Mental Illness and Violence Generally

The term “mental illness” encompasses a wide range of conditions and diagnoses. Researchers have concluded that there is little if any relationship in general between having a mental health diagnosis and being violent towards others.³ Most research has focused on “serious mental illnesses,” defined as conditions that can cause substantial disruptions in functioning and the ability to perform major life activities. Examples include schizophrenia, bipolar disorder, major depression, and other conditions that cause serious disruptions in functioning.⁴

Two definitive studies were conducted across multiple sites in the 1990s. The first study, the National Institute of Mental Health's Epidemiologic Catchment Area (ECA) study revealed that approximately 4 percent of all violence in the U.S. was attributable to serious mental illness. It should be noted that “violence” was defined broadly in the ECA study to include a broad range of physically assaultive behaviors, including hitting with a fist, pushing, shoving, or throwing things, in addition to more serious acts such as attacking or threatening to harm another person with a weapon.^{Reference Swanson, McGinty, Fazel and Mays5}

A second major study conducted in the mid-1990s focused on individuals released from psychiatric hospitals and followed for one-year post-discharge. Significantly, the study showed that substance abuse significantly increased risks of violence among persons with and without mental illnesses.^{Reference Steadman and Mulvey6} After controlling for substance abuse, mental illness was not significantly associated with violence.

More recent studies have similarly shown low overall rates of violence among people with mental illness as a class. However, rates of violence do increase among certain subgroups, including some individuals experiencing symptoms of psychosis such as delusions and hallucinations, particularly when substance abuse is also a factor.^{Reference Fazel and Gulati7}

Few studies have directly examined whether people with co-occurring mental illness and substance use disorders are more prone to violence than people with substance use disorders who do not have mental illness. A meta-analysis conducted by a team of researchers at Yale Medical School revealed that risks of violence among substance abusers was particularly high in people diagnosed with psychotic disorders such as schizophrenia or schizoaffective disorder.^{Reference Duke, Smith and Vitelli8}

Other factors, such as a history of violence and anti-social behavior, past victimization or trauma, youth, and male gender may also have an impact on rates of violence among people with serious mental illness (as well as among people who don't have serious mental illness). Socio-economic status and a person's living conditions and circumstances may also be a factor. Disproportionate numbers of people with serious mental illness live in economically disadvantaged neighborhoods with high rates of alcohol and drug abuse and violence. When these factors are controlled for, mental illness alone does not pose a significant risk for increased violence.⁹

Mental Illness and Mass Violence

Mass shootings, defined as acts in which 4 or more people are killed, constitute less than one percent of all gun violence in the U.S., yet garner much of the media attention and have a profound effect on public perceptions about gun violence.¹⁰ Mass shootings understandably create shock and in their aftermath often generate speculation that the shooters must have had a mental illness.

Since a number of those who committed mass violence in recent years have died during their attacks, information about their psychiatric histories is frequently based on prior reports or behaviors rather than evaluations after the tragedies. Nevertheless, researchers studying the characteristics of mass shooters in recent years have connected serious mental illness to between 23 percent^{Reference Stone11} and 40 percent¹² of these tragedies, a significantly higher percentage than with acts of gun violence generally.

While these percentages are quite high, it is important to consider them in context. According to Everytown for Gun Safety, deaths resulting from mass shootings constitute less than one percent of all gun deaths in the U.S. each year.¹³ This is not to suggest that efforts to identify individuals who pose risks for mass gun violence or to take steps to prevent such violence from occurring should not be undertaken. Extreme Risk Protection Orders, which are discussed later in this paper, may be one effective strategy for doing so. But, considering the low percentage of mass shootings relative to shootings in general, focusing on mass violence data to draw broad conclusions about overall threats posed by people with serious mental illness is neither fair nor an effective strategy for broadly reducing gun violence.

Gun Violence and Suicides

Although gun homicides generate more attention in the media, there are actually significantly far more suicides with guns than murders. In 2017, 60 percent of gun related deaths in the U.S. were suicides (22,854) whereas 37 percent were homicides (14,542). And, suicide rates have been increasing in recent years.^{Reference Hedegaard, Curtin and Warner14} Suicides were the second leading cause of death in 2016 for young people between the ages of 16 and 24.¹⁵

The epidemic of suicides has been particularly severe among veterans. In 2016, the U.S. Department of Veterans Affairs reported that about 20 veterans per day take their own lives. In that same year, veterans accounted for 14% of all suicide deaths in the U.S. Rises in suicide rates among young veterans between the ages of 18 and 34 have been particularly prevalent between 2006 and 2016.^{Reference Shane16}

Nearly half of all suicide deaths in the U.S. are with firearms, accounting for 60% of all gun deaths each year.^{Reference Gramlich17} Ninety percent of suicide attempts with guns result in deaths. Others result in serious disability.^{Reference Pallin and Spitzer18} By contrast, only one in twelve suicide attempts overall in the U.S. are lethal.^{Reference Carroll19}

Mental illness appears to be associated with a significant percentage of suicide deaths. The Centers for Disease Control (CDC) recently estimated that about half of all suicide deaths involved people diagnosed with mental illness, and they speculated that the actual rates may have been significantly higher because many others may have had mental health conditions but had not seen a mental health professional and thus were undiagnosed.²⁰

The statistics concerning suicides suggest that national conversations about reducing gun violence should focus as much on reducing gun related suicides as they focus on homicides.