A widely quoted clinical aphorism is that there are two kinds of therapists: those who have experienced the loss of a patient to suicide, and those who haven’t yet. Although the expression sounds a bit nihilistic, this adage conveys a warning to those clinicians who have not experienced a patient’s suicide: do not become overly confident in the ability to predict suicide. In fact, predicting suicide is the most challenging situation for a mental health practitioner– not only because it is literally a matter of life and death, but also because even the most skilled and empathic of clinicians can never really know what lies in the heart and mind of a patient. Even when there is full and open communication between caregivers and patients, life events and irresistible impulses can occur that destabilize patients’ emotional equilibrium and lead them to desperate acts.
Assessing someone’s risk for suicide involves understanding the biological factors that contribute to risk such as the presence of medical or psychiatric conditions associated with higher rates of suicide; psychological risk factors including an individual’s attitude about the future, religious beliefs and immediate emotional state; and social factors such as whether or not the person has available and supportive networks of friends and family. Familiarity with all the clinical and personal factors that need to be considered when attempting to predict suicide is an essential first step, and a great deal of research has been done which sheds light on the contributions of these factors.
In the United States, someone dies by suicide every 15 minutes and suicide is the tenth leading cause of death. Despite widespread efforts at public and professional education and at reducing the stigma of mental illness, the U.S. suicide rate has been increasing over the last 15 years. Approximately 90% of persons who die by suicide have diagnosable mental disorders; 60% of elderly suicide victims have consulted with a primary care practitioner in the months preceding their death. Given this knowledge, how does this serious public mental health crisis continue to grow?
Unfortunately, predicting suicide is an extraordinarily difficult challenge. We know a great deal about actuarial risk – persons who are at risk due to demographic, historical and clinical factors. We also know a fair amount about times of high risk for suicide, specifically times of loss and times of transition. However, we have much less knowledge about, and therefore a rather limited ability to predict, short term risk. Compounding this limitation, is the good and bad news that among all those persons at risk, and despite the fact that suicide is the tenth most common cause of death, the actual rate of suicide is “only” 12/100,000 (Centers for Disease Control and Prevention, 2009 data). Consequently, it is very difficult to predict who will actually attempt suicide in the near term from among those experiencing a dangerous combination of risk factors.
What can be done by a clinician trying to help individual patients in emotional pain and at risk for suicide? Clinicians must stay up to date with scientific knowledge and evidence-based practices. Most practitioners know that older white men, especially those with co-morbid substance abuse or medical illness, have the highest rate of suicide. But are we up to date with emerging information about the significant increases in suicide rates among both middle-aged men and women? And are we aware that women, who continue to make three times as many attempts as men but whose attempts are less likely to be lethal, are increasingly using dangerous means of suicide, including using more firearms? Staying current with the scientific literature is a daunting task, but in this internet age and thanks to organizations like the American Foundation for Suicide Prevention (www.afsp.org, an organization dedicated to supporting research, disseminating information and providing support to the survivors of suicide), lay persons and professionals can easily update their knowledge and access valuable resources. It is also incumbent on practitioners to be aware that the internet provides detailed information on how to commit suicide, making it all the more important that we ask our patients what they are studying and what means they have considered.
Clinicians must also do everything in their power to truly understand and foster meaningful connections with the individuals in their care. Although this sounds obvious and simplistic, our health care system does not treat it as obvious, and therefore it is not at all simple. Brief evaluations, over-reliance on patients’ self-reports, short follow-up appointments, and patients seeing many different providers who themselves may not be systematically passing along critical data at each “hand-off”, undermine the essential emotional bond between caregiver and patient, which is the cornerstone of a mutual and trusting relationship. Persons considering suicide often have lost hope and any ability to see a positive future for themselves; sometimes knowing that a concerned and involved caregiver believes there might indeed be relief and hope can be lifesaving. Where in our contemporary mental health system, which treats therapy like a commodity to be sought by “consumers”, is there recognition of the value of a genuine and enduring human connection between healer and those in need of healing? These comments have a tinge of the polemical about them but are intended to draw attention to some serious science.
In a perfect world, an established and enduring relationship between caregiver and patient would help prevent suicide because each would know the other well enough for the caregiver to recognize subtle clinical changes in his or her patient, and for the patient to comfortably confide thoughts and impulses deemed unacceptable, even unspeakable, anywhere else. Such relationships are increasingly rare in today’s healthcare system. However, clinicians can inform themselves about evidence-based methodologies for effective communication with persons who may be thinking of ending their own lives. Communicating with persons who are contemplating suicide or who may have already decided to do so, requires time, patience, understanding and the sophistication to not take “No, I won’t” for an answer. In 1998, Shea outlined the Chronological Assessment of Suicide Events (CASE), an invaluable method of inquiry which systematically elicits a patient’s history, critical attitudes, and beliefs. Employing such an approach could help any caregiver save someone’s life. This approach can be used during a psychotherapy session in a practitioner’s private office or while in an intense emergency room. It proceeds through a sequential exploration of a patient’s current suicidal behaviors and ideas, then on to a review of recently occurring ones. Next, it examines more distant past actions and ideas and concludes with a discussion of future or anticipated ones. Complemented by using techniques such as Pomeroy and colleagues’ idea of “gentle assumption,” wherein the clinician acts as though he or she just presumes that problematic or dangerous behaviors have occurred, and Shea’s own notion that a patient’s “denial of the specific” should not dissuade the interviewer from empathically probing for more detail on similar or related behaviors and beliefs, CASE provides a schema for connecting with the individual in distress and collecting thorough and critical data.
The opinions and observations of other caregivers and patients’ loved ones are critical sources of information. Clinicians may hesitate to aggressively pursue these sources for fear of breaching confidentiality. Asking a patient’s other caregivers for relevant information and asking a patient’s friends and family about unusual, erratic or dangerous behaviors can be crucial interventions. There may be times when disclosing professional concerns about a change in a patient’s presentation may be warranted. While it is a core principle of our professional work to faithfully maintain a patient’s confidentiality, “first, do no harm” is also a core principle. Sometimes safety does trump privacy. Clinicians may want to seek consultation or peer supervision when uncertain how to navigate complex ethical matters and contradictory expectations. Many clinicians outline confidentiality boundaries early on in treatment by anticipating with their patients those clinical circumstances which might require a recalibration of the delicate balance between confidentiality and safety.
Lastly, a word about the utility of conducting formal suicide risk assessments. In his 2011 book, Preventing Patient Suicide, Simon repeated his caution that “suicide risk assessment is a process, not an event.” Over the last thirty years, a variety of scales have been devised to assess suicide risk. No single one has been widely adopted, and all share the limitation of capturing only a snapshot in time of interacting risk and protective factors. Nonetheless, these scales can serve as organizing templates, checklists, and adjuncts to clinical evaluations. Perhaps the most compelling and multi-dimensional of these instruments is the Columbia-Suicide Severity Rating Scale. The Columbia scale examines four discrete aspects of suicidality: severity of ideation, the intensity of ideation, actual suicidal behaviors, and the lethality of the behaviors.
In summary, predicting suicide is almost as difficult as it is important. Much is known, and even more remains to be learned. The current rate of suicide poses a major public mental health challenge. We have a collective responsibility to our patients and to our professions to do all that we can to advance the study and prevention of suicide.