Professional News
How Successful Are Residents in Predicting Violence Risk?
Psychiatric News
Volume 47 Number 17 page 20-21

Mass shooting incidents always prompt renewed discussion about mental health clinicians’ ability to evaluate someone’s risk for violence.

Experience matters for that skill, and so do numbers, according to a study published online September 1 in Psychiatric Services in Advance. Attending clinicians did a better job than residents in assessing potential for violence among psychiatric inpatients at a California hospital, the study concluded.

“It’s reassuring to know that, but it isn’t as obvious as it seems,” lead author Alan Teo, M.D., now a Robert Wood Johnson Foundation clinical scholar at the University of Michigan, told Psychiatric News.

In fact, little research explores how training and experience affect risk assessment, wrote Teo and colleagues, who were all at the University of California, San Francisco, when working on the study.

Using a retrospective, case-control design, they compared evaluations of 151 patients who had physically assaulted staff with those of 150 randomly selected nonviolent patients.

A total of 38 psychiatry trainees, most a year or two into residency, had evaluated 52 patients at admission. Forty-one attending psychiatrists, on average 14 years after completing residency, assessed the other 249 patients.

The attendings predicted risk accurately for 70 percent of the patients, but the residents got it right only 52 percent of the time—roughly equal to flipping a coin.

The researchers then used a five-item clinical subscale of the Historical, Clinical, Risk Management-20 (HCR-20-C) to evaluate charts of all the patients.

The five questions covered active symptoms of major mental illness, impulsivity, lack of insight, negative attitudes, and unresponsiveness to treatment, said Teo. “Twenty items is just not feasible in a busy clinical setting.”

Use of the HRC-20-C would have improved the residents’ accuracy to 67 percent, almost that of their more-experienced colleagues. However, the latter would not have done much better using the list.

Still, said Teo, “It’s not every day that you can find a checklist to make up for that gap in clinical experience.”

Use of structured risk assessments like the HCR-20-C might make residents more aware of useful information arising during an admission workup and help them improve their skills, the researchers suggested. inline-graphic-1.gif

“The Relationship Between Level of Training and Accuracy of Violence Risk Assessment” is posted at http://ps.psychiatryonline.org/PSInAdvance.aspx.

Interactive Graphics


Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).
Related Articles