Philosophy, Psychiatry, & Psychology

With attention turning to the process of formulating the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-V; e.g., Kendler et al. 2008), the study by Flanagan and Blashfield (2007) of the similarities and differences between clinicians’ “folk” taxonomies and psychiatry’s official one is timely, and its lessons are in need of analysis. In this commentary I first address questions raised by the existence of multiple de facto taxonomies, and by the characteristics of those taxonomies themselves, before focusing attention on the DSM diagnostic system and its attributes that make acceptance by its constituencies far less than complete. Although the aim of Flanagan’s and Blashfield’s study is explicitly descriptive rather than normative, it provides an important opportunity to examine the extent to which our taxonomies serve the purposes for which they were constructed—whether by experts or by clinicians.

There are certainly reasons to expect significant deviations between clinicians’ everyday taxonomies and the official one. Perhaps the most glaring reason is the fact that—even if it were desirable—no one can be familiar with the enormous number of categories codified by the DSM. On the other hand, the remarkable hegemony of the DSM—in the education and training of physicians and clinical psychologists, in clinical communication, and in the ubiquitous administrative functions of billing and coding—makes such deviations at least somewhat surprising and in need of scrutiny and explanation. The first question, however, is whether one should be concerned by the existence (surprising or otherwise) of multiple taxonomies. Flanagan and Blashfield make the compelling case that, given the high rate at which Americans seek clinical attention for psychiatric problems, the stakes are commensurately high for our efforts at categorizing those problems in valid and useful ways. And the extent to which a gulf exists between the way psychiatry and clinical psychology are written about and taught on the one hand, and conceptualized and practiced by clinicians on the other, invites a multitude of problems—among which cynicism about our disciplines may be one of the most pernicious.

At least as important as their finding of only moderate correlations between clinicians’ taxonomies [End Page 271] and that of the DSM are the characteristics of clinicians’ diagnostic systems themselves. One potentially major artifact of their methodology, however, must be borne in mind: the instruction to the clinicians in their study to “[p]ut together the diagnoses that have similar treatments…” (emphasis mine). Although they cite previous work (Flanagan and Blashfield 2006) in defense of the proposition that such instruction did not unduly influence their findings, it was certainly an avoidable potential problem, which itself likely reflects the history of our clinical disciplines in which the treatment cart frequently pulled the diagnosis horse. It is, therefore, not unlikely that clinicians’ taxonomies might emphasize treatment considerations even if they had not been directed to do so. The fallacious reasoning behind such intuitively appealing practice, however, must be exposed. The implicit assumption that treatment response is a proxy for other important characteristics of illnesses is predicated on the false proposition that our treatments exert their effects via similar mechanisms across the various indications for their uses, thus allowing us to infer from such responses that what was awry in the first place among the conditions for which a given treatment is effective must be closely related. Categorizing mania and trigeminal neuralgia together by virtue of their shared responses to anticonvulsant therapy, or considering autoimmune platelet deficiencies to be closely related to abdominal trauma due to the fact that each sometimes requires splenectomy, are examples of the potential results of acting on that false premise.

Another parameter on which the clinicians in the current study based their nosologies—most explicitly by the “dimensional experts”—is the “environmental” versus “biological” one. Given that a coherent dichotomy would either distinguish environmental from genetic etiological factors, or would contrast psychological (i.e., subjective or first-person) from biological (i.e., objective or third-person) manifestations of illness, a taxonomy one of whose axes purports to represent an environmental/biological continuum cannot...