Psychiatry Should Stay Comfortable In Its Own Skin: No Good Comes From Overselling Our Science Base

Psychiatry is a wonderful specialty. We have highly effective medication and psychotherapy tools. Forty years of accumulated clinical research have given us a pretty clear idea of optimal treatment guidelines. With an accurate diagnosis and an appropriate treatment, most of our patients benefit greatly and many recover completely.

But there is one source of great and continuing frustration in our field. We are in the midst of a neuroscience revolution that has provided a miraculous and tantalizing window into normal brain functioning. But the vast accumulation of basic science knowledge revealing the mechanisms of normal brain functioning has shed relatively little light on the far greater complexity of what causes psychopathology. As a result, the neuroscience revolution has so far had almost no impact on how we diagnose and treat our patients. The inherent difficulty in translating from basic to clinical science guarantees that we will make only slow progress in unraveling the multitudinous heterogeneity of brain malfunctions that cause mental illness.

DSM-5 initially got into trouble because it was ambitious to jump-start a “paradigm shift” in psychiatry-well before there was sufficient scientific knowledge to make this possible. We would not have been burdened by all the dangerous DSM-5 suggestions for unproven diagnoses if its workgroups had not been given the green light to be recklessly creative in promoting their pet innovations. This same excessive ambition informs the hype surrounding the reorganization of the chapter headings proposed for DSM-5. Psychiatry does itself no good when we oversell ourselves.

In fact, the reorganization of chapters proposed for DSM-5 is a small and sensible change that (with only two very notable exceptions) will probably do no harm. The previously aggregated disorders that present in childhood and adolescence will be grouped with descriptively similar disorders that present in adulthood. Obsessive-compulsive disorder will lead a new category followed by a spectrum of possibly related disorders. Schizophrenia will similarly have its own spectrum. Disorders related to stress will be grouped together. Bipolar and unipolar mood disorders will be separated.

So far, so good. Most of these changes were also considered for DSM-IV. Each has pluses and minuses, but overall they are plausible and may provide a somewhat clearer organizational scheme.

Two of the proposed organizational changes pose much greater risks. The first is having a category for “Addictions” that includes “Behavioral Addictions.” This will likely extend the boundary of mental disorder where it doesn't belong into shopaholism, workaholism, hypersexuality, exercise and Internet addiction, and who knows where else. The second is the emasculation of the personality disorders section and the elimination of multiaxial diagnosis.

My purpose here though is not to evaluate the proposed reorganization in any detail, but rather to caution that it is being far oversold as the treasured product of some sort of scientific revolution. The DSM-5 news release gushes breathlessly: “This restructuring of the DSM's chapters and categories of disorders signals the latest scientific thinking about how various conditions relate to each other.” “They should facilitate more comprehensive diagnosis and treatment approaches for patients and encourage research across diagnostic criteria.” “The sequence of chapters builds on what we have learned about the brain, behavior and genetics over the past two decades.”

Let's get real. The impact of any DSM has very little to do with the organization of its categories. Instead, what counts are the actual disorders included and how they are defined. Unless it comes to its senses, DSM-5 will result in an explosive and unjustified inflation in psychiatric diagnosis because of its many new, unsupported, and high prevalence “mental disorders” and the many diagnostic thresholds it plans to lower. It really matters little how the manual chapter headings are organized-the fallout will come from the diagnoses that are included and how criteria sets are written.

Which brings us to the false promises in the DSM-5 announcements of its proposed reorganization. Contrary to claims, the proposal does not represent some revolutionary advance indicating that our basic neuroscience findings can now play a large role in psychiatric diagnosis and treatment. Such statements trumpeting a role for basic neuroscience in day-to-day clinical practice are wildly premature and overblown.

You will know precisely when there is a paradigm shift with translational research finally impacting on our clinical work. It will be marked by the development of biological tests that can be used to guide our diagnosis and treatment. Until then any reorganization is not much more than window dressing. The DSM-5 changes are best seen as merely editorial-not really reflective of any paradigm shift toward a new and deeper understanding of psychopathology.

What motivates the hype and premature claims? There are three likely explanations. First, claiming too much too soon may reflect profound disappointment that our progress has been so slow-that our incredibly powerful neuroscience tools have so far met their match in our incredibly complicated brains. It may be intellectually comforting to trumpet our small beachhead of available knowledge rather than face squarely how large is the continent of the still unknown.

There may also be a felt need to clothe psychiatry in the authority of neuroscience-to show that we are as science based as the rest of medicine. Leaving aside the fact that all of medicine is having the same great difficulty in making translational giant steps, we should not be surprised or ashamed that psychiatry lags a bit behind, despite all of our intense and successful research efforts. The brain is an exponentially greater riddle to solve than are the pancreas or heart or immune system. If it is so difficult to figure out diabetes or atrial fibrillation or rheumatoid arthritis, why should we expect schizophrenia to tumble easily? Finally, clothing DSM-5 in fancy science dress may be seen as a way to justify its existence and to sell more books.

An interesting historical parallel from over a century ago presaged the current overselling of neuroscience as a guide to clinical practice. In the late nineteenth century, there was great excitement in neurology occasioned by the discovery of the neurone, of neural networks, and of the brain as an electrical machine. This basic neuroscience revolution afforded tremendous halo prestige to the clinical neurologists who then treated what today would be considered the less severe mental disorders.

The two most popular diagnoses that emerged from neuroscience theorizing were neurasthenia and conversion hysteria. Both were based on then plausible (but now clearly wrong) models of brain functioning and of psychopathology. Then, as now, the wonderful science advances explaining normal brain functioning could not explain psychopathology-but this did not prevent Charcot, Freud, and Beard from developing models that now seem quaintly ambitious in their overreach.

Let's not repeat this history of over promising and disappointing false starts. Psychiatry should live comfortably within its own skin, not make excessive claims. We are largely successful at doing what we do best in our current clinical work. We are eager to advance and incorporate the ever advancing scientific understanding of mental disorders and how best to treat them. But (except for Alzheimer disease), psychiatry is likely decades away from anything resembling a paradigm shift.

It always best modestly to under promise and then strive to over deliver. The sad tale of DSM-5 is a succession of overblown promises and then disappointing and potentially dangerous under performance. Psychiatry should work hard at what we do well-without reaching beyond our current grasp or raising expectations we can't possibly fulfill.