Acompanhamento Terapêutico

DSM in Philosophyland: Curiouser and Curiouser, Allen Frances *

*  é um dos organizadores da última versão do Manual Diagnóstico e Estatísticos de Transtornos Mentais (DSM-IV) .

      First off, thanks to James Phillips for inviting these stimulating commentaries. Second, a confession. My last (and only) formal training in philosophy was a freshman course in college that went well over my head. Now I have been invited to share my (probably sophomoric) speculations on the meanings that swirl below the surface of psychiatric classification. I do so without any confidence they can survive rigorous analysis by  those more expert than I in the tools of philosophic inquiry. Much of what I say below may be simple minded or simply wrong. What I do understand (perhaps better than anyone) are the practical issues of creating a psychiatric manual and the many good and bad (intended, unintended) consequences it can have. My views on deeper meanings are given, and should be taken, with a large grain of salt.

The Epistemologic Game

        First Umpire: “There are balls and there are strikes and I call them as they are.”

Second Umpire: “There are balls and there are strikes and I call them as I see them.”

Third Umpire: “There are no balls and there are no strikes until I call them.”

As I recall it, the three umpires are replaying a marathon epistemological game that: 1) began with Plato; 2) continued in the medieval joust between the realists and Occam’s nominalists; 3)was revived in the post-renaissance debate between Descartes and Vico on the power and limits of rational thought; 4)was refined by Kant; 5)churned up by Freud; and 6)finally settled by quantum physicists who have sharply downgraded the capacity of the human mind to ever fully intuit (much less understand) reality. Closer to my turf, I like to think of Bob Spitzer as umpire #1,  me as umpire #2, and Tom Szasz as umpire #3.

Spitzer’s achieved a paradigmatic revolution in psychiatric diagnosis and nosology. He introduced the method of diagnostic criteria (originally developed for research purposes) into a tool for general clinical practice. For the first time, psychiatrists could agree on diagnoses and make interpretive judgments across the research/ clinical interface. Certainly, the level of reliability achieved by DSM- III was over sold, especially when it was used by the average clinician. But DSM III was a huge leap forward from the useless and neglected guidance offered by DSM-I and DSM-II. It gave hope that psychiatry could become scientific and join in the advances that were being made in the rest of medicine.

DSM-III resulted from and promoted the victory of biological psychiatry over the psychological and social models that until then were its serious competitors. In the early dawn of its triumph, the biological model was presented with a realist, reductionist flourish that would have done umpire #1 proud. Mental disorders were real entities that existed  “out there.” The process of scientific discovery would elucidate their etiology and pathogenesis using the powerful new methods of neuroscience, imaging, and genetics.

The next section will focus on the disappointing fate of this ambitious program, but one  central point belongs here. Biological psychiatry has failed to produce quick, convincing  explanations for any of the mental disorders. This is because it has been unable to circumvent the fundamental and inherent flaw in the biological, “realist” approach – mental disorders don’t really live “out there” waiting to be explained. They are constructs we have made up – and often not very compelling ones at that. It has, for example, become clear that there is no one prototype “schizophrenia” waiting to be explained with one incisive and sweeping biological model. There is no gene, or small subset of genes, for “schizophrenia.” As Bleuler intuited, “schizophrenia” is rather a group of disorders, or perhaps better a mob. There may eventually turn out to be twenty or fifty or two hundred kinds of “schizophrenia.” As it stands now the definition and boundaries of “schizophrenia” are necessaarily arbitrary. There is no clear right way to diagnose this gang and not even much agreement on what the validators should be and how they should be applied. The first umpire was called out on strikes when the holy grail of finding the cause of “schizophrenia” turned out to be a wild goose chase.

Szasz is the third umpire. He quickly saw through the epistemological “no clothes” of umpire #1 and led the fight against simple minded biological reductionism (even well before the biologists had discovered their own voice and began making their overly ambitious and naïve claims). Szasz vigorously presented the view that mental illness is a medical “myth.”  Mental disorders were no more than social constructs that in some cases served a useful purpose, but in many others could be misused to exert a noxious social control, reducing freedom and personal responsibility. The biological “realists” reacted predictably to Szasz’ “nominalist” attack. They dismissed it. “If schizophrenia is a myth, they crowed, it is a myth that responds to medication and has a genetic pattern.” But their triumphalism was premature and based on both weak philosophic and weak scientific grounds. It turned out that the neuroscience, genetics, and treatment response of “schizophrenia” follow anything but a simple reductionist pattern. The more we learn about “schizophrenia” the more it resembles a heuristic, the less it resembles a disease.

This brings us to me (a call’um as I see’um) second umpire. In preparing DSM-IV, I had no grand illusions of seeing reality straight on or of reconstructing it whole clothe from my own pet theories. I just wanted to get the job done – i.e., produce a useful document that would make the fewest possible mistakes, and create the fewest problems for patients. Following Vico, I accepted that much in real life ( and almost everything in psychiatric classification) is overlapping, fuzzy, and heterogeneous – anything but Cartesian and amenable to overarching rationalist principles or mathematical precision. Psychiatric classification is necessarily a sloppy business. The desirable goal of having a classification consisting of mutually exhaustive, non-overlapping mental disorders is simply impossible to meet.

Instead, the second umpire follows a down-to-earth brand of Bentham utilitarian pragmatism. His umpire’s eye is fixed on the end result of getting to what works best – not distracted by biological reductionism or rationalist models of how things should be constructed. A diagnosis is a call to action with huge and unpredictable results. No decision can be right on narrow scientific grounds if it winds up hurting people.

Descriptive Psychiatry Gets Long of Tooth

The Dodo: “Everyone has run and everone has won and all must have prizes”.

      Modern descriptive psychiatry just passed its 200 birthday – if we measure it from the milestone of Pinel’s creation of the first psychiatric classification that resembles our own. His work was born from the Enlightenment belief in a rational world – some underlying order could be imposed even on the obvious irrationality of mental illness. The premise was that any domain receiving systematic observation and classification would eventually display causal patterns.

This approach was enormously successful in each of the major paradigm shifts in science. Always a careful description preceded a causal model. Kepler’s astronomical observations led toNewton’s gravity. Linnaeus’ classification of plants and animals led toDarwin’s  evolution. Mendeleyev’s periodic table led to Bohr’s structure of the atom. There have been dozens of descriptive systems vieing to describe things so brilliantly that their truth would shine forth. “All have run, but none has won prizes.” Descriptive classification in psychiatry has so far been singularly unsuccessful in promoting a breakthrough discovery of the causes of mental disorder.

This is doubly disappointing given the miraculous advances in our understanding of  normal  brain functioning. The advances in molecular biology, brain imaging, and genetics are spectacular – their impact on understanding psychopatholgy almost nil. Why the disconnect? The answer lies in a paraphrase of the opening lines of Anna Karenina. All normal brain functioning is normal in more or less the same way, but any given type of pathological functioning can have many different causes.

This is also true for all the complex diseases in medicine. A genetics company using  the Icelandic registry had tremendous success in finding gene markers for a dozen diseases, including schizophrenia. It recently went bankrupt because, in each instance, the particular candidate marker explained fewer than three per cent of the cases of the particular disease. There appear to be no common genes even for the common illnesses. Psychopathology is heterogeneous and overlapping not only in its presentation but also in its pathogenesis. There will likely be hundreds of paths to schizophrenia, not one or just a few and perhaps no final common pathway. Where does that leave the descriptive system of psychiatry? Fairly high and dry. Nature has obviously chosen to deprive us of clear joints, ripe for carving. There is little indication of any imminent and sweeping etiological breakthrough. Everything points towards a slow and painstaking retail accumulation of explanatory power. It is not even clear that the DSM categorical approach is the best research tool. The NIMH is embarking on a project to correlate an integrated exploration of neural networks with psychopathology. They chose to study dimensions of behavior (e.g. anxiety, pleasure seeking, executive functioning)  – not with the standard psychiatric disorders which are deemed too complex to have any simple relationship with a given neural network. Our DSM categories may not lead the future charge in understanding psychopathology.

Our descriptive classification of disorders is old and tired. It has worked hard for us and  continues to have many valuable and irreplaceable functions (which we will discuss in the last section). Fiddling needlessly with the labels will not advance science and may actually do more harm than good in its effect on clinical care.

The Elusive Definition of Mental Disorder

Humpty Dumpty: “When I choose a word it means just what I choose it to mean.”

      When it comes to defining the term “mental disorder” or figuring out which conditions qualify, we enter Humpty’s world of shifting, ambiguous, and idiosyncratic word usages. This is a fundamental weakness of our field. Many crucial problems would be much less problematic if only it were possible to frame an operational definition of mental disorder that really worked.

Nosologists could use it to guide decisions on which aspects of human distress and malfunction should be considered psychiatric – and which should not. Clinicians could use it when deciding whether to diagnose and treat a patient on the border with normality. A meaningful definition would clear up the great confusion in the legal system where matters of great consequence often rest on whether a mental disorder is present or absent.

Alas, I have read dozens of definitions of mental disorder (and helped to write one) and I can’t say that any have the slightest value whatever. Historically, conditions have become mental disorders by accretion and practical necessity, not because they met some independent set of operationalized definitional criteria. Indeed, the concept of mental disorder is so amorphous, protean, and heterogeneous that it inherently defies definition. This is a hole at the center of psychiatric classification.  And the specific mental disorders certainly constitute a hodge podge. Some describe short term states, others lifelong personality. Some reflect inner misery, others bad behavior. Some represent problems rarely or never seen in normals, others are just slight accentuations of the everyday. Some reflect too little control, others too much. Some are quite intrinsic to the individual, others are defined against varying and changing cultural mores and stressors. Some begin in infancy, others in old age. Some affect primarily thought, others emotions, yet others behaviors, others interpersonal relations, and there are complex combinations of all of these. Some seem more biological, others more psychological or social. If there is a common theme it is distress and disability, but these are very imprecise and nonspecific markers on which to hang a definition.

Ironically, the one definition of mental disorder that does have great and abiding practical meaning is never given formal status because it is tautological and potentially highly self serving. It would go something like “Mental disorder is what clinicians treat and researchers research and educators teach and insurance companies pay for.” In effect, this is historically how the individual mental disorders made their way into the system.

The definition of mental disorder has been elastic and follows practice rather than guides it. The greater the number of mental health clinicians, the greater the number of life conditions that work their way into becoming disorders. There were only five disorders listed in the initial census of mental patients in the mid nineteenth century, now there are close to three hundred. Society also has a seemingly insatiable capacity (even hunger) to accept and endorse newly defined mental disorders that help to define and explain away its emerging concerns. As a result, psychiatry is subject to recurring diagnostic fads. Were DSM-V to have its way we would have a wholesale medicalization of everyday incapacity (mild memory loss with aging); distress (grief, mixed anxiety depression); defects in self control (binge eating); eccentricity(psychotic risk); irresponsibility (hypersexuality); and even criminality (rape, statutory rape).

Remarkably, none of these newly proposes diagnoses even remotely pass the standard loose definition of “what clinician’s treat.” None of these “mental disorders” has an established treatment with proven efficacy. Each is so early in development as to be no more than “what researchers research” – a concoction of highly specialized research interests.

We must accept that our diagnostic classification is the result of historical accretion and accident without any real underlying system or scientific necessity. The rules for entry have varied over time and have rarely been very rigorous. Our mental disorders are no more than fallible social constructs (but nonetheless useful ones if understood and applied properly).

The Conservative/Innovation Debate or Where Have All the Normals Gone?

Alice: “But I don’t want to go among mad people

Cheshire Cat: Oh, you can’t help it,we’re all mad here”