RESEARCH TRIANGLE PARK, N.C. — A new study in the International Journal of Methods in Psychiatric Research quantifies the effects of common methodological limitations that produce flawed results in much of the existing research on violence perpetrated by adults with mental illness. The study, which analyzed secondary data on 3,000 adults with mental illness and employed various simulation methods, was conducted by RTI International and North Carolina State University, and shows how results from some research should be questioned, especially as clinicians and policy makers frequently rely on scientific evidence to inform treatment and policy concerning people with mental illness.
“Violence risk screening and assessment, and subsequent management, are all critical parts of clinical care and rely upon valid, reliable and generalizable research,” said Kiersten L. Johnson, PhD, RTI research public health analyst and the paper’s lead author.
Recent high-profile incidents, including mass shootings, have resulted in increased scrutiny on clinicians’ ability to identify and target violence-related factors among adults with mental illness, even as there is little evidence linking mental illness to serious violence perpetration.
“Addressing these limitations is particularly important as current violence risk assessment and management practices, as informed by existing research, could be introducing unnecessary harms to the client being assessed and to the client–clinician relationship,” explained Richard Van Dorn, PhD, RTI senior mental health researcher, and principal investigator of the study.
The three common methodological limitations explored in the paper include:
- Insensitive measurement of violence that results in high levels of information bias,
- Use of data that measures the risk and protective factors and violent outcomes at the same time, and
- Use of data with long measurement gaps between risk and protective factors and violent outcomes.
As detailed in the study, not utilizing self-reported violence data and instead relying on treatment or arrest records, or collateral informant report, misses over half of all violent incidents. Furthermore, data that exclude self-reported violence produce biased associations between substance use and psychiatric symptoms and violence. Associations between baseline variables and subsequent violence attenuate over long time-lags and, when paired with violence information bias, result in fewer significant effects than should be present. Moreover, the true direction of the simulated relationship of some significant effects is reversed.
“It is our hope that this study will spur serious consideration about ways to improve the science and practice of clinical violence risk screening, assessment and management,” explained Sarah L. Desmarais, PhD, associate professor of psychology at North Carolina State University, director of the university’s Center for Family and Community Engagement and second author of the paper.
Funding for this study was provided by the National Institute of Mental Health (NIMH), Award R01MH093426.