Ronald Pies

Ronald Pies

Hello, I must be going.
I cannot stay,
I came to say
I must be going.
I’m glad I came
but just the same
I must be going.

–Lyrics from  a song from the Marx Brothers’ 1930 film Animal Crackers, written by Bert Kalmar and Harry Ruby

Peter Kinderman

Peter Kinderman

I very much appreciate Prof. Kinderman’s gracious response to my earlier criticisms, and I believe that on several points, we are largely in agreement. We agree, first, that controlled trials of various diagnostic (or non-diagnostic) approaches to mental health treatment are needed. Indeed, I am not wed to a DSM-type or categorical approach at all; on the contrary, I have argued that the best diagnostic system would be that which most consistently and effectively reduces the emotional suffering–and enhances the lives–of those we treat. I have referred to this criterion as “instrumental validity” [see].

Prof. Kinderman is also correct in arguing that the greater the heterogeneity in a particular diagnostic category, the less useful–in general–it proves to be. However, this is a problem throughout all of clinical medicine. Thus, a diagnosis of “myocardial infarct” (heart attack) conveys a wide range of possible outcomes, depending on the size of the infarct, its location, the person’s pre-infarct cardiac function, etc. But when a physician hears that a patient has just had an “MI”, he or she still knows certain useful things about the patient.

So, too, with psychiatric categories like schizophrenia or even the very heterogeneous major depressive disorder (MDD). Thus, an article just published in the American Journal of Psychiatry finds

that persons diagnosed with MDD (by DSM-IV criteria) showed work impairment that was directly related to the “burden of illness” (severity, duration,etc); and furthermore, that work productivity improved with acute phase antidepressant treatment. So, even this very heterogeneous category of MDD showed predictive validity, in this study (Trivedi et al, Am J Psychiatry 2013;170:633-641).

Furthermore, it is not the case, with all DSM-5 diagnoses, that two persons with the same diagnosis can have, as Prof. Kinderman opines, “absolutely nothing in common.” To be sure, this depends partly on what we mean by “in common.” For example, in order to be diagnosed with Bipolar I (BP-I)disorder, the patient must have had at least one manic episode–and this is carefully defined. So, 1000 patients with BP-I will all have a history of a manic episode in common.As with all polythetic constructs, the particular symptoms of a manic episode will differ from patient to patient.

Similarly, all patients with a diagnosis of Panic Disorder–100%–must meet DSM-5 criterion A; i.e., “recurrent, unexpected panic attacks,” even though the precise symptoms–dizziness, trembling, etc–will differ from patient to patient. Of course, as we learn more and more about these conditions, we hope to refine our criteria so as to generate more homogeneous categories, with greater predictive validity. In some cases, this may actually lead to the elimination of some diagnostic subtypes; e.g., DSM-5 has eliminated the various subtypes of schizophrenia (paranoid, catatonic, etc) because they showed

limited diagnostic stability, low reliability, and poor validity. These subtypes also have not been shown to exhibit distinctive patterns of treatment response or longitudinal course. Instead, a dimensional approach to rating severity for the core symptoms of schizophrenia is included in Section III to capture the important heterogeneity in symptom type and severity expressed across individuals with psychotic disorders. [American Psychiatric Association,Highlights of Changes from DSM-IV-TR to DSM-5].

And, despite the heterogeneity in the diagnosis of schizophrenia, the diagnosis does help predict such outcomes as high suicide risk (5-6%), occupational dysfunction, and reduced rates of marriage. That said, I agree with Prof. Kinderman that there are often stronger and more specific predictors for some of these outcomes; e.g., a past history of suicide attempts is a stronger predictor of subsequent suicide than is a diagnosis of schizophrenia. So Prof. Kinderman and I agree that clinicians should not “assume” a high or low risk of some outcome, simply on the basis of a categorical diagnosis. There must be a much more fine-grained assessment of risk factors (see, e.g., Knoll JL, J Correct Health Care. 2010 Jul;16(3):188-204. doi: 10.1177/1078345810366457. Epub 2010 May 12. Suicide in correctional settings: assessment, prevention, and professional liability).

I do appreciate Prof. Kinderman’s implicit acknowledgment that abuses of human rights can, unfortunately, occur in any profession–including, alas, both psychology and psychiatry–and I applaud the statement from the British Psychological Society condemning such abuses. The American Psychiatric Association has issued similarly strong statements; e.g.,

The American Psychiatric Association reiterates its position that psychiatrists should not participate in, or otherwise assist or facilitate, the commission of torture of any person. Psychiatrists who become aware that torture has occurred, is occurring, or has been planned must report it promptly to a person or persons in a position to take corrective action.

I am also relieved to learn that Prof. Kinderman has had extensive experience working on inpatient settings with patients, as he puts it, “who have psychotic experiences.” I am sure we agree that such experiences are in need of a comprehensive differential diagnostic process–both to rule out classic, “organic” causes, such as a brain tumor, and also to consider neurodevelopmental syndromes leading to psychosis. This includes–in my view–schizophrenia.

Since Prof. Kinderman and several of his colleagues have had experience with severely and acutely psychotic, agitated inpatients, he will undoubtedly agree with me that, in such cases, an approach based on a patient’s self-generated “problem list” is often impossible. First, the agitated, psychotic patient must be diagnosed (e.g., rule out delirium, drug intoxication, etc.); and often, treated with medication. (If Prof. Kinderman is aware of non-pharmacological alternatives to the treatment of the acutely agitated, psychotic or manic patient–alternatives that have been validated in randomized, controlled studies–I would be most grateful for the literature citations. Psychiatrists are certainly interested in non-pharmacological alternative treatments for such severely disturbed patients. I am aware, of course, that various behavioral techniques may be useful adjunctive approaches to the acutely psychotic patient).

Incidentally, I believe it is a serious error to assume that we do not know anything substantial as regards the neurobiology of schizoprenia and related psychoses, notwithstanding the dismissive and unfortunate remarks from Dr. Thomas Insel re: the DSM-5. To be sure: we don’t have “office-ready” laboratory tests for the group of conditions now conceptualized under the rubric of “schizophrenia”; but we know much more about the neurobiology of schizophrenia than is commonly supposed [see Howes et al., for a recent review]

Re: the DSM-5, I do not accept the ad hominem argument that there is something inherently prejudicial, unethical or disqualifying about DSM-5 workgroup members having had (to a limited degree) past financial relationships with pharmaceutical companies; nor do I accept the notion that there is some inherent contradiction between the deliberations of DSM-5 “committees” and an “evidence-based” approach to the issues. I do not know if Prof. Kinderman or other participants in this summit have actually gotten hold of the DSM-5. Having perused it in some detail, I am impressed by its depth of research; rich references to empirical data; and the broad participation of dozens of PhD psychologists, social workers, and other mental health professionals. [Disclosure: I had no official relationship with the DSM-5 work groups; receive no remuneration or other benefits from pharmaceutical companies; and have recently retired from clinical practice]. There is also a concerted attempt, in DSM-5, to present a “dimensional” approach to several disorders, particularly the personality disorders. And so, before summit participants launch into more criticism of the generic “DSM” concept, I would urge them to examine the “ocular proof”: the actual DSM-5, which differs substantially from earlier versions of the DSM. After all, you can’t criticize the “movie”  without seeing it!

Finally, I would like to clarify what must have been Prof. Kinderman’s misreading of my comments on “Hippocratic medicine” and the philosophy of Dr. William Osler. Contrary to Prof. Kinderman’s misapprehension, I am decidedly not a fan of “medieval medicine”! Nor do I advocate the thoughtless or casual use of medications, such as antipsychotics. Pharmacological agents should be used, in mental health care, only when absolutely necessary; at the lowest feasible dose; and for the shortest clinically feasible time.

The title of my last commentary (“Osler in Wonderland”) alluded to Osler’s view that physicians should not, as a first approach, simply treat “symptoms”; rather, they should try whenever possible to understand the disease process generating the symptoms. So, too, with regard to self-generated lists of so-called “problems” from our patients and clients. True: it is sometimes necessary to treat these complaints individually and empirically; but whenever possible, we should try to identify a unifying causal entity underlying the patient’s list of problems. This need not be a Platonic “category”–the phrase, “carving Nature at its joints” is from Plato’s Phaedrus–but sometimes, a categorical diagnosis may be useful and appropriate. Delirium is one example; bipolar I disorder is another. And it is simply incorrect to assert that the category utilized must be one for which the precise pathophysiology is already known. Only very recently, for example, has the pathophysiology of delirium been somewhat clarified.

Finally, while I share with Osler the view that diagnosis is indispensable, I also agree with his well-known comment that, sometimes, “It is much more important to know what sort of a patient has a disease than what sort of a disease a patient has.”

With best regards, and now signing off,

Ron Pies MD

Ronald Pies

About Ronald Pies

Dr. Pies is Professor of Psychiatry and Lecturer on Bioethics & Humanities at SUNY Upstate Medical University; and Clinical Professor of Psychiatry at Tufts University School of Medicine, Boston.