path of a mass shooter is peppered with points in which we, as a society, could have intervened and mitigated risk.

Bottom line

It’s hard and complex to explain evil. Focusing solely on mental illness as a way to make sense of mass shootings largely avoids confronting the real, more complicated causes.

Love, YLE

Much of this post was written during my “day job” with brilliant colleagues at the Meadows Mental Health Policy Institute.

---

If you want to read more about these complex issues, I highly suggest these two books. (I don’t know the authors personally and gain nothing from recommending, just really enjoyed reading them.)

• The Violence Project: How to Stop a Mass Shooting Epidemic (Amazon)

• Healing: Our Path from Mental Illness to Mental Health (Amazon)

In case you missed it, previous YLE posts on this topic include:

• We can reduce gun violence in the U.S.

• Epidemiology of Mass Shootings

---

“Your Local Epidemiologist (YLE)” is written by Dr. Katelyn Jetelina, MPH PhD—an epidemiologist, biostatistician, wife, and mom of two little girls. During the day she works at a nonpartisan health policy think tank, and at night she writes this newsletter. Her main goal is to “translate” the ever-evolving public health science so that people will be well equipped to make evidence-based decisions. This newsletter is free thanks to the generous support of fellow YLE community members. To support the effort, please subscribe here: