Unpacking the Relationship Between Mental Illness and Violence
There is an “intricate link” between violence and mental illness, and perceptions of this relationship are colored by misconceptions, stigma, and stereotypes. Understanding the difficulties related to studying violence among people with mental illness provides a nuanced view of the data that can be applied to policymaking.
Unexpected or inexplicable events sometimes elicit responses like “That’s crazy,” or “That’s insane.” Although these terms are used less frequently today to describe people with mental illness, their continued association with erratic or unfathomable actions and events exposes a persistent connection between mental illness and unpredictable behavior. People behaving unpredictably makes others feel less safe, even if the situation itself is not inherently unsafe. The lingering association between mental illness and inexplicable behavior is particularly clear when a highly publicized violent act occurs, as speculation about the perpetrator’s mental health often follows.
Sorting out the real relationship between mental illness and violence requires an understanding of assessment difficulties, public perception, and other risk factors.
Violence and Mental Illness: Assessment Challenges
Assessing the relationship between violence and mental illness is not a straightforward process. Although researchers have attempted to measure how frequently people with mental illness behave violently—as well as the proportion of violence in society that is attributable to mental illness—results are mixed. While some studies have found that people with mental illness are slightly more prone to violence, others have observed no difference between violent acts committed by people with mental illness and people without. These differences are attributable to methodology.
The two studies frequently cited as the most methodologically sound found that people with mental illness are somewhat more likely to commit violent acts compared to people without mental illness. One study measured minor or serious violence committed by people with schizophrenia, bipolar disorder, or major depressive disorder, excluding those with a co-occurring substance use disorder (SUD), over a 12-month period. Seven percent of people with at least one of those three mental illnesses and no SUD committed a violent act, compared to 2 percent of people with neither mental illness nor SUD. This study also concluded that only 4 percent of all violence committed in one year is attributable to these three mental illnesses alone. A different study found that 2.9 percent of people with serious mental illness committed violent acts in a year’s time, whereas only 0.8 percent of people without serious mental illness or SUD did. Many things likely contribute to the different findings in the larger body of research, including different definitions of mental illness as well as populations studied and study design.
What qualifies as mental illness?
Researchers can define “mental illness” in many ways. Studies that assess violence committed by people with mental illness sometimes limit their definition to “serious mental illness,” which includes disorders that substantially impair daily functioning and interfere with or limit at least one major life activity. Schizophrenia, bipolar disorder, and major depressive disorder are frequently categorized as serious mental illnesses; however, other disorders can also fit the definition. Other studies take a narrower scope, limiting their definition of mental illness to specific conditions (e.g., schizophrenia, schizophrenia spectrum disorders) generally associated with higher odds violent behavior. Even though definitions of mental illness have specific meanings, the nature of mental illness makes it difficult to assess its connection to violence.
Diagnosing mental illness is another challenge. One contributing element is the fact that symptoms can vary over time. For example, major depressive disorder can be episodic or persistent, responses to medications can change over time, and milder symptoms may not be recognized prior to formal diagnosis. Missed, delayed, or wrong diagnoses are common in mental healthcare due to variation and/or overlap in symptoms. This means a person may collect multiple diagnoses at different times that might not represent their actual mental illness—making it difficult to identify the correct population for assessing violent conduct.
Why does study design matter?
Studies sometimes use populations that are convenient rather than more representative of the public. In the case of mental illness and violence research, some studies have measured violence in specific clinical settings (e.g., inpatient care) or in criminal justice settings. This choice means the findings from these studies may not generalize to people with mental illness living in the community.
Another element of study design is whether researchers account for other variables that affect the relationship between the two variables of interest. This is known as controlling for confounding, and not all studies control for the same variables. Many variables confound the relationship between mental illness and violence because they also predict violent behavior. For example, a history of violence often predicts future violence—though not all studies address this association in their methods.
Other methodological considerations that challenge researchers measuring the association between violence and mental illness involve the context of the violent act.
One contextual factor studies can miss is whether a violent act was committed in retaliation against physical provocation by someone else—an important factor, as people with mental illness are frequently victims of violence and abuse. Although violence of any kind is unacceptable, people often fear unprovoked violence by strangers more than other types of violence.
Public Perception of Mental Illness
Despite increased awareness about mental health, mental illness—especially severe mental illness—is still highly stigmatized. Studies show that many people perceive individuals with mental illness as dangerous and desire social distance from them. While many factors contribute to these perceptions, the majority of the public has no personal experience with serious mental illness. So, when media coverage associates mental illness with violent actions, it reinforces misconceptions. Studies show that media coverage frequently links mental illness with dangerousness, violence, and crime—a bias that can increase social stigma and decrease support for public policies that would benefit people with mental illness (e.g., insurance parity, expanded funding for treatment resources). Another challenge with media coverage of violent events is that reporting can conflate non-pathological poor mental health with diagnosable mental illness. Mental health issues affect most people at some point in their lives, even if it doesn’t rise to the level of a mental illness; however, the media and the public might not make that distinction.
The public discourse also tends to focus on violence between strangers despite the fact that most violence occurs between family members and acquaintances. The Centers for Disease Control reported that 14.6 percent of all 2022 homicides in which the perpetrator is known were committed by strangers. Bureau of Justice Statistics numbers for 2023 show that strangers perpetrated 19.2 percent of all homicides, and Federal Bureau of Investigation numbers for 2024 put the percentage for stranger-perpetrated homicides among all homicides at 16.5 percent. Researchers have observed similar patterns among people with mental illness who commit violent acts. When people with serious mental illness do commit violent acts, it is most frequently toward family members rather than strangers. One study of patients discharged from civil inpatient facilities found that just 6 percent had committed a violent act against a stranger. Furthermore, even though violent actions are slightly more prevalent among people with mental illness, these individuals are much more likely to be victims than perpetrators.
Other Risk Factors for Violence
Although research suggests a slightly increased propensity to commit violent acts among people with mental illness, estimates suggest that most violence is not the result of mental illness.
Calculations from one large study imply that 96 percent of violence would still occur even if the increased risk were reduced to the average risk of a person without mental illness. One reason for this is that many of the social and individual factors associated with violence among people with mental illness are the same as for people without mental illness.
Violent behavior is hard to predict, but many social and psychological factors are associated with an increased likelihood of violence. Being young, male, or single is associated with higher likelihood of committing a violent act, as are a history of physical abuse, growing up with an incarcerated parent, experiencing childhood trauma, and having limited social support. (These risk factors are the same among people with and without a mental illness.) Living in poverty and being unemployed or financially strained are also associated with violent acts.
Substance use disorder is another condition associated with violent behavior among people with and without mental illness. Even if it doesn’t qualify as a disorder, alcohol use is strongly associated with violent behavior. And because alcohol is legal and widely available, it accounts for a significant portion of violence associated with substance use. One study of women offenders housed in a forensic treatment facility found that 70 percent of patients with alcohol use disorder had committed a violent crime, compared to 19 percent of patients with another substance use disorder. Another large study found that alcohol use disorder and cocaine use disorder were most strongly associated with intimate partner violence, whereas people with cannabis use disorder or opioid use disorder were more likely to be victims of intimate partner violence. Studies have also identified methamphetamine use as a risk factor for perpetrating violence, although the relationship is complex and many of the studies have methodological limitations. Other studies have shown that methamphetamine and heroin use are more strongly associated with property crime than violent crime.
Conclusion
There is an “intricate link” between violence and mental illness, and perceptions of this relationship are colored by misconceptions, stigma, and stereotypes. Understanding the difficulties related to studying violence among people with mental illness provides a nuanced view of the data that can be applied to policymaking. Additionally, recognizing that there are multiple risk factors for committing violent acts, many of which apply to people both with and without mental illness, limits the tendency to latch onto any suggestion that the perpetrator was mentally ill when violent incidents do occur.
