A test for suicide? by Elana Premack Sandler, L.C.S.W., M.P.H.

Should there be a "magic bullet" for suicide prevention?

This summer, a team of Harvard University psychologists released two tests that could change how mental health clinicians determine suicide risk.

These tests are objective measures that demonstrate an association between certain ways of thinking and a greater likelihood of suicidal behavior.

One test, which looks at how people pay attention to individual words, found that people who later attempted suicide paid more attention to words related to suicide. The second test, which measures unconscious associations between words, found that suicidal people made stronger associations between words related to "self" and words related either to "life" or "death/suicide." (You can read more about these tests and the related published studies here.)

I found out about these tests from my colleague, who raised a number of good questions that I'd like to reflect here.

1) Should there be, a "magic bullet" for suicide prevention?

A "suicide test" challenges conventional wisdom around suicide prevention in clinical settings. Traditionally, practitioners have struggled with how to ask about suicide and how to gauge what answers really indicate acute risk. A test that objectively predicted suicide risk would be quite a relief. But, that raises another question:

2) How might these tests take away the human dimension of a clinical assessment?

Even as I wrote out that question, I thought, "Well, the point of a clinical assessment is that there is some objectivity, and as humans we're rarely objective." So maybe the point is to take away the subjectivity, to make it clear - this person is suicidal, and this person is not. But, people aren't really that simple, are we?

3) Wouldn't it be equally useful to, in my colleague's words, research the helpful dialogues that get patients to tell practitioners that they have suicidal thoughts?

Yes, please. My colleague is a natural skeptic, so she really wants to put these tests to the test. Being the nerd I am, I'm excited to think that there could be a scientifically proven way to help more people that might be easier than asking really hard questions. At the same time, I wouldn't want to de-emphasize the value of the relationship between a clinician and a patient. Ideally, suicide assessment would involve a combination of an objective measure and a relationship-based conversation.

Obviously, and as usual, more questions than answers. What do you think?