Identifying the Best Approach to Patient-Centered Decisions About Serious Illness Care: What’s Autonomy Got to Do With It?

Propelled in large part by the American zeitgeist of self-determination, patient autonomy replaced beneficence as the primary ethical principle guiding medical decision making in the United States during the last century. This autonomy-driven model assumes an informed patient who fully understands the implications of available treatment choices and can identify those most aligned with their values and goals. Furthermore, this model tends to assume the internal consistency of and ability to accurately identify such values and goals for patients’ current and future selves. However, there is increasing recognition that this conceptualization of patient autonomy is intricate and fallible, owing largely to the complex interrelationships among human cognition, emotions, and behavior, especially in the context of serious illness. We need to improve our understanding of the best approaches for guiding patients and their families to the right decisions for them in the context of serious illness—those decisions that best match their individual values and goals.

Rubin and colleagues¹ report an elegant and innovative randomized clinical trial using psychological experimentation with hospitalized patients to compare intuitive vs deliberative approaches to medical decision making. Participants consisted of 199 hospitalized patients aged 60 years or older with serious illness who were asked to express their treatment preferences for a series of hypothetical, albeit relevant, clinical scenarios using either intuitive or deliberative decision making. In the intuitive group, patients were subjected to a cognitive load (remembering a series of numbers) and instructed to provide their treatment preferences immediately based on instinct. In the deliberative group, patients were not subjected to a cognitive load and were instructed to think critically about their answers, take a full minute before answering, and explain why they made their decisions. Patients in both groups chose very similar treatments, reported very similar decisional certainty, and were equally likely to have their treatment choices match their stated personal values. This experiment provides persuasive evidence that encouraging hospitalized patients to deliberate on life-sustaining treatment decisions does not change the content or improve the quality of their decisions. The one potentially important exception is that patients in the deliberative group were more likely to endorse a palliative philosophy for goals of care, but this did not translate to different treatment choices. The study seems adequately powered to exclude moderate or large differences in decision making with deliberation—at least deliberation for 1 minute. It is possible that these patients had already formed decisions about intensity of care as more than half endorsed having a living will or engaging in prior discussions about end-of-life care, or that longer deliberations are necessary. However, these limitations seem unlikely to explain the absence of any important effect of deliberation on treatment decisions. Instead, this study raises the question of whether such complex medical decisions are always made with an intuitive approach and that additional efforts to promote deliberation, such as decision aids, may be doomed to fail the mission of changing decisions inherently anchored to initial instinctual responses.

This study also raises the important question of how to best facilitate informed decisions for seriously ill patients on their medical care that are in line with their individual values and goals. There are several important knowledge gaps that need to be addressed to understand the appropriate role of patient autonomy and how best to support seriously ill patients and their families in medical decision making.

Patients with serious illness are commonly asked to provide their preferences regarding highly consequential and complex medical decisions, often decisions that may affect their own deaths. Even in a shared medical decision-making model, there is often an implicit assumption that patients are able to fully understand complex options, in the context of illness and distress, and to participate as an informed partner. However, patients who are naive to treatments such as endotracheal intubation, tracheostomy, and feeding tubes may not possess the level of understanding necessary to make the decisions they would if they had the training of clinicians. This is not an argument for paternalistic decision making, but rather an argument for research to explore ways to support patients with serious illness, and their family members, in making decisions that are true to their individual values and goals.

The correct role for patients and family members in these decisions varies by context and over time. A significant majority of healthy people (65%) report valuing the ability to choose their own treatment in the context of a hypothetical cancer diagnosis, but only 12% of patients with cancer actually want to have the same level of control over decisions.² Atul Gawande, MD, MPH, articulately describes his own desires to have another physician make decisions about his ill child: “I wanted them to decide, doctors I had never met before…I couldn’t bear the possibility of making the wrong call.”³ Patients’ own preferences for involvement in medical decision making can change over the course of a serious illness, especially in the context of psychological adjustment to serious illness. Further, negative mood impairs logical reasoning processes: anxiety increases risk aversion, sadness increases risk tolerance, and fear increases the perceived affect and consequences of negative outcomes.^4,5 People also tend to make decisions based on affective forecasting, which constitutes predictions about how future events will make them feel, and yet studies show that most people are poor affective forecasters, overestimating the intensity and duration of emotions in response to an event.⁶ In addition, people often underestimate their resilience and adaptability to bad situations.⁷ Again, the solution is not paternalistic decision making by physicians, but rather a better understanding of effective methods to identify the best approach to shared decision making for each patient and his or her family, in the context of the specific decision and socioemotional milieu.

The effect of decisional role on future mental health is also an important consideration. Matching family members’ preferences for decisional role to the role they are asked to play is associated with decreased symptoms of posttraumatic stress disorder after the death of a loved one in the intensive care unit.⁸ Cognitive dissonance theory suggests that people are adept at constructing narratives to rationalize decisions they made as the best of all potential alternatives. However, in the context of serious or even tragic choices regarding end-of-life care, cognitive dissonance may not help alleviate the emotional consequences including psychological distress, regret, and grief. A combination of a sense of personal responsibility with the ability to imagine better alternatives had one chosen differently may increase decisional regret and serious distress.⁹ For example, perceived personal responsibility for making tragic decisions about withdrawing life support for a terminally ill child seems to result in much greater distress and grief for parents than if the same outcome resulted from a choice made by physicians.¹⁰

We have much to learn about the best way to support patients with serious illness and their family members in patient- and family-centered decision making. Shared decision making is clearly the goal, but what does that mean in the context of serious illness? In striving for the best possible decisions for each patient, there are many complex nuances including the influence of cognitive biases, the effect of emotions on effective decision making, the ultimate role of intuitive decision making, and the emotional consequences of decision making about serious illness. Rubin and colleagues¹ are trailblazers in this area with this randomized clinical trial examining the effects of deliberative vs intuitive decision making. We need additional studies that incorporate the novel use of psychological experiments to examine the effect of cognitive biases, mental stress, and emotional valence on decision making, decisional outcomes, and other patient- and family-centered outcomes. Simulating the real-life context in which difficult decisions are made allows us to create approaches to supporting seriously ill patients and their families in the most effective, patient-centered decision making. Rather than supporting deliberation for the sake of patient autonomy, the goal should be making the best medical decisions with, or if they desire, for patients and their families based on the patient’s values and goals.

Published: January 25, 2019. doi:10.1001/jamanetworkopen.2018.7872

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2019 Lau N et al. JAMA Network Open.

Corresponding Author: J. Randall Curtis, MD, MPH, Division of Pulmonary, Critical Care, and Sleep Medicine, Harborview Medical Center, University of Washington, 325 Ninth Ave, PO Box 359762, Seattle, WA 98104 (jrc@u.washington.edu).

Conflict of Interest Disclosures: None reported.