Director’s Blog: Words Matter
By Thomas Insel on
NIMH just reached a milestone -- our first grant was awarded 65 years ago last month. Rather than celebrating, this anniversary has been allowed to pass quietly. With so much progress in genomics and neuroscience, we at NIMH have mostly been trying to keep up. But these kinds of anniversaries afford a good time to take stock -- all of us at NIMH would be remiss not to consider how far we have come since 1947. There have been many achievements: Nobel Prizes, great technologies, new treatments, and a vast enterprise for exploring the brain and behavior. But looking back is also sobering. Our original charge, from President Truman, was simply an executive order to fix the problems of America’s returning veterans who were struggling with “shell shock” or “combat neurosis.” Last month, we received another Presidential executive order. The topic – you guessed it: PTSD and TBI. Mission not accomplished.
During the same decades when scientific discovery has led to the eradication of many infectious diseases, has converted childhood leukemias from 95% fatal to 95% curable, and has reduced cardiovascular mortality by nearly 70%, our success rate with PTSD has been no better than our success at reducing war or trauma. In fact, for all mental disorders, while we have treatments, we lack cures, we lack vaccines, and we lack diagnostic biomarkers. Most of all, we lack a rigorous understanding of the disorders, at least on a par with our understanding of infectious diseases, childhood cancer, or cardiovascular disease. We need better science at every level from molecular biology to social science. Serendipity helps, but science, science that is rigorous and deliberate and even disruptive, is our North Star. That is why NIMH uses as its tag line that “research = hope.”
But there are many barriers to progress, not all of them are scientific. Some involve policy, some involve poverty, and remarkably, some are simply linguistic. In mental health, we are stymied by our language. The most obvious linguistic problem can be found in our current diagnostic terms, what my predecessor Steve Hyman has called “fictive categories.” Terms like “depression” or “schizophrenia” or “autism” have achieved a reality that far outstrips their scientific value. Each refers to a cluster of symptoms, similar to “fever” or “headache.” But beyond symptoms that cluster together, there should be no presumption that these are singular disorders, each with a single cause and a common treatment. Recall that Bleuler, who first introduced the term schizophrenia over a century ago, referred to “the schizophrenias.” And with new genetic discoveries, scientists are beginning to describe “the autisms,” a group of neurodevelopmental disorders of diverse causes.
Those who constructed the DSM were looking for a common language to describe symptoms, not a common biology or a common treatment. As someone who entered psychiatry pre-DSM-3, I can attest to the value of a common language. But there have been costs as well. In DSM-4, for instance, the diagnostic criteria for depression require 5 of 9 features, so it would be possible for two people with 1 of 9 criteria in common to have this same diagnosis. Not exactly “precision medicine,” but this approach has delivered diagnostic reliability. What is missing is validity. DSM never presumed to confer validity or explanatory value, but the field has imbued these symptom clusters with biological meaning, perhaps understandable in the absence of biomarkers or diagnostic tests. Ironically, this linguistic oversight has precluded the development of biomarkers that might confer validity. One reason we do not have biomarkers for mental disorders is our presumption that the biomarker is only valid if it maps on to a “fictive category,” rather than developing diagnostic categories based on the experimental data, as proposed by RDoC, our version of “precision medicine.”
Language traps us in even more subtle ways. There is no shortage of problematic words in our field. The term “stigma” may perpetuate a sense of being victimized with the unintended consequence of increasing discrimination and exclusion. There is an interesting ongoing debate about calling PTSD a “disorder” when it is unequivocally an injury. And conversely, for some in the autism community, a presumption that autism is an injury when much of the evidence points to autism as a neurodevelopmental disorder.
As a provocative question for our 65th birthday, I was recently asked if we should continue to be identified as NIMH when we study mental disorders more than mental health? Does the inclusion of “mental health” in our name (in contrast to the National Cancer Institute, the National Institute for Allergy and Infectious Diseases, the National Institute for Neurological Disorders and Stroke) reveal an ambivalence about our mission to transform the understanding and treatment of mental illness, especially serious mental illness? There is no ambivalence, but I appreciate the spirit of the question.
Some linguistic problems are easily solved. We can improve our current diagnostic categories via RDoC. We can find words that improve on “stigma.” Other linguistic issues, like the name of our institute, require literally an act of Congress. But on all of these issues, we need a broad conversation to help us understand how our language may be holding us back, limiting not only our impact but our imagination. Words matter, often in ways that are both subtle and profound.
Why has it taken so long for biological psychiatry to develop clinical tests and what to do about it? Kapur S, Phillips AG, Insel TR. Mol Psychiatry. 2012 Aug 7. doi: 10.1038/mp.2012.105. [Epub ahead of print] PMID 22869033