Case and Commentary
Feb 2017

Polarities in Clinical Thinking and Practice

John Z. Sadler, MD
AMA J Ethics. 2017;19(2):141-146. doi: 10.1001/journalofethics.2017.19.2.ecas1-1702.


This analysis of a case of a bereaved patient that poses two treatment options—watchful waiting or medication—focuses on five “polarities” in clinical practice: (1) the normal and the pathological, (2) the individual and the diagnostic collective, (3) the primary care physician and the consultant, (4) the expert and nonexpert, and (5) the moment and the process. These polarities can accentuate ethical problems posed by this case, for example, by creating stark contrasts that mask the complex contexts of care and characteristics of patients. These stark contrasts can create false dilemmas that may obscure simpler, shared decision-making solutions. Alternatives to conceiving cases in terms of polarities are discussed.


Dr. Jones sees a new patient, Mr. Thompson, a 68-year-old man in her outpatient, primary care clinic today. In reviewing his intake forms, Dr. Jones sees that Mr. Thompson scored 18 points on the Patient Health Questionnaire (PHQ-9), suggesting “moderately severe depression.” When she asks Mr. Thompson about how he is feeling, he tells her that his wife died three weeks ago but that he did not want to tell Dr. Jones about this because he did not want to trouble her.

In further talking to Mr. Thompson, Dr. Jones discovers that he has lost appetite, interest in his activities, and the ability to concentrate at work. He feels tired yet has had trouble falling and staying asleep. Finally, family and friends tell him that he seems a little “distracted” or slower lately. Dr. Jones asks whether Mr. Thompson has thought of hurting himself, and Mr. Thompson says no. She also asks if he owns a gun, and he says no.

Dr. Jones begins to consider how to diagnose and treat Mr. Thompson. She knows that recent changes to the Diagnostic and Statistical Manual of Mental Disorders (DSM) regarding bereavement have caused significant debate among her psychiatric colleagues. She often refers psychiatric patients to either Dr. Taylor or Dr. Martinez, who seem to have very different stances on the issue as she learned on discussions during recent consults. Dr. Taylor worries that the DSM-5 is medicalizing normal grief even more than previous editions had: “Researchers find that grieving people typically yearn for their loved ones roughly every other day at one year after losing them, but the new DSM will lead many to label such people diseased.” He prefers to attentively watch such patients to see how their symptoms progress over time. Dr. Martinez shares Dr. Taylor’s worry about overmedicalization and “false positive” diagnoses but believes that, generally, clinicians should initiate treatment of possible depression if the symptoms are severe enough: “It’s entirely normal for one’s connective tissue to sever under blunt force trauma and for one’s body to react with fever and other symptoms to a viral invasion. It is easier to distinguish such dysfunction with ‘normal compensation’ in physical medicine from ‘normal function.’ Yet, in psychiatry, it is difficult to distinguish dysfunction with normal compensation from a ‘problem of living.’”

On the one hand, Dr. Jones feels that if she refers to Dr. Taylor, Mr. Thompson will receive watchful waiting, but she is concerned that his insurer will fight against reimbursement without a diagnosis. On the other hand, she is worried that a referral to Dr. Martinez will result in treatment with selective serotonin reuptake inhibitors (SSRIs) that is unnecessary and that Mr. Thompson may come to see himself as diseased rather than normal. Hence, she feels that her choice will ultimately determine Mr. Thompson’s diagnosis and treatment. She is unsure of what to do.


If I were Mr. Thompson and knew about Dr. Jones’s deliberations, I would want to stick with Dr. Jones, knowing that I was in good hands with such an insightful, knowledgeable, and thoughtful clinician. The following describes why.

This case poses what might be considered practice “polarities”—false dilemmas presented as either/or decision points to clinicians. These polarities could include (1) the normal and the pathological, (2) the individual and the diagnostic collective, (3) the primary care physician and the consultant, (4) the expert and nonexpert, and (5) the moment and the process. Each of these polarities play into the clinical problem posed here. One of the key points of this essay is that clinicians should be wary of such polarities because they oversimplify the complexities of clinical judgments and clinical relationships. The discussion that follows illustrates these points.

Five Practice Polarities and the Making of False Dilemmas

The normal and the pathological. The authors of the Diagnostic and Statistical Manual of Mental Disorders (DSMIV-TR and DSM-5 are careful to note that the manuals are “not meant to be used in a cookbook fashion” [1] and should be used with practical judgment sensitive to the clinical context [2]. They recognize with humility that most of the DSM disorders are without a specified pathoetiology, and DSM disorders remain works-in-progress in understanding mental illnesses, perhaps the most complex of conditions faced by clinicians. While the aspirations of the DSM are to provide empirically based, rigorous constructs for clinical, administrative, and research use, clinicians should know that ambiguous DSM cases are common, as is the one faced by Dr. Jones. Fortunately, Dr. Jones does not need to declare normality or disorder, other than as a diagnostic preliminary for a medical record. She can make her best estimate at Mr. Thompson’s initial visit and revise her assessments through clinical observations and responses to treatments, if any, in later visits. Indeed, she needn’t make a decision to refer in this clinical moment, but rather, can reserve some time to discuss options with Mr. Thompson in this and a prompt follow-up visit.

The individual and the diagnostic collective. As clinicians trained in the methods of science, we often forget that a patient with a disease or a disorder is a unique individual whom we encounter in his or her wholeness, while our knowledge of diseases and disorders is based on collections of people thought to resemble each other in specific ways—what might be called “diagnostic collectives” or groupings, which are abstract categories, remote from the complexity of the singular person. Those who define such diagnostic groupings—the World Health Organization, publishers of the International Statistical Classification of Diseases and Related Health Problems (ICD), and the American Psychiatric Association (APA), publishers of the DSM—wield considerable practical power and exert diagnostic authority no matter how conscientiously the authors of the diagnostic manuals may wish to constrain their authority. While Mr. Thompson might meet DSM-5 criteria for a major depressive episode, clinicians should keep in mind that the DSM-5 criteria are based upon shared characteristics of large groups of people, some of whom exhibit some, all, or none of the DSM criteria. How representative a diagnosis is and how much clinical utility it has are ongoing questions for psychiatric researchers and DSM committees. Thus, in clinical practice, making a diagnosis is just the beginning in finding out “what is going on with the patient” [3]. The clinician working with the unique patient adds living flesh to the bare-bones diagnostic category or categories that the patient seems to fit. Other considerations, of course, apply to “what is going on with the patient,” from the patient’s personal values, to his sociocultural context, to how his medical care is paid for, to name a few. These limitations of DSMcategories are why the DSM authors advise the DSM to be used with an eye towards its clinical utility (or not) [4]. While the DSM categories have the authority of the APA and teams of experts, the clinician is the ultimate arbiter in diagnosing her patient, and responsible clinicians will consider conventions as well as controversies in applying DSM categories, just as Dr. Jones does here.

The primary care physician and the consultant. Primary care physicians can use consultants in different ways. One way is to transfer total care for a particular condition to the consultant. In Dr. Jones’s case, she might want to let one of the psychiatrists simply manage Mr. Thompson’s depression. Alternatively, Dr. Jones might want a second opinion from one or more of the psychiatrists in deciding Mr. Thompson’s care, with advice in management or additional referral (to a psychotherapist or minister, for example). The sketchy details in the case make choosing any of these options difficult to substantiate. What’s clear is that Dr. Jones seems to be confident in predicting each of her consultants’ therapeutic leanings. Assuming her judgments are valid, what seems indicated to me is for Dr. Jones to discuss these possibilities with the patient and solicit the patient’s input in shared decision making [5, 6]. From this discussion, what the patient wants may become obvious, and in such an ambiguous treatment selection situation, Dr. Jones would have a poor justification to refuse Mr. Thompson’s preferred direction. In any case, she should monitor Mr. Thompson herself to address the excesses or neglect of one of the psychiatrists, if that were to happen. That, among other things, is what primary care is for.

The expert and nonexpert. This polarity is most closely attuned to the theme of legitimacy/authority. The DSM authors are experts in their field and experts in the diagnostic collectives they are dedicated to constructing. But, as noted in discussion of the individual and the diagnostic collective above, the physicians’ expertise stops at the patient they have not seen, whom they don’t know, and whom they have no relationship with. The “expert” may be a specialist as described above. However, the expert about Mr. Thompson, at least from the medical point of view, is certainly Dr. Jones. This medical expertise is complemented by Mr. Thompson himself as an expert “by experience” [7-9]. In the conclusion, I discuss how clinical decisions should emerge from this dual expertise of patient and clinician.

The moment and the process. Polarities of practice tend to prompt us to make quick decisions. But with the exception of the medical/surgical emergency or intensive care, quick decisions are not required and may represent unreflective, impulsive practice. Insurance company billing requirements and managed care also (seem to) demand quick decisions. But patients and their diagnoses change as their illnesses and lives change, regardless of how industry or experts describe patients’ maladies. In the case here, Dr. Jones does not have an urgent-care decision to make; provisional diagnoses and choices can be discussed with Mr. Thompson and can be made, tested, and revised over a series of brief outpatient encounters and ultimately submitted to an insurance company. Ethical dilemmas that seem so urgent in the moment melt away in the face of actual ongoing relationships, particularly ones that are ongoing in a primary care setting. While “the system” urges us to make quick decisions, they are rarely required even by the system. Although I do not think diagnosis based upon reimbursement rates is an honest way to practice medicine, the case here presents a genuine ambiguity that deserves a provisional, and only provisional, diagnosis. A relevant consideration in choosing a provisional diagnosis is whether the diagnosis permits (e.g., funds) the patient’s proper monitoring, but the clinician’s first obligation is ensuring that the patient, not the insurance company, receives proper care.


My comments about medical polarities have much to do with habits of thinking that are perpetuated by social and academic conventions. The case presented here is cast in a classical medical literature genre wherein the terms of the case are framed from the clinician’s—Dr. Jones’s—point of view. The case poses questions for Dr. Jones, and for Dr. Jones only. Dr. Jones has a clinical problem—referring a bereaved patient for treatment with either medications or psychotherapy—which seems to be solely hers. The problem with this genre convention is that the dialogical, interpersonal, intersubjective nature of the patient-clinician relationship is lost. The case presented here omits almost any salient information about Mr. Thompson’s values, psychosocial circumstances, personal preferences, ways of thinking, patterns of participating in health care, economic and insurance circumstances, and so on. Thus, the case seems highly problematic because a key portion of the patient-clinician relationship (that is, the patient) is missing from the case. In a more elaborated context and dialogue, Dr. Jones may find Mr. Thompson to be an individual who inhabits one or more of these contexts and has one or more of these characteristics:

  1. Doesn’t like to take medications, especially psychiatric medications
  2. Is already in grief counseling with his minister, and the minister recommended medication for him, as he is struggling more than most.
  3. Doesn’t like to go to consultants, because he thinks they are only in it for the money.
  4. Is a loner who would rather take a pill to ease his pain.
  5. Has great faith in Dr. Jones and would prefer her to make a treatment decision.

These, of course, are only five of the countless contexts and characteristics that arise in the patient-clinician relationship and, indeed, in shared decision making generally. Each of these in isolation suggests a relatively obvious course of action, once the patient’s perspective is understood. Unfortunately, our genre conventions of medical ethics cases often do not respect the patient’s standpoint, and patients can be presented, as here, as generic stand-ins for real people. What we need are ethics cases and pedagogies that embrace stakeholder voices and that support shared decision making, thereby avoiding polarization, false dilemmas, and oversimplifications. The Virtuous Psychiatrist: Character Ethics in Psychiatric Practice [10] and Narrative Psychiatry: How Stories Can Shape Clinical Practice [11] are examples of two books that demonstrate new genre forms of ethics cases as multistakeholder, dramaturgical processes that avoid false dilemmas and promote nuanced, collaborative practices with patients.


  1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-IV-TR. 4th ed. Arlington, VA: American Psychiatric Association; 2000:32.

  2. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Arlington, VA: American Psychiatric Association; 2013:21.

  3. Sadler JZ. The psychiatric significance of the personal self. Psychiatry. 2007;70(2):119.

  4. First MB. Clinical utility in the revision of the Diagnostic and Statistical Manual of Mental Disorders (DSM). Prof Psychol Res Pr. 2010;41(6):465-473.
  5. Barry MJ, Edgman-Levitan S. Shared decision making—the pinnacle of patient-centered care. N Engl J Med. 2012;366(9):780-781.
  6. Lipkin M. Shared decision making. JAMA Intern Med. 2013;173(13):1204-1205.
  7. Faulkner A, Thomas P. User-led research and evidence-based medicine. Br J Psychiatry. 2002;180(1):3.

  8. Fox J. The importance of expertise by experience in mental health services. Int J Leadersh Public Serv. 2008;4(4):39-43.
  9. Noorani T. Service user involvement, authority and the “expert-by-experience” in mental health. J Polit Power. 2013;6(1):49-68.
  10. Radden J, Sadler JZ. The Virtuous Psychiatrist: Character Ethics in Psychiatric Practice. New York, NY: Oxford University Press; 2010.

  11. Lewis B. Narrative Psychiatry: How Stories Can Shape Clinical Practice. Baltimore, MD: Johns Hopkins University Press; 2011.


AMA J Ethics. 2017;19(2):141-146.



The people and events in this case are fictional. Resemblance to real events or to names of people, living or dead, is entirely coincidental. The viewpoints expressed in this article are those of the author(s) and do not necessarily reflect the views and policies of the AMA.