Shared Mind: Communication, Decision Making, and Autonomy in Serious Illness

INTRODUCTION

In the context of serious illness, individuals usually rely on others to help them think and feel their way through difficult decisions. Yet, we know very little about why, when, and how individuals involve trusted others in their care. Recognizing that information, deliberation, and decisions often occur within relationships, health professionals and social scientists have called for richer conceptualizations of autonomy and decision-making processes.^1–14

In this article, we explore how relationships can inform decision-making in clinical practice, considering how information, emotions, values, and autonomy are shared among patients, family members, and health professionals. Providing high-quality information about disease-related factors has been the focus of many efforts to improve decision making, yet we know little about how information is used by patients and their families. Shared decision making has often been conceptualized as a process of matching of choices to patients’ values and preferences with the goal of promoting individual autonomy; however, people often rely on others to define and weigh the values that shape their decisions.⁵ These influences can be both positive (eg, clarifying a discussion) or negative (eg, mindlessly adopting another’s perspective as one’s own). Emotions can strongly shape preferences¹⁵; even so, although patients report strong emotional bonds to their physicians, most emotions in patient-physician interactions are unspoken.¹⁶ There has been progress in helping patients to be more active in decisions, but patients, physicians, and communication experts have different criteria for shared decisions.¹⁷

In this article, we propose an interactional approach as one anchor for a continuum of decision-making processes in which relational autonomy emerges through shared information, shared deliberation, and shared mind. By shared mind, we mean situations in which new ideas and perspectives emerge through the sharing of thoughts, feelings, perceptions, meanings, and intentions among 2 or more people.¹⁸ Shared mind can be conceptualized as an achievement (eg, consensus), but we are more interested in shared mind as an interpersonal process (eg, becoming attuned) that is present to varying degrees in human interactions.

The concept of shared mind coalesces some of the recent thinking about decision making and derives from 2 observations: (1) that much of what we consider personal or individual (eg, our preferences, values, and attitudes) is socially mediated^18,19; and (2) that humans communicate and understand each others’ subjective experiences through multiple pathways—spoken language, expressed emotion, meaningful actions. In our view, shared mind in decision making is both discovered (as a naturally occurring phenomenon) and created (resulting from effort on the part of clinician or patient). With awareness, clinicians and patients can learn how to recognize shared mind when it is occurring naturally, recognize when it is not, and, when appropriate, change their behavior to promote it. To illustrate, we will first consider a clinical situation, then describe attributes and processes underlying shared mind.

Richard Grayson (a pseudonym, based on an actual patient) was a retired and intellectually active professor of epidemiology with recently diagnosed cancer of the pancreas. After consulting 2 oncologists and a surgeon, he was offered conflicting recommendations and a limited range of options. Surgery was unlikely to improve survival or quality of life. Although initially he thought he would never consider chemotherapy for an incurable cancer, Mr Grayson was surprised that chemotherapy offered a 30% chance of life extension by 2 to 3 months, and might improve his quality of life—but with the risk of side effects. Still, no one could know whether he would be one of the lucky 30% whose cancer would respond or whether palliation alone might offer more comfort. In addition to local options (choice between 2 chemotherapy regimens and a phase 3 clinical trial comparing high- and low-intensity regimens), he also considered phase 1 trials of promising new drugs and alternative regimens currently favored in Europe. Yet, as an epidemiologist, he was exquisitely aware that he was both older and had more extensive tumor burden than the clinical trial study populations, limiting their applicability to his own condition. He also knew that his logical analysis of the situation was hijacked by strong affect; he was often overwhelmed rather than enlightened by more information.

Despite being well-informed and having good social support, excellent access to medical care, and a trusting patient-physician relationship, he felt lost. Mr Grayson, accompanied by his partner, asked his family physician, “What would you do if you were me?” His physician responded, “I might have different values and preferences than you do, so, let’s focus on what’s most important to you.” This commonly advocated approach assumes that Mr Grayson has a set of personal values that would guide him through this particular situation, and that his preferences would be clear if his physician only asked. It assumes that his values are static, and exist in him as an individual, rather than as a dynamic dialogue that also engages his close social networks. Although Mr Grayson had written advance directives and could articulate broad values and preferences—his desire to live, avoid suffering and avoid iatrogenic harm—he had difficulty translating general principles into a concrete choice in this specific context. Furthermore, contemplating the choice only accentuated Mr Grayson’s sense of burden.

Dr Porter, his family physician, also found it difficult to provide clear answers. The effectiveness data were ambiguous. Chemotherapy might make things better or might make things worse. Traveling to other centers would be tiring and reduce contact with social supports. Dr Porter wondered whether she should help the patient make a decision, make a stronger recommendation, or engage others in the discussion to spare the patient the burden of deciding.²⁰ In trying to honor Mr Grayson’s individual autonomy (eg, clarify his preferences, make wise decisions), she also recognized his limited ability to do so without the input from trusted others.

In seeking others to help him understand his values and feelings, cope with uncertainty, sort out his options, and navigate his way, Mr Grayson changed from being uninformed but certain (that chemotherapy would not be useful in this circumstance) to being informed and uncertain (it might be); his preferences also changed from being stable (no chemotherapy) to being unstable (not knowing what to think). After some time, a set of values, preferences, and plans emerged that not been previously voiced by any of the individuals—in this case, a time-limited clinical trial of second-line chemotherapy with the option for dose reduction, concurrent palliative care, and changing from full-code to do-not-attempt-resuscitation status. Mr Grayson, his family, and his physicians believed that the decision was mutually endorsed, clinically reasonable, emotionally calming, and somehow made sense in the context of the mutually constructed set of goals and values. This decision—in the face of a complex and overwhelming situation—was not made by an individual. Rather, the decision was built²¹ within partnerships that enabled Mr Grayson to participate meaningfully in care.

ARE TWO MINDS BETTER THAN ONE?

Observations about Mr Grayson and other patients have led us to consider an interactional view of care. Clinicians, family members, and relevant others did not merely help Mr Grayson clarify his own preferences so he could make an individual autonomous decision. Rather, they interacted in such a way that all parties considered new information, perspectives and options—the decision emerged out of collective thinking and feeling. Put another way, a 2-way conversation considers 3 minds—the patient’s, the clinician’s, and that which is shared between them. We take this view for several reasons.

First, in general, important health decisions are usually not made in isolation. They are usually made in the context of social networks with friends, family, other social contacts, and health care professionals.¹² These social networks emerge and change when patients develop serious or chronic illnesses. Whereas a healthy person might have a sustained relationship with one primary care physician or with a small number of health care professionals, those with serious and chronic illnesses have multiple connections to the health care system—they interact with many interrelated physicians, nurses, and therapists. Their family members, caregivers, and friends have additional health-related interactions with each other, the patient, and the clinicians involved in patients’ care.²² Professional networks of clinicians can grow increasingly complex when caring for seriously ill patients. Yet, we know little about when and how decisions are made within these complex, emergent social networks.

Second, patients’ preferences may be vague, unstable, and uninformed. Although some patients have underlying values and preferences that clearly apply to the health care decision at hand, often these values and preferences have not been previously examined, are general (eg, longevity and quality of life) rather than contextualized to the specific decision (eg, whether to choose disfiguring surgery with a low chance of cure), and are in conflict with other articulated values and preferences.²³ In simple, familiar, and anticipated situations, preferences tend to be stable, predictable, and consistent. Preferences in complex, ambiguous, unfamiliar, and affectively charged situations, however, often are not^24,25; in such situations preferences tend to be remarkably sensitive to subtle effects of framing, ordering, tone of voice, word choice, and recent experiences.²⁵ Rationales for each preference may sound quite logical, but they are often constructed as an afterthought.²⁶ Furthermore, people often are not adept at predicting how they will feel after a future event has occurred (ie, affective forecasting)^27,28 and often underestimate their own ability to adapt to adversity. Thus, unassisted, patients’ decisions might be neither truly informed nor autonomous, and patients may have limited insight into their own cognitive biases and limitations. Clinicians’ preferences and decision-making processes are similarly affected.

Third, patients often make medical decisions to take care of someone else’s psychological needs, sometimes compromising their own physical or emotional well-being. A patient may opt for chemotherapy in advanced cancer “because my family would be too upset if they thought I was giving up.” Although the moral implications of such decisions are complex, the presence of others’ needs points to fundamental flaws in the assumption that individuals consider a good decision solely in terms of their own health. Rather, exposing this kind of logic allows it to be examined and challenged, and patients can be offered other ways to address both their own needs and those of their family members.

Finally, an otherwise cognitively intact person with a serious illness often has difficulty processing complex information and ambiguity. Among seriously ill hospitalized adults who have no evidence of dementia, one study reported that cognitive testing was at a level similar to a 10-year-old—including concrete thinking and inability to attend to more than one idea at a time.²⁹ Furthermore, in this diminished cognitive state, patients often have to deal with uncertainty, sort through massive amounts of decontextualized information, consider multiple options, and face the emotional impact of the illness. Faced with complex, ambiguous, and emotionally laden circumstances (cognitive overload), a person will tend to ignore data, simplify tasks, and make hasty decisions, that is, revert to rules of thumb and stereotypes rather than explore the full range of his or her values and the scientific data.^30–32 Thus, seriously ill patients may need, and often seek, help from others to recall and process information. Clinicians also suffer from cognitive overload, increasing their propensity to rely on similar mental shortcuts and not question their own oversimplifications.^33–41

SHARED MIND

Mr Grayson’s situation shows how shared mind includes cognitive (eg, shared ideas, values, goals, or decisions), affective (feelings, emotions), and motor (eg, mirroring each others’ facial expressions and gestures) elements.^42–48 Here, we explore the literature on shared mind¹⁸ and related areas—collaborative cognition, attunement, and sensemaking. Although these studies have been largely in nonmedical contexts, these ideas can be applied to clinicians’ experiences with patients.

COMMUNICATION THAT PROMOTES SHARED MIND

Relationships and organizations are networks of conversations. Mr Grayson, his family, and his clinical team had conversations that were characterized by cooperative efforts to create coherence and meaning.⁷⁸ Clinicians listened attentively, expressed curiosity and interest, explored emotions, and responded to his requests for help,^79–82 which, in turn, promoted not only sharing of information but also shared deliberation.⁸⁰ Shared mind might happen during a single conversation, or it might be the result of a history of interactions among patients, family, and clinicians. Shared mind may be facilitated by personality factors, such as openness; relational factors, such as respect and trust; and organizational factors that promote and value time spent with patients.

INTERACTIONAL CARE AND RELATIONAL AUTONOMY

An interactional approach amends prior models of clinical care in its focus on relationships. Shared mind promotes relational autonomy,^5,10,93 a view which recognizes that humans are social beings and that trusting relationships and personal knowledge, in fact, enhance autonomy by helping patients to process complex decisions that otherwise overwhelm the cognitive capacity of a single individual.

All communication between physicians and patients, including attempts to achieve shared mind, has dangers, the greatest of which is self-deception. Because clinicians’ influences on decision making are ever present, self-awareness is key in helping clinicians provide appropriately tailored and balanced information, explore patients’ values, and be attentive to their concerns.^75,94,95 Lack of clinician self-awareness, by contrast, can lead to blindness to power differences between clinicians and patients (and within families); clinicians can unknowingly cross the line from collaboration to substituted decision making or coercion. Similarly, unwanted paternalism can masquerade as shared mind when the unexamined values of the physician overwhelm unspoken values of the patient and if the patient does not feel empowered to speak for him or herself. Misunderstandings or mistrust can lead clinicians to believe that verbal assent indicates the presence of shared mind. Thus, shared mind depends on enhancing clinicians’ capacity for mindfulness and self-monitoring,^96,97 as well as helping patients to be fully engaged in care.⁹⁸

Shared mind, like caring, may be difficult to measure, but that difficulty should not diminish its importance. Direct observation of clinical encounters and patient and clinician surveys reveal some markers of shared mind, such as, for example, the degree to which clinicians and patients have beliefs, emotions, and values in common, and, when they do not, the degree to which they can articulate each others’ perspectives.⁸⁶ Checking for mutual understanding and consensus are behaviors that can be observed, measured, and monitored. Ultimately, for the practicing clinician shared mind may prove to be a useful concept, as are presence and compassion, for identifying qualities of interactions worth cultivating and identifying when its absence interferes with clinical care.

By drawing attention to shared mind, clinicians can observe when it is present and when it is not, be attentive to the mutability of patients’ preferences, and note ways in which they and others can make helpful contributions to decision-making processes and enhancing patient autonomy. Medical decision making should consider not only the individual perspectives of patients, their families, and members of the health care team but also the perspectives that emerge from the interactions among them.