A required reading for my students at The University of Utah is a brief essay by Dr. Martin Schwartz entitled “The importance of stupidity in scientific research.” In this essay, Schwartz talks about the importance of productive stupidity, which refers to being ignorant by choice. He notes that the most important questions we face often force us into the uncomfortable position of not knowing. However, being comfortable with ignorance and the possibility of being wrong is what enables us to make progress towards answering these important questions. In other words, the most important advances are typically achieved by those who are willing to admit, “I don’t know.” When it comes to suicide prevention, if we’re unwilling to acknowledge what we don’t know, this unwillingness may lead us to cling to faulty assumptions that cause us to become stuck in our efforts to prevent suicide.
Faulty Assumption about the Association of Suicide Ideation and Suicidal Behaviors
Perhaps the most prevalent faulty assumption about suicide is that the severity (or frequency) of suicide ideation is linearly associated with risk for suicidal behaviors. According to this perspective, as suicide ideation becomes more severe, the likelihood of a suicide attempt increases. This assumption is largely based on research like that conducted by Greg Simon and colleagues, who found that patients who endorsed thoughts about death and self-harm on a daily basis were 10 times more likely to attempt suicide or die by suicide than patients who denied such thoughts. That’s pretty notable. What most people are not aware of is another finding from that same study: over 96 percent of patients reporting thoughts about death or self-harm (even those reporting these thoughts on a daily basis) did not attempt suicide or die by suicide. This finding has been replicated in numerous other studies as indicated in the recent meta-analysis by Joseph Franklin and colleagues.
Another reason suicide ideation is an unreliable predictor of suicidal behaviors is because suicide ideation changes over time, sometimes very quickly and to a large degree. Although this characterization of ideation is not new – David Rudd first articulated it in the fluid vulnerability theory over a decade ago - it has only been within the past few years that researchers have provided the scientific support for it. The dynamic nature of suicide ideation explains why it is not a reliable indicator of suicidal behavior: asking someone to report the severity of their ideation during the past week (or two weeks) simply does not provide an accurate measure of that ideation. To put this into perspective, imagine asking someone to report their blood pressure during the past week; blood pressure varies enough on a moment-to-moment basis that asking an individual to report a single score that describes an entire week would have only limited accuracy.
Faulty Assumptions about the Transition from Suicidal Thoughts to Action
Another common, but faulty assumption is that suicidal behavior is the endpoint of a single, common pathway. This assumption underlies many contemporary theories and models of suicide, which presume that there is a handful of variables that account for suicidal behavior. However, accumulating data suggest that there are multiple different processes and pathways associated with suicide. Clinical experience leads me to believe that at least two “subtypes” of suicide seem to exist. The first involves a process characterized by highly variable suicidal thinking and risk factors and frequent crises (what I often call the “roller coaster” subtype) whereas the second involves a process characterized by relatively stable, low-level suicidal thinking that can easily remain undetected (what I often call the “slow simmer” subtype). As you can imagine, these very different subtypes likely respond differently to different suicide prevention strategies. The roller coaster subtype may respond better to treatments such as brief cognitive behavioral therapy for suicide prevention (BCBT) or crisis response planning (CRP), two strategies shown to reduce suicide attempts among military personnel. The slow simmer subtype, in contrast, may be difficult to detect and may therefore benefit more from enhanced means safety procedures. We should consider multiple pathways to suicide to increase the effectiveness of our prevention efforts.
Faulty Assumptions Lead Us to Ask the Wrong Questions
Why aren’t we better at identifying and detecting military personnel at risk for suicide? Because we’re asking the wrong questions. Yes, asking military personnel if they are thinking about suicide is an important question to ask and is a question we should continue asking, but we must accept that the answer to this question, whether positive or negative, is remarkably limited in accuracy. The uncomfortable reality is that many military personnel who are thinking about suicide will not disclose these thoughts to others and asking about such thoughts more often is unlikely to increase disclosure rates. Indeed, some military personnel have noted that the increased frequency of suicide risk screening is sufficiently obnoxious that they will actually deny suicidal thoughts in order to avoid having to answer additional questions about suicidal thoughts and behaviors. When we focus too intently on suicide ideation, we may be decreasing our ability to detect suicide risk.
To better identify and detect military personnel at risk for suicide, we need to develop assessment and detection methods that do not hinge on the honest self-disclosure of suicidal thoughts.
The Way Forward
Despite the expansion of suicide risk screening methods and the implementation of suicide prevention programs across the DoD, military suicide rates remain elevated, which calls into question the utility and validity of prevention strategies based on traditional assumptions (e.g., screening for suicide ideation, gatekeeper training). This is not to say that we should completely abandon these methods. Not asking about suicidal thoughts, for example, is not an acceptable solution. Rather, we need to be honest with ourselves that our traditional assumptions may be insufficient.
To save lives, we must become productively stupid and therefore aware of what we don’t know. Productive stupidity may be as simple as challenging our deep-rooted assumptions by asking two questions:
How do we know this?
What if we’re wrong?
Productive stupidity will help us develop assessment and detection methods that do not hinge on the honest self-disclosure of suicidal thoughts. Productive stupidity will help us identify more pathways to suicide and then develop interventions to deal with life’s obstacles along those pathways.
In summary, if we don’t know that something is true, we should be willing to be skeptical and conduct unbiased research on the idea. To prevent suicide, we must abandon faulty assumptions, even those we have promulgated for generations. Why? Because the cost of being wrong about suicide is too high.
Bryan is a board-certified clinical psychologist who researches military and veteran suicide. He served on active duty in the U.S. Air Force and deployed to Iraq in 2009. As executive director of the National Center for Veterans Studies, he oversees multiple research studies aimed at developing and testing new treatments to prevent suicidal behavior among military personnel.
The views expressed in Clinician's Corner blogs are solely those of the author and do not necessarily reflect the opinion of the Psychological Health Center of Excellence or Department of Defense.