Loading field map
Shared decision making (SDM) and clinical communication are two intertwined pillars of patient-centered clinical practice. Clinical communication is the broader field concerned with how clinicians and patients exchange information, build relationships, and navigate the emotional and practical realities of illness. Shared decision making is a specific, structured application of that communication: a process in which clinicians and patients jointly arrive at a medical decision when more than one reasonable option exists, integrating the best available evidence with the patient’s values, preferences, and circumstances.
The two are not synonymous. Clinical communication encompasses everything from taking a history and breaking bad news to motivating behavior change and coordinating care among multiple providers. Shared decision making is narrower—it concerns the deliberative phase of a clinical encounter where a choice must be made—but it depends on a foundation of effective communication. A clinician cannot engage in shared decision making without first establishing trust, eliciting the patient’s concerns, and conveying information in an understandable way. Conversely, shared decision making gives clinical communication a concrete purpose: it moves conversation from information exchange toward action.
The field addresses a fundamental tension in medicine: clinicians possess technical knowledge that patients lack, but patients possess unique knowledge of their own lives, values, and tolerance for risk. The central question is how to combine these two bodies of knowledge responsibly. Should the clinician recommend and persuade, or present options neutrally? How much technical detail should a patient absorb before a decision is truly informed? What happens when a patient’s preference conflicts with what the clinician believes is medically best?
The stakes are substantial. Medical decisions routinely involve trade-offs—between survival and quality of life, between aggressive treatment and palliation, between immediate side effects and long-term benefits. A decision that is right for one patient may be wrong for another with identical clinical findings. When decisions are made without eliciting patient values, the result can be overtreatment, undertreatment, regret, and erosion of trust. Conversely, when patients are left to decide without adequate support, they may feel abandoned or make choices based on misunderstanding. The field therefore asks not only how decisions should be made, but who should bear responsibility for them, and what constitutes a good decision when outcomes are uncertain.
The modern field emerged in the late twentieth century, but its intellectual roots reach further back. In mid-twentieth-century Western medicine, the dominant model was paternalistic: the physician, drawing on expert knowledge, determined the appropriate treatment and expected patient compliance. This model was challenged from several directions. The patients’ rights movement of the 1960s and 1970s, the rise of bioethics with its emphasis on informed consent and patient autonomy, and a growing body of research showing wide unexplained variations in medical practice all contributed to a reassessment. If equally qualified physicians in different regions made systematically different decisions for similar patients, then clinical judgment alone could not be the sole legitimate basis for choice.
The term “shared decision making” began to appear in the medical literature in the 1980s, associated with efforts to articulate a middle path between paternalism and what some saw as the opposite extreme: “informed choice,” in which the clinician merely presents facts and the patient decides alone. The middle path held that the clinician’s role is not to dictate but also not to withdraw—rather, to engage in a genuine dialogue, offering a recommendation when appropriate while remaining open to the patient’s priorities.
A crucial catalyst was the evidence-based medicine movement of the 1990s. Evidence-based medicine emphasized that clinical decisions should rest on systematic research rather than anecdote or authority. But evidence applies to populations; individual patients must still decide whether a treatment’s average benefit is worth its average harm for them. This created a natural partnership: evidence-based medicine could supply the facts, and shared decision making could supply the process for applying those facts to a particular person. The development of patient decision aids—booklets, videos, and interactive tools that present options and their outcomes in accessible language—provided practical instruments for the field. Research on these aids, synthesized in systematic reviews, showed that they improved patients’ knowledge, reduced decisional conflict, and helped patients choose options consistent with their values.
The field is not organized into sharply opposed schools, but several distinct approaches and research traditions coexist, each addressing a different aspect of the problem.
One influential tradition, rooted in the work of researchers such as Angela Coulter and Glyn Elwyn, emphasizes the elicitation and integration of patient values as the core of shared decision making. The clinician’s task is to present options fairly, describe their benefits and harms in terms the patient can grasp, and then help the patient weigh those outcomes against what matters most to them. This approach has produced practical models, such as the “three-talk model” (choice talk, option talk, decision talk), which breaks the process into teachable steps. Its strength is its clarity and applicability to clinical practice; its limitation is that it can become procedural, reducing a complex human interaction to a checklist. Critics note that patients do not always arrive with pre-formed values—values often emerge and shift during the conversation itself.
A second tradition, drawing on qualitative research, medical humanities, and the work of scholars like Rita Charon and Trisha Greenhalgh, emphasizes that illness is experienced within a life story, not as a set of abstract options. In this view, the clinician’s task is not merely to present evidence but to understand the patient’s narrative—how they make sense of their illness, what they fear, what they hope for. Decision making emerges from this understanding rather than from a structured deliberation over options. This approach is less prescriptive than the values-based model and harder to teach as a protocol, but it captures something the procedural models miss: that decisions are embedded in relationships and identities. Its limitation is that it offers less concrete guidance for clinicians who need to move a conversation toward a decision within a time-constrained visit.
A third tradition, associated with researchers like Albert Mulley and Michael Barry, focuses on identifying which decisions are genuinely “preference-sensitive”—where the best choice depends on how the patient values different outcomes—and on developing tools to support those decisions. This approach has driven the creation of patient decision aids and the measurement of decision quality. It asks: Did the patient know the relevant facts? Did they choose the option consistent with their stated values? This tradition has been influential in health policy, where it argues that preference-sensitive decisions should not be driven by clinician habit or financial incentive but by informed patient choice. Its limitation is that it can overstate the extent to which patients want to make decisions; many patients, particularly in acute or serious illness, prefer to delegate decisions to their clinician, and a rigid insistence on patient choice can itself be a form of abandonment.
A fourth tradition, rooted in medical education and the work of researchers like Suzanne Kurtz and Jonathan Silverman, treats clinical communication as a set of teachable skills. This tradition produced the Calgary-Cambridge guide, a widely used framework that breaks the clinical encounter into tasks: initiating the session, gathering information, building rapport, explaining and planning, and closing. Shared decision making appears within this framework as one component of the “explanation and planning” phase. This approach has been enormously influential in medical education, where communication skills are now routinely taught and assessed. Its strength is its practicality; its limitation is that it can fragment the encounter into techniques, losing sight of the overall purpose of the conversation.
These approaches are not rivals in the way that competing scientific paradigms are rivals. They overlap and inform one another. The values-based model needs the relational understanding of the narrative approach to avoid becoming mechanical; the communication-skills tradition provides the behavioral substrate on which the other approaches depend. In practice, most clinicians and researchers draw on several traditions, and the field’s major debates are less about which approach is correct than about how to balance competing goods: patient autonomy versus beneficence, procedural rigor versus relational depth, efficiency versus thoroughness.
Several concepts recur throughout the field and are essential for understanding its current landscape.
Informed consent is the legal and ethical prerequisite for treatment, but it is not the same as shared decision making. Informed consent typically requires disclosing risks, benefits, and alternatives; it can be satisfied by a signature on a form. Shared decision making goes further, requiring that the patient actually understand the information and that the decision reflect their values.
Decisional conflict is the state of uncertainty a patient experiences when facing a choice among options that involve trade-offs. The field treats decisional conflict not as a psychological failing but as a normal and often appropriate response to a genuinely difficult choice. The goal of shared decision making is not to eliminate this conflict but to resolve it in a way the patient can live with.
Decision aids are tools designed to support shared decision making. They present options, their outcomes, and their probabilities in accessible formats, often including value-clarification exercises. Systematic reviews have found that decision aids improve knowledge and reduce decisional conflict, but they are not a substitute for conversation; they are best used as a preparation for discussion, not a replacement for it.
Risk communication is the art and science of presenting probabilistic information. It addresses questions such as: Should risks be presented as percentages, frequencies, or visual displays? Should the focus be on relative risk reduction or absolute risk reduction? How should uncertainty be conveyed? Research in this area has shown that the framing of risk information can substantially influence decisions, which raises ethical questions about how clinicians should frame information without being manipulative.
The option of not deciding is a concept that receives increasing attention. Some patients, after hearing the options, prefer to defer the decision to their clinician or to a family member. The field has had to grapple with whether this is a failure of shared decision making or a legitimate outcome of it. Most current thinking holds that a patient’s informed choice to delegate is itself a valid expression of autonomy, provided the patient genuinely understands that they are delegating and that the clinician is aware of the patient’s values.
Contemporary shared decision making and clinical communication face several durable tensions.
Implementation remains the central challenge. Decades of research have produced robust evidence that shared decision making improves outcomes, and many health systems have formally endorsed it. Yet observational studies consistently show that it is not routinely practiced in most clinical encounters. Clinicians cite time pressure, lack of training, and the difficulty of applying a deliberative process to acute or urgent situations. The field has responded with efforts to integrate shared decision making into clinical workflows, to train clinicians in communication skills, and to develop decision aids that can be used within the flow of a visit. Progress has been real but uneven.
Measurement is contested. How does one know whether shared decision making has occurred? Some researchers use audio recordings of encounters, coded for the presence of specific behaviors. Others use patient-reported measures of whether they felt involved. Still others use decision quality measures—did the patient know the facts and choose consistently with their values? Each approach captures something different, and the field has not settled on a gold standard. This matters because health systems increasingly tie reimbursement or quality ratings to measures of patient involvement, and the choice of measure shapes what clinicians do.
Cultural and contextual variation is increasingly recognized. The field’s early research and models were developed largely in North America and Western Europe, where patient autonomy is a dominant value. In other cultural contexts, patients may expect clinicians to make decisions, or may expect family members to be centrally involved. The field has moved from assuming that shared decision making is universally appropriate to asking how it should be adapted to different cultural norms. This is not a settled question: some argue that the core principle—that decisions should reflect patient values—is universal, while the process must be adapted; others argue that the very concept of individual autonomy is culturally specific and that imposing it is a form of epistemic imperialism.
Digital technology is reshaping the field. Telemedicine has changed the communication environment, raising questions about how to build rapport and share decisions without physical presence. Artificial intelligence tools that provide risk predictions or even treatment recommendations introduce a new actor into the decision-making process: the algorithm. The field is beginning to ask how clinicians should present machine-generated recommendations, and whether the principles of shared decision making apply when the “evidence” is a black-box prediction rather than a clinical trial.
Health literacy and equity are persistent concerns. Shared decision making presupposes that patients can understand and engage with complex information. Patients with limited health literacy, cognitive impairment, or language barriers are at risk of being excluded from the very process designed to empower them. The field has responded with plain-language decision aids, interpreter services, and communication techniques tailored to low-literacy populations, but the gap between the ideal and the reality remains wide.
The field’s durable contribution is the recognition that medical decisions are not purely technical. They are human decisions, made by people with lives, fears, and hopes, and the quality of the conversation that precedes them matters as much as the quality of the evidence behind them. The ongoing work of the field is to make that recognition operational—to build systems, train clinicians, and develop tools that make genuine dialogue possible within the constraints of real-world medicine.