How does the difference between objective evidence and subjective evidence relate to the practice of psychiatry?
July 2006, Vol. XXIII, Issue 8
Several recent statements from various Hollywood celebrities and authors in the mental health field have raised serious questions about psychiatry's “objectivity.” These very public dust-ups have merely highlighted an ignoble tradition of psychiatry bashing that goes back to the early 1960s.
What do philosophers mean when they speak of “subjective” and “objective” evidence? According to philosopher and economist Amartya Sen, they often mean something like this: when I say, without having observed your house, “I truly and deeply believe that your house is on fire,” I am making a subjective claim. In contrast, if 2 people simultaneously witness what they believe is smoke coming from your house, and say, “We believe your house is on fire,” they are making a type of objective statement. This does not necessarily mean that your house is on fire--after all, someone inside might have been producing a gray vapor of some sort that merely resembles smoke. Thus, the veridical, nature or “truth value,” of objective statements cannot automatically be assumed. But whereas, in theory, we might choose to run that gray vapor through an electronic “vapor analyzer” at some point, we surely would not hesitate to call 911 immediately.
How does all this apply to clinical psychiatry? For philosophers such as Sen, if I say, without having assessed you, “I believe deep in my heart that you have a thought process disorder,” I am making an essentially subjective claim. On the other hand, if my chief resident and I sit with you for an hour, attending carefully to your speech, and neither of us has a clue as to what you have been saying, we are beginning to develop an “objective” frame of reference. If both the resident and I can point to your use of frequent neologisms and unconventional syntax, as well as to your shifting from idea to idea within the same sentence, we are continuing to develop an “objective” basis for saying that you have a thought process disorder of some sort--our agreement being a modest example of “interrater reliability.” If, on standardized neurolinguistic testing, we can confirm that, indeed, your use of grammar, syntax, logic, and concept formation are all abnormal, we have further objective evidence of a thought process disorder.
Does all this mean that you have a "mental illness" or a “disease” of any kind? Of course not. To determine that, we need a much broader construct than that of thought process disorder, and a much wider array of objective data. In philosophical terms, we need many more observations that can be confirmed by multiple observers. For example, the resident and I may agree that a patient with thought disorder also has a blunted or flat affect (based, in part, on our shared experience of “normal” affect). We also observe that he is muttering to himself when sitting alone. We may learn, from a spouse or family member, that the patient has not been showering, feeding himself, or changing his clothes for the past month, and that this represents a dramatic change in his usual behavior.
We may learn, from both the patient and the family member, that he is "hearing voices" when nobody is in the room, and that these voices are telling him, “You deserve to burn in Hell for all eternity!” We may also learn that, on more than one occasion, the patient has broken into tears and slashed himself with a razor, upon hearing these “voices.” At this point, we have built a stronger objective case for saying that the patient has an illness or disease of some sort--perhaps a schizophreniform or psychotic depressive disorder, although other possibilities must be considered (eg, an endocrine disturbance, a brain tumor, a dementing process).
If, in addition, we can find abnormalities on neurologic testing, brain imaging, or laboratory testing, so much the better. (It is clear, on this view, that “objectivity” is not an all-or-none quality, but one that exists along a continuum of evidence.) With such testing, we certainly would be strengthening our objective database--but laboratory studies are not necessary for objectively claiming, in the first place, that the patient has a “disease” of a psychiatric nature.
Historically, the construct of “disease” developed from the notion of suffering and incapacity in the absence of an obvious external cause, such as a knife wound. The term disease was originally derived from the notion of “dis-ease” or discomfort. It is no coincidence that the word patient is derived from the Latin pati, meaning “to suffer” or “to bear.” As physicians, we first recognize dis-ease and treat suffering, based on our clinical (from Gk. kline, “bed”) observations. In general, it is only subsequently that we invoke imaging or laboratory studies to bolster or confirm our diagnosis.
As Schwartz and Wiggins have argued, “[W]e legitimately reason about people's experiences and behaviors in the same manner that we might reason about breathing problems . . . such reasoning might--or might not--lead us to lesions. But the goal is the relief of suffering, the promotion of health, and the amelioration of the illness.” Of course, a neuropathologist might not offer a diagnosis of neuroborreliosis until the organism causing Lyme disease has been reliably identified. But even a neuropathologist would not deny that a patient who complained, without dissimulation, of profound memory impairment, disorientation, visual hallucinations, and paresthesias had a disease of some kind--much less deny that this person merits our care and treatment.
Indeed, when the average neurologist diagnoses, say, migraine headaches, he or she rarely uses laboratory or imaging studies, except to rule out other disease entities, such as a CNS lesion. Rather, the neurologist relies, in the first place, on the patient's subjective (or phenomenologic) claims: eg, “Doc, I get a persistent, throbbing, left-sided pain in my head, along with nausea and sensitivity to light.” This claim is then weighed in the context of objective data derived from the medical history and neurologic examination. This process is not radically different from the holistic approach a psychiatrist takes in diagnosing schizophrenia or major depression. Nor does it differ from the way in which most emergency room physicians would make a presumptive diagnosis of angina pectoris, even if the patient's ECG were normal. That some “researchers” may successfully fool clinicians by presenting bogus complaints of hallucinations, headache, or chest pain does not impugn the objective basis of medical diagnosis. Neither does the sad fact that some of our colleagues fail to gather a sufficiently detailed set of phenomenologic and objective data before they make a diagnosis or offer treatment.
In psychiatry, as in the rest of clinical medicine, it is the patient's unique experience of suffering and incapacity--not a blood test or MRI result--that first prompts diagnosis and treatment. Indeed, we would do well to remind ourselves of that famous dictum from Maimonides: “The physician does not cure a disease, he cures a diseased person.”
Dr Pies is clinical professor of psychiatry at Tufts University. His most recent books include Creeping Thyme, a Collection of Poetry (Brandylane Publishing); Zimmerman's Tefillin, a Short Story Collection (PublishAmerica); and Handbook of Essential Psychopharmacology, 2nd ed, (American Psychiatric Publishing).
The author wishes to thank Dr Scott Lilienfeld and Dr Michael A. Schwartz for their seminal contributions in the area of psychiatric diagnosis, and expresses appreciation to Dr Daniel Pistone for his stimulating ideas.
Spitzer RL, Lilienfeld SO, Miller MB. Rosenhan revisited: the scientific credibility of Lauren Slater's pseudopatient diagnosis study. J Nerv Ment Dis. 2005;193:734-739.
Slater L. Opening Skinner's Box: Great Psychological Experiments of the Twentieth Century. New York: WW Norton; 2004.
Slater L. Reply to Spitzer and colleagues. J Nerv Ment Dis. 2005;193:743-744.
Lilienfeld SO, Spitzer RL, Miller MB. A response to a nonresponse to criticisms of a nonstudy: one humorous and one serious rejoinder to Slater. J Nerv Ment Dis. 2005;193:745-746.
Pies R. Psychiatry clearly meets the “objectivity” test: viewpoints. Psychiatr News. 2005;40:17.
Pies R. Moving beyond the “myth” of mental illness. In: Schaler JA, ed. Szasz Under Fire. Chicago: Open Court; 2004:327-353.
Sen A. Objectivity and Position: Assessment of Health and Well-being. Paper presented at: Harvard University, December 11, 1990. Available at: http://www.grhf.harvard.edu/frame6/frame6.html. Accessed June 1, 2006.
Szasz TS. The Myth of Mental Illness. New York: Harper & Row; 1974.
Nagel T. The View From Nowhere. New York, Oxford: Oxford University Press; 1986.
Schwartz MA, Wiggins OP. Psychiatry fraud and force? A commentary on E. Fuller Torrey and Thomas Szasz. J Human Psychol. 2005;45:403-415.