Beyond DSM-5, Psychiatry Needs a “Third Way”

February 9, 2012
Ronald W. Pies, MD
Ronald W. Pies, MD

Dr Pies is Professor Emeritus of Psychiatry and Lecturer on Bioethics and Humanities, SUNY Upstate Medical University; Clinical Professor of Psychiatry, Tufts University School of Medicine; and Editor in Chief Emeritus of Psychiatric Times (2007-2010). Dr Pies is the author of several books. A collection of his works can be found on Amazon.

The recent spate of Op-Eds in the New York Times says it all: both the psychiatric profession and the general public have strong feelings about the pending DSM-5-what many in the media like to call “Psychiatry’s Bible.” These emotions are certainly understandable.

The recent spate of Op-Eds in the New YorkTimes says it all: both the psychiatric profession and the general public have strong feelings about the pending DSM-5-what many in the media like to call “Psychiatry’s Bible.” These emotions are certainly understandable. Apart from the many ethical and philosophical problems raised by changes in psychiatric diagnosis, there are millions of dollars at stake: depending on how the DSM-5 evolves, individuals with major depression, Asperger’s syndrome, ADHD, and a number of other diagnoses may or may not be eligible for coverage by third-party payers.

Most of the controversy over the DSM-5 has focused on “boundary” issues: where do we draw the line between the introspective and the autistic individual, or between the bereaved widow experiencing ordinary grief and the one with a major depressive episode? But most of the vociferous critics of the DSM-5 have been nibbling around the edges of the diagnostic system as a whole. Few have argued that the entire DSM approach is in need of a drastic overhaul.

The DSMs over the past 30 years have attempted to “carve Nature at its joints”-a metaphor first used by Plato in his Phaedrus. By requiring certain necessary and sufficient conditions for diagnosing a disorder, the DSMs have provided researchers with a means of ensuring “inter-rater reliability”-in effect, making sure that those investigating schizophrenia or bipolar disorder have identified subjects with the same condition. This categorical approach is based on the notion that there is an “essence of schizophrenia” or an “essence of panic disorder,” much like the Platonic “forms” that defined the nature of triangles or spheres. Critics of this approach have argued that “Nature” may not have “joints” and that human disease states are too varied and complex to conform to essentialist definitions. Call this the “anti-pigeonhole” critique of the DSMs.

Some of these critics have argued for a “dimensional” approach to diagnosis, in which patients are rated on a number of different measures of psychopathology; for example, Mr Jones might be categorized as having high anxiety, moderate depression, and minimal cognitive disorganization. Proponents of the dimensional approach believe it allows clinicians to address the patient’s problems in a more targeted and fine-grained way, without reifying the patient’s condition as a specific disease entity. But a dimensional approach makes it much harder to group patients together for research purposes, and nearly impossible to exchange “shorthand” information between clinicians. After all, it’s much easier to tell a colleague, “I saw an elderly man with bipolar I disorder” than to break down the patient’s problem into 4 or 5 symptom dimensions.

One thing we know for sure: many clinicians routinely ignore the present DSM system. For example, a study by Dr Mark Zimmerman and colleagues found that nearly 25% of psychiatrists used the DSM-IV criteria to diagnose major depression less than half the time-and this was true of more than two-thirds of the non-psychiatrist physicians studied. Many clinicians find the cut-and-dried DSM criteria too superficial to capture the nuances of the patient’s condition. Others (this author among them) complain that the DSM embodies the worst of both worlds: it lacks validated biochemical “markers” for the major disorders, while also ignoring the kind of depth-psychology emphasized by psychoanalysts and existential therapists. Furthermore, the convoluted “exclusion rules” demanded by the DSM criteria-for example, in applying the “bereavement exclusion” for major depression-are often too confusing for most clinicians.

It’s too late now to do very much about the DSM system: the DSM-5 will probably be coming our way within the next year or two, warts and all. But we can begin to think beyond the DSM-5, and beyond “categorical” versus “dimensional” models: we can begin to consider a “third way.”

Decades ago, the psychiatrist Karl Jaspers (1883-1969) advocated the use of so-called ideal types in psychiatric diagnosis. Similarly, in recent years, other clinicians have argued for the use of “prototypes”-basically, narrative descriptions of typical disease presentations, sometimes in the language patients themselves use to describe their experience (ie, phenomenological prototypes). Prototypes-unlike the rigid DSM categories-have “fuzzy” boundaries. They try to capture the core features of an illness like schizophrenia without specifying an “essence,” or a list of necessary and sufficient criteria. In this sense, they reflect developments in the philosophy of language first described by Ludwig Wittgenstein (1889-1951).

Rather than proffering a Platonic “essence” to define a category, Wittgenstein referred to “family resemblances”-like blue eyes and blond hair in 3 of 5 members of the Jones family, and tallness in 4 of the 5. No single feature invariably identifies a Jones family member, but the “blue eyes, blond hair, and tall” prototype helps us do so. My colleague, Dr Nassir Ghaemi, and I used a modified prototype approach in developing a screening tool-the Bipolar Spectrum Diagnostic Scale (BSDS)-for identifying bipolar spectrum disorders. Rather than relying on a fixed set of criteria, the BSDS presents a kind of narrative “story” to the patient, who is then asked how closely the bipolar prototype fits his or her condition. The BSDS has proved useful in identifying patients on the “softer” end of the bipolar spectrum. Similarly, I have recently used the prototype method to develop a potential screening tool-the Post-Bereavement Phenomenology Inventory (PBPI)-for distinguishing ordinary grief from clinical depression.*

Of course, insurance companies are bound to oppose the use of prototype diagnoses-after all, it’s much easier to deny (or approve) a claim based on a hard-and-fast set of criteria. We can always retain such strict criteria in a separate manual, designed for researchers, or as an appendix to a prototype-based manual. But psychiatry needs to move beyond the DSM system, and beyond the perennial categorical versus dimensional debate. I believe that work-a-day clinicians will welcome a “third way”-one that allows us to appreciate the fluidity, complexity, and depth of our patients’ problems.

For further reading

Ghaemi, SN, Miller CJ, Berv DA, et al. Sensitivity and specificity of a new bipolar spectrum diagnostic scale. J Affect Disord. 2005;84:273-277.

Pies R. Why psychiatry needs to scrap the DSM System. PsychCentral.

*Pies R. After Bereavement: Is It “Normal Grief” or Major Depression? (In process: Submitted to Psychiatric Times).

Schwartz MA, Wiggins OP, Norko MA. Prototypes, ideal types, and personality disorders: the return to classical psychiatry. J Pers Disord. 1989;3:1-9.

Vzquez GH, Romero E, Fabregues F, et al. Screening for bipolar disorders in Spanish-speaking populations: sensitivity and specificity of the Bipolar Spectrum Diagnostic Scale-Spanish Version. Compr Psychiatry. 2010;51:552-556.

Wittgenstein L. The Blue and Brown Books. New York: Harper Torchbooks; 1965.

Zimmerman M, Galione JN, Ruggero CJ, et al. A different approach toward screening for bipolar disorder: the prototype matching method. Compr Psychiatry. 2010;51:340-346.

Zimmerman M, Galione J. Psychiatrists’ and nonpsychiatrist physicians’ reported use of the DSM-IV criteria for major depressive disorder. J Clin Psychiatry. 2010;71:235-238.