Medical Education
May 2020

How Should Shared Decision Making Be Taught?

Dong-Kha Tran, MD and Peter Angelos, MD, PhD
AMA J Ethics. 2020;22(5):E388-394. doi: 10.1001/amajethics.2020.388.


As the field of medicine shifts from a paternalistic to a more patient-centered orientation, the dynamics of shared decision making become increasingly complicated. International globalization and national socioeconomic differences have added unintended difficulties to culturally sensitive communication between physician and patient, which can contribute to the growing erosion of clinician empathy. This article offers a strategy for teaching students how to enter into conversations about shared decision making by bolstering their empathy as a result of exposing them to the many variables outside of their patients’ control. Patients’ historical and cultural context, gender identity, sexual orientation, and common assumptions about clinicians as well as institutional biases can severely limit students’ ability to integrate patients’ value-laden preferences into shared decision making about health care.


Once viewed as a paternalistic-oriented profession characterized by physicians’ overprotection of care recipients, medicine has now shifted towards more person-centered care.1 In conjunction with international globalization and national socioeconomic disparities, this shift has complexified medical training. Memorable words from the authors’ first day of medical school— that “medicine is a service profession”—continue to echo. Our service to patients is to bring extensive medical training and knowledge to bear on their personal, real-life experiences to generate a foundation for shared decision making.

We believe that teaching empathy is the best way to prepare students to serve patients. It is not to be assumed that those who enter the profession are inherently empathetic or compassionate. Empathy has been defined, but oversimplified, as “an ability to understand the patient’s inner experiences and perspective and a capability to communicate this understanding.”2 However, as the current sociopolitical climate has helped make increasingly clear, the disparities and bias suffered by many patients are such that students, let alone heath care professionals, might have incredible difficulty in finding common ground with their patients. Institutional and systemic barriers and biases can have drastic but nearly invisible effects on the patient-clinician relationship. Although empathy serves as bridge between clinicians’ and patients’ experience, especially if clinicians have had personal experiences with a particular disease process, we argue that the combination of patients’ life experiences and illness experience is unique. This degree of complexity encourages a patient-centered rather than a paternalistic approach.

Teaching empathy is no small undertaking, given the insufficient evidence of the effectiveness of educational interventions designed to enhance empathy.3,4 We believe that, prior to implementing any such intervention, clinicians must first build a pedagogical foundation. As we shall describe, laying this foundation requires exposing students to diverse patient backgrounds rooted in a complex variety of factors—including, but not limited to, race, gender, culture, lifestyle, and socioeconomic status—through coursework, patient panels, and patient encounters during rotations, residency, and beyond. This cumulative experience leads to broadening of students’ perspectives, which foundation is required to support not only empathy but also shared decision making. The goal of this training is for the student to become patients’ and other stakeholders’ teacher, translator, and guide in the complex medical field where stakeholder preferences are folded into health care decision making, informed consent, and care planning.

Patient Background

Race. There is evidence that minorities’ preferences for treatment and disclosure may differ from those of whites. Dula and Williams argue that common assumptions about end-of-life care contradict those of African-Americans, who tend to prefer more aggressive care.5 Lack of understanding of apparently irrational demands for treatment can lead to clinician frustration and inability to incorporate patients’ preferences in decision making. Contrasting cultures’ preferences for diagnosis disclosure are illustrated in “What You Don’t Know” (now the film The Farewell).6 This story, about a Chinese-American family that decides not to inform the terminally ill grandmother of her prognosis, brought into the mainstream the author’s struggle straddling 2 cultures and raised many ethical questions about how Western medical assumptions about patient preferences might not be in line with those of other cultures.

Gender and sexual identity. Over the years, there have been large societal shifts in sexual orientation and gender identity. Studies have shown compelling evidence of increased negative health indicators and higher rates of victimization in lesbian, gay, bisexual, transgender, and queer youth.7 Although it is clear that health care practitioners must be comfortable discussing sexuality and sexual orientation, in one survey of lesbian, gay, and bisexual young adults, 78% of respondents reported that these issues were never discussed at all with their clinician during adolescence and 67% that they would have liked to have had such conversations.7

Socioeconomic status. Although there has been a focus on race, gender, sexual orientation, and culture in social psychological analysis of identity, socioeconomic factors such as income and education have also been shown to affect personal and social identities.8 Patients of lower socioeconomic status might believe they get lower-quality care based on clinician assumptions about the treatment or medications that patients with fewer resources deserve. Higher education has been shown to increase patient self-advocacy, enabling clinicians greater understanding of presenting concerns and symptoms that may lead to more accurate and timely diagnoses.9

As these examples hopefully demonstrate, it is imperative that students be aware of patients’ cultural background. Awareness is a valuable tool for health care professionals to open a dialogue and have care conversations with patients that draw on patients’ real, individual experiences.

Structural and Individual Bias

In 2003, the Institute of Medicine (IOM) released an extensive report, Unequal Treatment: Confronting Ethnic and Racial Disparities in Health Care,9 which found that, regardless of socioeconomic and sociodemographic status, racial and ethnic disparities in health care persist and are associated with worse outcomes. The broader historical and contemporary inequalities experienced by minorities in the United States contribute to complex structural and individual biases in many clinical encounters. Since the landmark IOM report, perceptions of health care discrimination have decreased among Latino, Asian, and immigrant individuals but remained consistently high among black individuals.10 Nevertheless, all of these minorities10—and among the chronically ill, blacks11—continue to perceive that they are subject to health care discrimination at higher rates than whites.

Historical and contemporary inequalities in the US contribute to complex structural and individual biases in many clinical encounters. 

Multiple barriers can prevent patients from receiving high-quality medical care. Language barriers, for example, contribute to inadequate patient understanding and informed consent, exacerbating health systems problems such as lack of resources, knowledge, and institutional priority. Outside the health system, medications are vastly undersupplied in predominantly minority neighborhoods compared to predominantly white neighborhoods.5 Moreover, considerable research in the field of psychology has shown that the most well-meaning clinicians are socialized to have implicit and explicit stereotypes.5,12 Clinicians making judgments under the pressure of time and resources are susceptible to information shortcuts (ie, stereotypes) due to lack of information.

Structural biases are present in legal, regulatory, and policy-making areas. The IOM noted that racial and ethnic minorities are more likely than whites to be enrolled in “lower-end” health plans with stricter limits on covered services, with the result that health care financing and delivery are fragmented by socioeconomic status.3,5 Conversely, at the opposite extreme is the highly publicized case of Steve Jobs reported in 2009 by the Wall Street Journal. California-resident Jobs had the means to receive a liver transplant in Tennessee with a median waiting list time of 48 days (compared to the national average of 306 days).13

All of the above factors may contribute to patient mistrust and treatment refusal, from which physicians can wrongly infer that patients are unsophisticated and uneducated.5 However, in the context of systemic prejudices, students may begin to understand how patient decisions may be completely logical.

Compassion Fatigue

Compassion fatigue, a component of burnout, was first studied in crisis counselors and mental health practitioners14 and is now commonly studied in palliative care nursing clinicians.15 It is a result of repeated exposure to stressors, including death and dying, vicarious trauma, expanding workloads, and a feeling of impotence to do more to help in the face of limited resources.4,14,16 Although we cannot speak with authority on the matter, communication barriers—such as misunderstanding of patient assumptions and perceptions—arguably contribute to clinician frustration and compassion fatigue. This misunderstanding can lead to patient nonadherence that can then lead to ineffective treatments—all of which are possibilities that are exacerbated by the time constraints of clinical practice. Tempering and modulating what we assume about our patients can facilitate starting a health care conversation.

Strategies for Education

In light of factors that contribute to disparities in health care, teaching students how to facilitate shared decision making will require more than knowledge of disease pathophysiology and its treatment that medical education currently requires. Despite limited evidence of the effectiveness of empathy-enhancing interventions, we believe empathy should be taught through longitudinal exploration and discussion of the many variables that affect patients’ psychosocial identities and contribute to health care disparities. Broadening students’ cultural perspectives through teaching empathy would facilitate their understanding of patients’ preferences and prior health care decisions and thus facilitate shared decision making.

It would be easy to infer from the discussion thus far that we believe that paternalism is bad and respect for autonomy is good, but the role of the 2 orientations in shared decision making is much more complicated. Ideally, paternalistic attitudes stem from protectiveness and thus imply benevolence, but they risk overprotectiveness—resulting in loss of patient autonomy and possibly negative impacts on patients’ mental health.1 Respecting patient autonomy, however, requires that patients be competent, and their level of competence is dependent on their psychosocial environment.1 It is thus necessary for the student to be able to balance paternalism and respect for autonomy in different scenarios.

As described by Lerner and Caplan, bioethical education and discourse must be used to “historicize but not minimize past ethical transgressions” so as to emphasize why and how such events happened.17 Just as students are implored to evaluate academic research in the context of a priori and retrospective biases, so deconstructing history can reveal how complex historical circumstances can lead to experimentation that is unethical in hindsight. Lobotomies, for example, are viewed today as barbaric, but before the advent of antipsychotic medications they were believed to be the best last-resort option for patients who suffered immensely from psychosis.17 Similarly, the infamous US Public Health Service Syphilis Study at Tuskegee, which knowingly denied accepted treatment to poor African-Americans with syphilis, arguably was ethically justified, at least early in the study when, prior to the advent of penicillin, treatment was prohibitively expensive for many study participants, such that these participants might have been no worse off had they not participated. By arming students with this knowledge and making them aware that the fallacy of Whig history (that society becomes morally progressive over time) also applies to medicine and that past transgressions can be regarded as at least explicable, if not excusable,17 students can critically evaluate how best to obtain informed consent.


Empathy is poorly defined, and interventions to cultivate empathy have failed to translate into clinical practice. It has been shown that undergoing communication-enhancing interventions does not improve medical students’ attitudes toward patient centeredness,5 and it has been posited that unprofessional students can “fake” professional behavior in exam settings.18 We do not disagree with the difficulty of empathy education. Given the brief overview provided of complex factors that go into the patient-clinician relationship, we argue that undertaking shared decision making with patients is complex to the point of requiring cumulative and longitudinal experience that is outside the scope of many studies. We suggest a strategy by which educators introduce cultural and historical perspectives as early as possible in medical training so that each student can be exposed to many social and cultural worlds over time beyond their own social and cultural bubble. This greater awareness, in turn, would help each student understand patients’ preferences and decisions, thereby laying the groundwork for shared decision making. Before interventions can truly take hold, we must first ensure that all members of the medical profession understand that the factors discussed in this paper exist.

It can be argued that the addition of empathy training places an unrealistic burden on trainees struggling to demonstrate competency at a time when compassion fatigue and burnout are garnering attention. We argue that teaching empathy in medical education would best support patient-centered care. As our profession evolves, physicians and educators must also consider the many factors that contribute to patient health beyond medicine. To ignore these would be a disservice not just to our patients, but also to future medical professionals who have the right to be aware of the factors that complicate care.


In order for new resident physicians to overcome the many frustrations that might be involved in the shared decision-making process, medical educators must cultivate empathy early in students’ careers. Given the complexity of patient backgrounds in conjunction with institutional and clinician bias and socioeconomic disparities, it is vital that medical students understand the sociodemographic and cultural factors that influence patients’ preferences and decisions regarding health care. Only then can the student begin to apply the knowledge of medicine by focusing it through the lens of the patient. And only when the physician’s medical training and the patient’s unique real-life experiences are bridged can the patient be truly informed and the physician and patient begin to have effective and efficient conversations.


  1. Fernández-Ballesteros R, Sánchez-Izquierdo M, Olmos R, Huici C, Ribera Casado JM, Cruz Jentoft A. Paternalism vs autonomy: Are they alternative types of formal care? Front Psychol. 2019;10:1460.

  2. Hojat M, Gonnella JS, Nasca TJ, Mangione S, Vergare M, Magee M. Physician empathy: definition, components, measurement, and relationship to gender and specialty. Am J Psychiatry. 2002;159(9):1563-1569.
  3. Aponte-Soto L. Minorities and bias: the big picture. Science. 2016;353(6297):357-358.
  4. Batt-Rawden SA, Chisolm MS, Anton B, Flickinger TE. Teaching empathy to medical students: an updated, systematic review. Acad Med. 2013;88(8):1171-1177.
  5. Dula A, Williams S. When race matters. Clin Geriatr Med. 2005;21(1):239-253, xi.
  6. Wang L. What you don’t know. This American Life. Public Radio Exchange. December 12, 2017.

  7. Coker TR, Austin SB, Schuster MA. The health and health care of lesbian, gay, and bisexual adolescents. Annu Rev Public Health. 2010;31(1):457-477.
  8. Manstead ASR. The psychology of social class: how socioeconomic status impacts thought, feelings, and behaviour. Br J Soc Psychol. 2018;57(2):267-291.
  9. Smedley BD, Stith AY, Nelson AR, eds; Institute of Medicine. Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington, DC: National Academies Press; 2003.

  10. Schulson LB, Paasche-Orlow MK, Xuan Z, Fernandez A. Changes in perceptions of discrimination in health care in California, 2003 to 2017. JAMA Netw Open. 2019;2(7):e196665.

  11. Nguyen TT, Vable AM, Glymour MM, Nuru-Jeter A. Trends for reported discrimination in health care in a national sample of older adults with chronic conditions. J Gen Intern Med. 2018;33(3):291-297.
  12. FitzGerald C, Hurst S. Implicit bias in healthcare professionals: a systematic review. BMC Med Ethics. 2017;18:19.

  13. Kane YI, Lublin JS. Jobs had liver transplant. Wall Street Journal. June 20, 2009. Accessed March 11, 2020.

  14. Peters E. Compassion fatigue in nursing: a concept analysis. Nurs Forum. 2018;53(4):466-480.
  15. Cross LA. Compassion fatigue in palliative care nursing: a concept analysis. J Hosp Palliat Nurs. 2019;21(1):21-28.
  16. Sinclair S, Raffin-Bouchal S, Venturato L, Mijovic-Kondejewski J, Smith-MacDonald L. Compassion fatigue: a meta-narrative review of the healthcare literature. Int J Nurs Stud. 2017;69:9-24.

  17. Lerner BH, Caplan AL. Judging the past: how history should inform bioethics. Ann Intern Med. 2016;164(8):553-557.
  18. Rees CE, Knight LV. The trouble with assessing students’ professionalism: theoretical insights from sociocognitive psychology. Acad Med. 2007;82(1):46-50.

Editor's Note

Background image by Ryoko Hamaguchi.


AMA J Ethics. 2020;22(5):E388-394.



Conflict of Interest Disclosure

The author(s) had no conflicts of interest to disclose.

The viewpoints expressed in this article are those of the author(s) and do not necessarily reflect the views and policies of the AMA.