An Unholy Trinity in Current Suicide Risk Management

Image: fizkes/Shutterstock

I was a graduate student in neuroanatomy before entering medical school. My well-considered plan anticipated an eventual career in neurosurgery. Yet, severe immune dysfunction to surgical latex permanently ended my chosen specialty.

As a struggling, self-punishing, expatriated surgical resident, I frantically searched for stand-in training. Without insight, and too proud to ask for help, I discovered an immediately available, and tangentially-related residency prospect. I entered emergency psychiatry training in one of the state’s most agitated programs. Paradoxically, I had stood before that department’s ED door an hour after resigning from surgery, and not as a prospective trainee. Absurdly, no one answered, and I walked away, in the end, to enter the field.

My unusual training history, and nearly four decades of experience, in a now essentially extinct ED specialty, have provided unique personal and professional perspectives. Indeed, patient outcomes from the existing, unholy trinity of ineffective access, useless assessments, and a system unable to provide adequate training are dreadful.

Clearly, current behavioral health practice, and suicide research for that matter, is a complex mix of organizational, community, and patient-centered factors. It is compulsory to ask new and disturbing questions of a shattered system. Data confirm the staggering tragic consequences experienced by suffering individuals and families.

The plural of anecdotes is not evidence. Miguel de Cervantes, a 17th century Spanish writer, and whose Don Quixote was translated into more languages than any other book, excepting the bible, wrote “Facts are the enemy of the truth.” John Adams, the second U.S. President, and defense lawyer for British soldiers following the Boston Massacre, argued “Facts are stubborn things.”

Favorable outcome is a cross factor product of actual, timely access and effective assessment. In turn, it is contingent on a competent workforce and training at a level equal to, or greater than, licensure. What are the relevant facts? According to the most recent data from the CDC, National Council on Behavioral Health, U.S. Department of Education, American Association for Suicidology, Association of Medical Colleges, and the peer-reviewed literature:

  1. 55 percent of U.S. states are experiencing a severe shortage of general psychiatrists.
  2. 95 percent of U.S. states are experiencing a severe shortage of child and adolescent psychiatrists.
  3. Emergency psychiatry training program funding by the Centers for Medicare and Medicaid Services is essentially nonexistent.
  4. Psychiatry is no longer a top 10 residency preference training program.
  5. Between 2003 and 2013, there was a 10 percent reduction in the U.S. psychiatric workforce, and projected greater than 12 percent reduction by 2025.

Attention to U.S. suicide base rates, absolute numbers, and recency of clinical contact before suicide death demonstrates that outcomes, as a function of insufficient and ineffective access and training, and other factors, are appalling.

  1. Suicide rates have increased by 25 to 30 percent in the U.S. since 1999, with a 50 percent increase in women.
  2. Between 1999–2017, the U.S. youth suicide rate increased among most age groups, and highest among young girls 10 to 14 years old. In total, over 100,000 U.S. youth, ages 8–24, committed suicide, and 1309 were children ages 5 to 12.
  3. More than half of mental health visits between 2014–2016 were with psychiatrists.
  4. 50 percent of those who died by suicide were not diagnosed or recognized with a mental disorder.
  5. Up to 10 percent of deaths occur within hours of out-patient, ED, or in-patient psychiatric discharge, and 20 percent within days.
  6. Nearly 80 percent of patients, including youth, who committed suicide denied ideation in their last contact with a mental health practitioner.
  7. Some in-patient psychiatric post-discharge suicide rates are equivalent to the U.S. annual number of out of hospital cardiac arrests (OHCA).
  8. Systematic review in young people reported no differences between treatment and control groups in reducing suicide risk except for one small study. Some suicides rates for mental health cohorts with treatment (including veterans) are greater than suicide rates for the same disorders without treatment.

These disturbing findings contradict the widely-held view that current access to assessment is sufficient to foresee suicide, and further substantiate the fact that suicide behaviors are neither straightforward or low base-rate disorders.

Like the absence of chest pain in women with impending MI, the immense difficulty is that symptoms are often elusive rather than blatant. For example, there are considerable differences between acute disturbance in mental health, such as command hallucinations and toxic inhibitions, contrasted with absence of psychological health, as in repetitive, deliberate self-harm. Nevertheless, the common denominator is deterioration of or current change in the person’s mental status suggesting emerging danger.

The current epidemiological evidence tragically confirms that the benefit of asserted, state-of-the-art ideation-centric assessments is not only hard to determine, but inadequate. What, then, does this signal? 1. Current risk assessment is problematic; 2. Preoccupation with current risk factor analysis, without consideration to suicide phenotypes, is not sufficient to foresee suicide; 3. Increased suicidal risk in hypothesized low and high-risk groups may eventually be elucidated by an individual’s special vulnerability, including neurologic, to heretofore, unrecognized unique or combination risk factors; The absence of ideation is not a benign finding.

In a follow-up article, the ideation/depression construct in suicide assessment will be challenged. This dated, two-dimensional theory will be improved with a more accurate, 3D rotational assessment of danger (see Copelan R, American Journal of Emergency Medicine, Sept 2006). The discussion will follow the example of a chest x-ray. Although the acceptable standard of care in medicine recommends initial, plane views of the chest, MRI is used for better resolution to stage pathology. Similarly, the author will provide a new, statistically significant assessment matrix to inform the difficult diagnostic question “What is the impact of ideation and nonideation states on attempt rates across subgroups?”

More from Op-Med