Skip to main content
Research and Innovation

Researchers ID Risk Factors That Predict Violence in Adults With Mental Illness

For Immediate Release

Researchers have identified three risk factors that make adults with mental illness more likely to engage in violent behavior. The findings give mental health professionals and others working with adults with mental illness a suite of characteristics they can use as potential warning signs, allowing them to intervene and hopefully prevent violent behavior.

“Our earlier work found that adults with mental illness are more likely to be victims of violence than perpetrators – and that is especially relevant to this new study,” says Sarah Desmarais, an associate professor of psychology at North Carolina State University and co-author of a paper describing the work. “One of the new findings is that people with mental illness who have been victims of violence in the past six months are more likely to engage in future violent behavior themselves.”

The researchers compiled a database of 4,480 adults with mental illnesses – including schizophrenia, bipolar disorder and depression – who had answered questions about both committing violence and being victims of violence in the previous six months. The database drew from five earlier studies that focused on issues ranging from antipsychotic medications to treatment approaches. Those studies had different research goals, but all asked identical questions related to violence and victimization.

The researchers assessed the data to determine which behaviors, events and characteristics were most predictive of violent behavior over a six-month period. Violent behavior, in this context, ranged from pushing and shoving to sexual assault and assault with a deadly weapon.

The researchers found three risk factors that were predictive of violent behavior: if an individual is currently using alcohol; if an individual has engaged in violent behavior over the past six months; and if an individual has been a victim of violence within the past six months.

“We found that these risk factors were predictive even when we accounted for age, sex, race, mental illness diagnosis and other clinical characteristics,” Desmarais says.

In contrast, the researchers found that current drug use was not predictive of violent behavior, when age, sex, race, mental illness diagnosis and other clinical characteristics were considered.

“This is useful information for anyone working in a clinical setting,” Desmarais says. “But it also highlights the importance of creating policies that can help protect people with mental illness from being victimized. It’s not only the right thing to do, but it makes for safer communities.”

The paper, “Proximal Risk Factors for Short-Term Community Violence Among Adults with Mental Illnesses,” is published online in the journal Psychiatric Services. Lead author of the paper is Kiersten Johnson, a Ph.D. student at NC State. Co-authors were Kevin Grimm of the University of Arizona; Stephen Tueller and Richard Van Dorn of RTI International; and Marvin Swartz of Duke University. The work was supported by the National Institute for Mental Health under grant number R01MH093426 to Van Dorn.


Note to Editors: The study abstract follows.

“Proximal Risk Factors for Short-Term Community Violence Among Adults with Mental Illnesses”

Authors: Kiersten Johnson and Sarah Desmarais, North Carolina State University; Kevin Grimm, University of Arizona; Stephen Tueller and Richard Van Dorn, RTI International; Marvin Swartz, Duke University

Published: Feb. 29, Psychiatric Services

DOI: 10.1176/

Abstract: OBJECTIVE: This study examined the role of static, distal indicators and clinically-relevant proximal indicators in the prediction of short-term community violence in a large, heterogeneous sample of adults with mental illnesses. METHODS: Data were pooled from five prospective studies of adults with mental illnesses (N=4,484). Follow-up data were available for 2,579 participants (57.5%). A hierarchical linear regression assessed the incremental validity of a series of variable clusters in the prediction of violence risk at six months: (1) static characteristics (i.e., age, sex, race/ethnicity, primary diagnosis); (2) substance use (i.e., alcohol use and drug use at baseline); (3) clinical functioning (i.e., psychiatric symptoms at baseline, recent hospitalization); (4) recent violence; and (5) recent victimization. RESULTS: Results demonstrated improved prediction with each step of the model, indicating that proximal indicators contribute to the prediction of short-term community violence above and beyond static, distal characteristics. When all variables were entered, current alcohol use, recent violence, and recent victimization were positive predictors of subsequent violence, even after controlling for participant characteristics. CONCLUSIONS: This study provides empirical evidence for three proximal, clinically-relevant indicators in the assessment and management of short-term violence risk among adults with mental illnesses: current alcohol use, recent violence, and recent victimization. Consideration of these indicators in clinical practice may assist in the identification of adults with mental illnesses at heightened risk of short-term community violence.