Empathy, Collaboration, and Trust in Suicide Risk Assessment
In their recently published Scientific American article “Suicide Risk Assessment Doesn’t Work,” Declan Murray and Patrick Devitt note the limitations of using risk factors in assessing suicide risk and address a significant problem in mental health. We agree with the conclusion that over-reliance on formulaic assessment can interfere with real engagement with patients. We cannot agree, however, that the best recommended practice is to send suicidal patients home with the reassurance that statistically they are likely to stay alive "no matter what we do."
Murray and Devitt begin their article describing the threat from a distraught husband, "if you don't admit her to the hospital, you will have blood on your hands." While this might be an accurate representation of the nightmares of early-career psychiatrists, this is not a representative description of the challenges real therapists and psychiatrists face in evaluating potentially suicidal patients. Murray and Devitt frame their analysis in terms of an adversarial relationship, in which the primary concern of the practitioner is to avoid moral and legal risk. In reality, there are many ways that suicidality can present, both outside the hospital and in hospital or emergency room settings, and often the negotiation around suicide assessment is collaborative, and based on empathy and trust.
It is a serious error to conflate a full clinical assessment with reliance on standardized instruments and risk factors. The authors suggest that the patient is likely to remain alive "no matter what," but a thorough risk estimation is undoubtedly related to how the patient describes her dilemma and what the husband has observed. While we do not have specific research evidence to prove that an individual who has been found with a gun in her mouth and her finger on the trigger is more likely to die from suicide, we would nonetheless be unwilling to disregard this observation. Similarly, if a patient said to one of us (as some have), "I feel so terrible right now I can't stand it and will kill myself if I don't get help," we would be likely to conclude that the patient is indeed in need of urgent assistance.
We therefore cannot support the "unequivocal good response" of the psychiatrist, as described by Murray and Devitt, to this distressed husband. The authors' conclusions do not flow logically from the literature that they cite. In a situation such as this, we can agree that risk assessment instruments may not be helpful, but the clinician at a minimum should speak to the patient, do a full assessment, assess the patient's level of distress and the presence or absence of suicidal intent, and what situation has led to the current crisis. In our view, the application of the low population base rate (by declaring that there is "a 99.9% likelihood that she will be alive on Monday") shows poor understanding of how to translate and apply empirical findings to clinical practice. Should the hypothetical patient commit suicide, the authors' description of this as "an unpredictable tragic outcome" would not be accurate.
In a pointed critique of the uses and misuses of evidence-based medicine, Smith and Pell (2003)1 have observed that there is insufficient evidence to demonstrate that individuals jumping from airplanes have a statistically greater chance of survival if they implement the use of parachutes because there has not yet been a large enough sample of individuals prepared to test this hypothesis. Similarly, it is unlikely that there has been a sufficient number of clinicians willing to send home patients seeking emergency hospitalization with clear suicidal intent to generate meaningful comparison groups that would allow us to test the effectiveness of this intervention. It is a misuse of evidence-based medicine to reject the dictates of common sense and competent clinical assessment. Moreover, as a model of clinical intervention, this interaction seems most likely to enrage and frustrate the panicked spouse. A more helpful and ethical approach would be to perform a complete assessment of the situation, and convey to both the patient and family member clinical recommendations using the best information available at the time. The focus of the clinician should be on appropriate assessment, intervention, and empathic engagement, and not the application of statistics to limit legal liability.
1Smith, G., Pell, J. 2003. “Parachute use to prevent death and major trauma related to gravitational challenge: systematic review of randomized controlled trials.” British Medical Journal 327: 1459-61.
Strategic Initiatives: Suicide Research and Education
In April 2016 the National Center for Health Statistics reported that in the US, the suicide rate increased 24 percent between 1999 and 2014. Suicide is the 10th leading cause of death in the US, with just over 44,000 suicides per year in this country (afsp.org/about-suicide/suicide-statistics/). For mental health professionals, the suicide of a patient is the worst possible outcome of treatment. While many causes of death are not preventable, suicide is one cause of death that often leaves survivors with the sad feeling that it could have been otherwise…if only.
When someone dies by a person’s own hand, the loss deeply affects each of those closest to him or her. What may often be missed, however, is how that loss also affects the therapist trained to keep this suicide from happening. In the popular mind, it may seem inappropriate to be concerned with the professional on the sidelines after a successful suicide attempt, but for the human being behind that degree, a death can have many repercussions both professionally and personally.