Psychotic hallucinations, delusions rarely precede violence

Mass shootings at the hands of unhinged loners - such as those in Aurora, Colorado; Santa Barbara, California, and Newtown, Connecticut - perpetuate a commonly held belief that mental illness triggers violent crimes.

But a new study from the University of California, Berkeley, shows that hallucinations and delusions associated with psychiatric disorders seldom foreshadow acts of aggression.

In a painstaking review of 305 violent incidents in the United States, the researchers found that only 12 percent were preceded by psychosis. While numerous studies have found that brutality and bloodshed are more likely to be sparked by anger, access to firearms and substance abuse, this latest analysis is the first to look at the regularity of psychosis-induced violence among the mentally ill.

The results, recently reported in the online edition of the journal, Clinical Psychological Science, challenge the media-fueled stereotype of homicidal mayhem.

"High-profile mass shootings capture public attention and increase vigilance of people with mental illness. But our findings clearly show that psychosis rarely leads directly to violence," said study lead author Jennifer Skeem, a clinical psychologist and associate dean of research at UC Berkeley's School of Social Welfare.

Skeem and fellow researchers at the University of Virginia and Columbia University focused on the most violent patients tracked in the MacArthur Violence Risk Assessment study, a major 1998 analysis of more than 1,100 offenders who had been discharged from psychiatric facilities.

Specifically, the researchers looked at a subgroup of 100 high-risk patients, who had been involved in two or more violent incidents in the year after they were discharged from a psychiatric facility, to establish their mental states at the time they committed acts of violence.

"We wanted to examine the small group of people with repeated violence and see how consistently these violent incidents were caused by hallucinations and delusions," Skeem said.

In addition to reviewing records, they interviewed former patients about what they were thinking and feeling immediately before they engaged in violence, and sought the perspectives of their friends and family members. The results revealed that psychosis preceded only 12 percent of the violent acts they committed following their release. Moreover, while psychosis drove one violent incident, it was rarely implicated in subsequent ones, the study found.

The study defines violence as battery resulting in physical injury, sexual assault, and assaults or threats with a weapon. Mental illnesses ranged from schizophrenia and bipolar disorder to severe anxiety and depression.

While mass shootings account for a fraction of U.S. gun deaths, each one can influence public policy. For example, the 2014 shooting spree in Isla Vista near Santa Barbara, in which 24-year-old Elliot Rodgers killed six people, spurred the U.S. House of Representatives to pass an amendment to boost funding to add more mental health records to the nation's background check system for firearm purchases.