Skip to main content
Research and Innovation

Study IDs Key Indicators Linking Violence and Mental Illness

Photo credit: Stu. Shared under a Creative Commons license. Click for more information.

For Immediate Release

New research from North Carolina State University, RTI International, Arizona State University and Duke University Medical Center finds a host of factors that are associated with subsequent risk of adults with mental illness becoming victims or perpetrators of violence. The work highlights the importance of interventions to treat mental-health problems in order to reduce community violence and instances of mental-health crises.

“This work builds on an earlier study that found almost one-third of adults with mental illness are likely to be victims of violence within a six-month period,” says Richard Van Dorn, a researcher at RTI and lead author of a paper describing the work. “In this study, we addressed two fundamental questions: If someone is victimized, is he or she more likely to become violent? And if someone is violent, is he or she more likely to be victimized? The answer is yes, to both questions.”

The researchers analyzed data from a database of 3,473 adults with mental illnesses who had answered questions about both committing violence and being victims of violence. The database drew from four 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. For this study, the researchers used a baseline assessment of each study participant’s mental health and violence history as a starting point, and then tracked the data on each participant for up to 36 months.

Specifically, the researchers assessed each individual’s homelessness, inpatient mental-health treatment, psychological symptoms of mental illness, substance use and as victims or perpetrators of violence. The researchers evaluated all of these items as both indicators and outcomes – i.e., as both causes and effects.

“We found that all of these indicators mattered, but often in different ways,” says Sarah Desmarais, an associate professor of psychology at NC State and co-author of the paper. “For example, drug use was a leading indicator of committing violence, while alcohol use was a leading indicator of being a victim of violence.”

However, the researchers also found that one particular category of psychological symptoms was also closely associated with violence: affective symptoms.

“By affect, we mean symptoms including anxiety, depressive symptoms and poor impulse control,” Desmarais says. “The more pronounced affective symptoms were, the more likely someone was to both commit violence and be a victim of violence.

“This is particularly important because good practices already exist for how to help people, such as therapeutic interventions or medication,” she adds. “And by treating people who are exhibiting these symptoms, we could reduce violence. Just treating drug or alcohol use – which is what happens in many cases – isn’t enough. We need to treat the underlying mental illness that is associated with these affective symptoms.”

The research also highlighted how one violent event could cascade over time.

For example, on average, the researchers found that one event in which a person was a victim of violence triggered seven other effects, such as psychological symptoms, homelessness and becoming perpetrators of violence. Those seven effects, on average, triggered an additional 39 effects.

“It’s a complex series of interactions that spirals over time, exacerbating substance use, mental-health problems and violent behavior,” Van Dorn says.

“These results tell us that we need to evaluate how we treat adults with severe mental illness,” he adds.

“Investing in community-based mental health treatment programs would significantly reduce violent events in this population,” says Desmarais. “That would be more effective and efficient than waiting for people to either show up at emergency rooms in the midst of a mental-health crisis or become involved in the legal system as either victims or perpetrators of violence.

“We have treatments for all of these problems, we just need to make them available to the people that need them,” Desmarais says.

The paper, “Leading indicators of community-based violent events among adults with mental illness,” was published online Dec. 21 in the journal Psychological Medicine. The paper was co-authored by Kevin Grimm of Arizona State, Stephen Tueller and Kiersten Johnson of RTI International, and Marvin Swartz of Duke University Medical Center. 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.

“Leading indicators of community-based violent events among adults with mental illness”

Authors: R.A. Van Dorn, S.J. Tueller and K.L. Johnson, RTI International; K.J. Grimm, Arizona State University; S.L. Desmarais, North Carolina State University; and M.S. Swartz, Duke University Medical Center

Published: Dec. 21, Psychological Medicine

DOI: 10.1017/S0033291716003160


Background. The public health, public safety and clinical implications of violent events among adults with mental illness are significant; however, the causes and consequences of violence and victimization among adults with mental illness are complex and not well understood, which limits the effectiveness of clinical interventions and risk management strategies. This study examined interrelationships between violence, victimization, psychiatric symptoms, substance use, homelessness and in-patient treatment over time.

Method. Available data were integrated from four longitudinal studies of adults with mental illness. Assessments took place at baseline, and at 1, 3, 6, 9, 12, 15, 18, 24, 30 and 36 months, depending on the parent studies’ protocol. Data were analysed with the autoregressive cross-lag model.

Results. Violence and victimization were leading indicators of each other and affective symptoms were a leading indicator of both. Drug and alcohol use were leading indicators of violence and victimization, respectively. All psychiatric symptom clusters – affective, positive, negative, disorganized cognitive processing – increased the likelihood of experiencing at least one subsequent symptom cluster. Sensitivity analyses identified few group-based differences in the magnitude of effects in this heterogeneous sample.

Conclusions. Violent events demonstrated unique and shared indicators and consequences over time. Findings indicate mechanisms for reducing violent events, including trauma-informed therapy, targeting internalizing and externalizing affective symptoms with cognitive–behavioral and psychopharmacological interventions, and integrating substance use and psychiatric care. Finally, mental illness and violence and victimization research should move beyond demonstrating concomitant relationships and instead focus on lagged effects with improved spatio-temporal contiguity.

Leave a Response

Your email address will not be published. All fields are required.