The Lure of the Context–Dependent Psychiatric Diagnosis

… the king grew vain;

Fought all his battles o’er again

And thrice he routed all his foes,

And thrice he slew the slain.

—Alexander’s Feast, John Dryden

In “Major Depression After Recent Loss Is Major Depression—Until Proved Otherwise” (Psychiatric Times, December 2008, page 12), Dr Pies highlights one of the more provocative questions encountered when we train in clinical psychiatry: “Suppose your new patient Mr Jones, tells you he is feeling ‘really down.’ He meets all DSM–IV symptomatic and duration criteria for a major depressive episode (MDE) after having lost his wife to cancer 2 weeks ago. Should you diagnose MDD?”

He then remarks, “I’m guessing most psychiatrists would be reluctant to do so. Indeed, DSM–IV permits a ‘bereavement exclusion’ in such circumstances, provided the patient lacks ‘certain symptoms that are not characteristic of a ‘normal’ grief reaction.’”

Why such contention about diagnosing depression based on a rich symptom presentation? Why is the descriptive syndromal approach questioned? Hasn’t this battle been waged and decided? Does one need, as does the besotted King in Alexander’s Feast, to slay the already thrice slain?

The immediate reason, as put forth by Dr Pies, is the recent attention paid to a commentary by Allan V. Horwitz, PhD, and Jerome C. Wakefield, PhD (“An Epidemic of Depression,” Psychiatric Times, November 2008, page 44), authors of The Loss of Sadness: How Psychiatry Transformed Normal Sorrow Into Depressive Disorder.¹ These professors, citing Pies, advocate a context–laden or context–driven approach to psychiatric diagnosis. This approach already has several names, including a biopsychosocial approach and a Meyerian approach. In contrast, a content–poor (or formal) approach only stresses empirical, phenomenological, or descriptive psychiatric strategies. These decontextualized approaches bring no theory of mind to the diagnostic setting (Freudian, behavioristic, evolutionary, and others) and are therefore referred to as atheoretic.

Of course, there is no absolute dichotomy proposed or possible, but the two approaches are markedly different. For example, consider a patient who experiences auditory hallucinations of a derogatory nature in the second person. Using a contextualized approach, the treating psychiatrist might try to read meaning and cause into the content of the hallucination. However, a treating psychiatrist who uses a formal approach would not credit this or any interpretation, because such interpretations are not licensed by any empirical or therapeutic data. A particularly important point is that a context–dependent approach to diagnosis hardly exhausts causation and, at our present state of ignorance, perhaps engages none. And of what immediacy would we attribute folk interpretations of such content to a patient with second–person command auditory hallucinations who might be in withdrawal, febrile, or subject to neuroimmunological insults, of which we are dimly aware?

In his commentary, Dr Pies provides a tactful, considerate, yet unqualified rejection of the supposed value (and evokes consideration that there may be a false value) of using the contextualized approach in a psychiatric strategy when making a diagnosis and determining treatment. Indeed, Dr Pies’ commentary entails the very real possibility that the contextualized approach, with its inherent antinosological and antitreatment bias, favors withholding of treatment in many situations, particularly early in a clinical course seemingly heralded or triggered by an environmental “psychosocial cause,” although data in support of such triggers remains wanting. But why is the descriptive syndromal approach being questioned again? Why resurrect the contextualized diagnostic approach?

We see similar failings in progress in many areas of general medicine, yet we do not hear a clarion call to prior failed therapies and theories. We don’t go back to humoral theory in medicine or to bleeding a febrile patient. Those who practice internal medicine continue to look for basic causations that are currently beyond our understanding. Using reason and experience we hypothesize and then go out and test the hypothesis (we take measurements) against nature and refine and retest accordingly.

But this does not appear to be the case in psychiatry in which a prior and very much unsatisfactory if not frankly disproven approach is re–heralded. And this is more disquieting because in psychiatry almost nothing is known and all we have is the scientific method to use to explore encouraging ideas, which may turn into testable hypotheses and, over time, valid theories of illness.

The reason for the resurrection of context–driven psychiatry is very powerful (and far from puzzling). It is the result of an often–irresistible temptation, pull, or lure of the ego narrative that puts us at the center of life’s drama and employs all of folk psychology.

Donald Klein² spoke to this nearly 40 years ago:

The wishful belief that the descriptive diagnosis has been superseded by a desirable dynamic etiological viewpoint has proved unfounded. If anything, this viewpoint has simply succeeded in so blurring diagnostic distinctions and so undermining the interest of current psychiatrists in these distinctions that it is positively difficult to reestablish the value of descriptive characterization without appearing a hidebound reactionary.

Hagop Akiskal recorded Klein’s comment in an important chapter on psychiatric diagnosis.² In this chapter, Akiskal also cites Frank Fish (acknowledged by Akiskal as the leading authority on phenomenological approaches to psychiatry); the quote warns specifically against conflating observation and theory—a marriage that Wakefield–Horwitz’s theory (un)happily promotes: Frank Fish “taught that observation had to be distinguished from theorizing and inference, for without that, all the rest of psychiatry, from biochemistry and genetics to sociology and psychodynamics, float in the air without a base.”²

And finally, one further citation from the same chapter speaks to the antidiagnostic bias and antitreatment danger of the context–rich approach. Frank Miele wrote, “It is common knowledge that American psychiatry and clinical psychology, the former under psychodynamic influence, the latter under both psychodynamic and learning theory, have an animus against formal diagnosis.”²

These are voices from authorities on diagnostic theory from our own profession more than 35 years ago. But at mid–20th century, there was also a red flag against narrative context–dependent diagnosis raised by Gilbert Ryle, a reigning authority on the philosophy of mind, who coined the term “the ghost in the machine.” The warning specifically regards issues of personal identity and therefore holds particular import for psychiatric thinking. Ryle wrote, “gratuitous mystifications begin from the moment that we start to peer around for beings named for our pronouns.”³

Ryle was hardly the first post–Cartesian to caution us against the role of a contextualized psychiatric approach based on the concept of a well–defined ego or personal pronoun. Spinoza famously stated 300 years earlier, “experience and reason accord to establish that men believed themselves free only because they are conscious of their actions and not of what determines them.”⁴

Our forebears have waged a battle to defeat the false access to understanding of behavior predicated on context–rich formulations foundered on mystical beings. This battle has been more than twice told and more than thrice resolved. In Alexander’s Feast, the tired and besotted king was called on to retell his martial victories. The lure of the contextualized psychiatric approach, with its reliance on folk and folksy psychology and appeal to our supposed privileged access to our own nature, will continue to exert its pull, and credit is given to Dr Pies for helping slay the slain yet again.

Karen Venuto New York

Andrew Lautin, MD New York

Ms Venuto is a graduate student in public health at New York University. Dr Lautin is clinical professor of psychiatry at New York University School of Medicine.

References

1. Horwitz AV, Wakefield JC. The Loss of Sadness. New York: Oxford University Press; 2007.
2. Akiskal HS, Webb WL. The joint use of clinical and biological criteria for psychiatric diagnosis. In: Akiskal HS, Webb WL, eds. Psychiatric Diagnosis: Exploration of Biological Predictors. New York: Spectrum Publications; 1978.
3. Ryle G. The Concept of Mind. Chicago: University of Chicago Press; 2002.
4. de Spinoza B. Ethics. Curley E, trans. New York: Penguin Books; 1996.

Dr Pies responds:

I thank Ms Venuto and Dr Lautin for their thoughtful and philosophically sophisticated analysis. I believe their position is generally very close to that of my editorial. However, I do wish to clarify some aspects of my argument against the Wakefield–Horwitz thesis and perhaps also to suggest some subtle differences between my claims and the conclusions of Ms Venuto and Dr Lautin.

In the example of the patient who “experiences auditory hallucinations of a derogatory nature,” Ms Venuto and Dr Lautin are correct in noting that “using a contextualized approach, the treating psychiatrist would possibly try to read meaning and cause into the content of the hallucination” in contrast to what they call a “formal” (ie, descriptive–observational) approach. Nonetheless, I believe it is sometimes helpful to hypothesize “meaning” and “cause” when we are evaluating a patient with delusions or hallucinations. I merely object to the notion (implicit in the Wakefield–Horwitz thesis) that such contextual hypotheses are relevant to the determination that “disorder” or “disease” is present.

It may turn out that the patient’s derogatory hallucinations are indeed etiologically related to “a parental figure derogating the patient” as well as to excessive dopamine(Drug information on dopamine) in his limbic system and a host of genetic, environmental, and social factors. However, in my view, these causal considerations are not relevant to our determination that the patient manifests clinically significant disease or disorder. If there is substantial and prolonged suffering and incapacity associated with the patient’s hallucinations, then the patient has dis–ease and warrants professional treatment—regardless of “context.”

Moreover, I by no means reject the possibility that etiological hypotheses may be useful in the psychotherapeutic treatment of the patient. For example, there might well be a role for exploring the patient’s early traumatic experiences as a contributing factor in the genesis of auditory hallucinations. Of course, etiology becomes vitally important when there is a reversible medical or neurological cause for the patient’s symptoms (eg, alcohol(Drug information on alcohol) withdrawal, cocaine intoxication, etc), as Ms Venuto and Dr Lautin recognize. Indeed, there is no sharp dividing line between a descriptive–observational diagnosis and an interpretive–contextual diagnosis, to the extent that the latter is informed by careful empirical observation. Thus, the patient who reports derogatory auditory hallucinations and who also shows tachycardia, hypertension, and reactive dilated pupils may merit the “contextual interpretation” of “probable cocaine– or amphetamine–related hallucinosis.”

Problems arise—as Venuto and Lautin clearly appreciate—when we are “lured” into premature diagnostic closure by seductive but nonverifiable “narrative” explanations of the patient’s suffering; eg, “Mrs Jones tells me she is depressed because she lost her husband to cancer 2 months ago, and this is clearly the genesis of her problem.” Even more serious problems arise, as they do in the Wakefield–Horwitz thesis, when we create dichotomous “disordered”/“non–disordered” classifications based on such speculative (and sometimes spurious) narratives—what Venuto and Lautin aptly describe as “folksy psychology.” As I argue elsewhere, the clinician needs to preserve “diagnostic neutrality” in the early phases of the doctor–patient relationship, lest he or she be lured into premature diagnostic closure.1 As Venuto and Lautin recognize, claims about the etiology of the patient’s symptoms require “testable hypotheses” that are confirmed or disconfirmed as empirical evidence emerges in the therapeutic relationship.

In this regard, however, I believe that more is “known” in psychiatry than Ms Venuto and Dr Lautin acknowledge. For example, we may not know the biomolecular correlates of mania, but we do know a great deal about mania’s phenomenology. By this term, I do not have in mind merely the classic signs and symptoms of mania (decreased sleep, pressured speech, etc) but also the patient’s “felt experience” of mania—the sense of Godlike power, inflexible entitlement, superhuman wisdom, and so on, that constitute the inner world of the patient in a manic episode. These data are no less “real”—indeed, by some lights, no less “objective”—than data gathered by internists or rheumatologists.² As I suggest in another essay,³ the long–term goal of psychiatric nosology should be to integrate this phenomenological knowledge with epidemiological, biochemical, and genetic knowledge— ultimately striving for what my colleague Nassir Ghaemi calls (referring to psychiatrist Karl Jaspers) a “biological existentialist” approach to the nature of psychiatric disease.⁴

Finally, I agree with Venuto and Lautin that an “animus against formal diagnosis” in psychiatry poses potential risks for our patients. It sometimes seems that this animosity is founded on the mistaken belief that formal diagnosis inevitably “reduces” the patient’s problems to some trivial bit of aberrant biochemistry. In truth, the etymology of “diagnosis”(“dia” from across, between; “gnosis” from knowledge) points to its true nature: knowing the difference between one condition and another. Diagnosis does not diminish the patient’s humanity; it provides the basis for healing and recovery.

Ronald Pies, MD Boston

Dr Pies is professor of psychiatry and lecturer on bioethics and humanities at SUNY Upstate Medical University in Syracuse and clinical professor of psychiatry at Tufts University in Boston.

References

1. Pies R. Narratives, normality and diagnostic neutrality. Psychiatr News. In press.
2. Pies R. Psychiatry clearly meets the “objectivity” test. Psychiatr News. 2005;40:17
3. Pies R. The anatomy of sorrow: a spiritual, phenomenological, and neurological perspective. Philos Ethics Humanit Med. 2008;3:17. http://www.peh–med.com/content/3/1/17. Accessed May 4, 2009.
4. Ghaemi SN. On the nature of mental disease: the psychiatric humanism of Karl Jaspers. Existenz. 2008;3:1–9.