The Imperfection of Protection Through Detection and Intervention: Lessons from Three Decades of Research on the Psychiatric Assessment of Violence Risk
Posted: 27 Jul 2011 Last revised: 3 Dec 2012
Date Written: March 1, 2009
Abstract
This article reviews what we currently know about the accuracy of risk assessments and the ability of mental health professionals to “connect the dots” that foreshadow violence. The problem with anticipating and intervening to stop rare but serious violence is not really a matter of connecting dots, but of having dots (more precisely, “risk factors”) that are imperfect indicators of actual future violence.
The article first summarizes the Tarasoff decisions, which provided the first formulation of a protective duty that mental health professionals now take for granted. Section II explains why “warning signs” of violence that are apparent in retrospect merely provide an illusion that mental health professionals can detect violence in advance and stop it. Sections III and IV summarize research on the connection between violence and mental illness and on the ability of mental health professionals to assess the risk of (or less formally, to “predict”) violence. Section V uses some hypothetical mathematical examples to show what would happen if professionals attempted to stop violence by mentally ill persons through efforts to predict and intervene through confinement. The Conclusion suggests a different approach, akin to that used in public health efforts, that might reduce violence as a “side effect” of better treatment for persons with severe mental illness.
Keywords: violence, risk factor, risk assessment, violence prediction
JEL Classification: C00, I10
Suggested Citation: Suggested Citation