This summer's shootings in New York City, Oak Creek, Wisconsin, and Aurora, Colorado have led to a lot of soul-searching, and not just about gun laws. Many have called for better mental health screening and treatment to help heal would-be shooters before they commit crimes. It's worth remembering that people with mental illness are more likely to be victims of mental illness than perpetrators, but it's still reasonable to ask whether the mental health system could be doing more to spot signs of violence. And a recent study may offer one way for it to do so.
Psychiatrist Alan Teo and his coauthors looked at 41 trained psychiatrists and 38 still-training residents on four locked psych wards to see how they had evaluated the violence risk posed by a series of patients. Then they compared the evaluations with the patients' actual history of violence while on the wards. They found that while the psychiatrists were able to predict with moderate success which patients would become violent, the residents' predictions were no better than chance.
However, the study authors then combined the residents' evaluations with another form of risk assessment: a scale called the Historical, Clinical, Risk Management–20 clinical subscale (HCR-20-C). This scale requires that clinicians look for the following five characteristics in a patient:
• lack of insight
• negative attitudes
• active symptoms of major mental illness
• unresponsiveness to treatment
For each of these characteristics, the clinician assigns patients a 0 (absent), 1 (possibly present), or 2 (present). When Teo and his team combined the residents' evaluations with the patients' HCR-20-C scores, the combined data were about as good at predicting violence as the psychiatrists' evaluations were.
So basically, the addition of a very simple tool can help a resident predict whether a patient will be violent just as accurately as a psychiatrist can. This is significant because psychiatrists are in short supply in many parts of the country, and many people who enter psych wards or emergency rooms with severe mental illness will be seen, at least initially, by a resident. The HCR-20-C could help residents keep other patients and themselves safe.
The study is also encouraging, in a broader way, since it identifies a simple method for evaluating someone's potential for violence that has empirical support. Teo told BuzzFeed Shift that scales like the HCR-20-C could be "one piece in the puzzle" of preventing mass shootings. He explained, "if clinicians in this country, particularly those who are less experienced, start using a tool like the HCR-20-C to evaluate risk of violence, we are going to do a better job at figuring out who is at a higher risk and who is at a lower risk of violence. High risk cases that would have been missed previously might get detected with this tool." Once they are detected, doctors will have to figure out the best way to mitigate their risk to others, through treatment that may include hospitalization or medication.
Of course, James Holmes had been seen by a psychiatrist before he opened fire during The Dark Knight Rises — and she apparently reported her concerns about him to campus police. Helping mental health professionals identify potentially violent patients may be a step in the right direction, but police need to know what to do with that information too. Reducing the likelihood of another mass shooting is going to be a complicated proposition, of which any scale, as Teo says, will form only a part.