Article Text

Download PDFPDF

Shared decision-making as a method of care
  1. Victor M Montori1,
  2. Merel M Ruissen1,2,3,
  3. Ian G Hargraves1,
  4. Juan P Brito1,
  5. Marleen Kunneman1,2
  1. 1 Knowledge and Evaluation Research Unit, Mayo Clinic, Rochester, Minnesota, USA
  2. 2 Department of Biomedical Data Sciences, Section of Medical Decision Making, Leiden University Medical Center, Leiden, Netherlands
  3. 3 Department of Medicine, Division of Endocrinology, Leiden University Medical Center, Leiden, Netherlands
  1. Correspondence to Dr Juan P Brito, Department of Medicine, Mayo Clinic, Rochester, Minnesota, USA; brito.juan{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Care happens in interaction between the patient and the clinician, in conversation where the patient and clinician uncover or develop a shared understanding of the problematic situation of the patient and identify, discover, or invent ways to make that situation better, given what each patient prioritises and seeks.1 Thus, to get the right care for each patient, patient and clinician collaborate and deliberate together to figure out what to do.2

Shared decision-making (SDM) has been traditionally defined as a collaborative approach by which, in partnership with their clinician, patients are encouraged to think about the available care options and the likely benefits and harms of each, to communicate their preferences, and help select the best course of action that fits these.3 This definition is limited to situations in which the problem and the pertinent options to address it can be defined a priori, and the main task is to find the option that best matches the patient’s preferences. Thought in this way, SDM may involve distributing decision aids for patients to come prepared for the consultation and using tools to elicit and document patient preferences. This practice has been proposed as an expression of patient-centred care, a way of involving patients, an antidote to medical paternalism, and as a way to promote high-value care.3

But anytime, a patient and clinician figure out together what to do about the patient’s situation, they are doing SDM. Although there are multiple models and accounts of what SDM is and is not,2 4 5 in practice, SDM starts by determining the nature of the problematic situation the patient is experiencing. This often requires considering insights that only the patient and perhaps their family can share, insights about both the patient’s biology and biography. Then clinicians must mobilise their competence and compassion to work with patients to develop a sensible care plan that responds to the situation as understood, is based on relevant evidence, attends to the emotional aspects of the problem, and is feasible and sustainable for the patient.6 7

Therefore, we believe SDM is not ‘another thing clinicians must do’, that is, to help patients select the best evidence-based option given their preferences, but that it is a method of care, as central to the clinician’s art as history taking, the physical examination, the selection and interpretation of diagnostic tests, and patient education and counselling.

The practical method to implement SDM as a method of care proposed below seeks to make as few demands as possible of both patients, who are taxed by the demands of self-care and of navigating a labyrinthine healthcare system while responding to the demands of living,8 and of clinicians, who, despite some evidence of the contrary,9 often express their worries about SDM adding time to their encounters.10

Practicing SDM as a method of care

Here, we propose a simple method to implement SDM in practice (see summary box 1).

Box 1

Steps for shared decision-making (SDM) in practice

(1) Foster a conversation

Key elements: problem definition, iteration, co-creation

(2) Purposefully select and adapt the SDM process

Matching preferences

Reconciling conflicts


Meaning making

(3) Support SDM

Protect the space

Make the most of participation

Deploy useful tools

Advocate for care

(4) Evaluate and learn SDM

Evaluate beyond outcomes

Share the evaluation

Seek joint improvement

Foster a conversation

The first step in implementing SDM in practice is to foster conversations that invite patients and clinicians to collaborate, supports their collaboration and leads to the formulation of a codeveloped care plan.

In this conversation, the clinician curiously works to understand which aspect of the patient’s problematic human situation requires action.4 This could be an unaddressed medical problem, such as a new symptom, concern, sign or complication. It could also be a change in life circumstances that affects how the medical condition manifests or that affects the practicality of the existing plan to address it. It could be signs that the plan of care in place is not helping, or even hurting, or that it has become impractical or unfeasible. The patient and clinician must collaborate to arrive at a useful formulation of the problem. For example, will the change in insurance coverage change the patient’s ability to afford the treatment prescribed given their income and other expenses they must cover? Is the increase in pain in the feet impairing living? Is the increase in pain in the feet impairing living? The inquiry then seeks to uncover the action that the situation requires.1 It may be necessary to conduct new tests, change the care plan, or provide additional support.

The process of noticing and responding is iterative,1 11 and continues until a response emerges that makes intellectual sense (ie, it is an evidence-based response12 to the situation as understood), practical sense (ie, it is feasible and minimally disruptive of personal and social routines) and emotional sense (ie, it accounts for the emotional dimensions of the situation and feels like the right thing to do now) to patients and clinicians.7 Confronting the actions available to respond to the situation may lead to reframing the situation itself and reformulating the problem to address. For example, a patient facing a cancer with a very poor prognosis and seeking a cure may discover that the treatments have a low likelihood of success and a high likelihood of harm. The unattractiveness of these options may lead to a recasting of the situation as one in which the patient is facing death because of cancer and now seeks ways to die well. The options identified in the first instance fail to be a sensible response to this new formulation of the situation and new options must be identified, uncovered, or invented. As it is cocreated and implemented, the patient is best positioned to provide feedback about the plan, as to its adequacy, its acceptability by itself and in interaction with other treatments and daily routines, and its efficacy in improving the situation.

It may be helpful to be aware of some stumbling blocks to fostering SDM conversations:

  • In some cases, that is, when there is no one technically correct solution to their problem, patients may not be aware that SDM is an appropriate method of care. Clinicians can helpfully state so and invite the patient into the process of collaborately figuring out together what to do.13 This renders irrelevant that the patient ‘did not go to medical school’, creating the space and momentum for collaboration.

  • When made prematurely, a clinician’s recommendation or a patient’s demand can abort the exploration for new insights; when unacceptable to the other party, these can complicate the patient–clinician relationship.14 Also, a desire to avoid conflict may lead to either party acquiescing, which is why policies (eg, guidelines, pathways, formulary restrictions, preauthorisations) and marketing campaigns (eg, detailing to clinicians, direct-to-consumer advertising and ‘ask your doctor about…’ ads) can unduly shape care.15 16

Purposefully select and adapt the SDM process

There are four distinct ways in which patients and clinicians can work together to address the patient’s problematic situation: (A) focusing on matching preferences, (B) reconciling conflicts, (C) problem-solving or (D) meaning making.4 Each of these forms of SDM is best suited to address one of four different kinds of problematic situations (tables 1 and 2). Clinicians need to be aware of these forms of SDM to intentionally select the form best suitable to respond to the situation at hand, avoid selecting the wrong one and nimbly switching to a different form when the situation becomes clearer or changes.17 In our observations, clinicians and patients who do SDM well, work within a form of SDM until a better one becomes apparent and they flexibly, gracefully and perhaps intuitively switch according to the challenges uncovered during the conversation.17 For example, a conflict requiring reconciliation (‘I will never use insulin because I am not allowed to use needles at my job’) can become a problem requiring solving (‘Is there a way to use insulin such that it is only administer at home?’).

Table 1

Forms of shared decision-making (SDM)

Table 2

Practical differences between the forms of shared decision-making (SDM)

Also in our observations, the situations adverse to care emerge when clinicians use an unhelpful SDM form or inflexibly insist on using a particular SDM form after it has proven unsuitable. This can be observed when, for example, a clinician, documenting advanced directives, requires from distressed patients and family members to select what they prefer from a menu of life-sustaining therapies.18 Selecting the right approach requires clinicians to be present, competent, flexible and attuned to whether the conversation is helping the patient with what they are struggling.

Support SDM

Finding and deploying useful, usable and desirable ways of supporting SDM in each encounter promotes the practice of SDM as a method of care.

Protect the space

SDM is work for both patients and clinicians.19 The conversation is the workspace within which this work takes place. The space for the conversation must be set up to be supportive of this work.20 Clinical spaces can be cluttered with visual (posters behind office doors, clinical equipment) and auditory (overhead announcements, ringtones) distractions. Demands for entries from the medical record system can interrupt conversations. Thus, clinicians must be deliberate about protecting the space and the time allocated for these conversations. This is less about new investments in interior design and more about securing agreements and arrangements (eg, team policy to avoid interrupting clinicians when in consultation with patients; minimise pop-up alerts and mandatory data entry in the design of medical records) that eliminate distractions, disruptions and interruptions. The setup should clearly signal the intention: the clinician and patient are here to have an unhurried conversation—not necessarily a long one—to work through what to do about the patient’s problems today and going forward.

Make the most of participation

Having set the stage for an unhurried conversation,21 it is necessary to determine who should participate in that conversation. Patients and clinicians in continuous relationships of care may be optimally situated to have unhurried conversations. When the issue requires specialised technical knowledge, or access to educational materials, longer consultations and decision-making tools, it may be optimal to bring into the conversation clinicians specialised in the matter, either to cocreate the plan of care with the patient or to assist the established patient–clinician dyad in their decision-making process. A similar choice needs to be made about the participation of informal caregivers, who in their roles at the patient’s side, often have expertise about and experience with the patient and may be responsible for the plan’s implementation.

Deploy useful tools

Clinicians and patients may want to thoughtfully consider which tools are allowed into the conversation, including specialised tools designed to support specific SDM forms that have shown to be useful, usable, and desirable. Given the situation at hand, different tools can support the decision-making process:

  • Self-management logs, patient-reported outcome trends, results from ancillary laboratory and imaging tests can all support the problem-solving mode of SDM.

  • Patients and clinicians could consider using home visits, photographs, narrative accounts of daily living, the ‘My Healthcare, My Life’ conversation tool,22 23 and other ways to develop a joint understanding of the social and economic challenges the patient faces routinely, and how these conditions promote or hinder health and the implementation of treatments. Stories and accounts of how patients lived their lives may also be helpful to their family and clinicians in determining together whether and for how long to implement intensive life-support interventions in the care of a critically ill patient.

  • Tools to support SDM conversations can help patients and clinicians select together which treatments to implement to reduce the risk of adverse disease outcomes.24 25 These tools should be easy to use, use helpful ways to communicate pertinent evidence and numerical risk information,25 26 and should support the conversation without intruding. Some tools which have been found to be useful in randomised trials are available free of use (see

  • Teach-back could be used to verify that patients and clinicians understood the information shared by each other.27

Advocate for care

Access and efficiency imperatives abbreviate and accelerate consultations to the point that SDM and other forms of care cannot be adequately implemented. Algorithms and guidelines may enable bypassing the messy process of co-creation, offering a right answer for ‘patients like this’, which may or may not fit ‘this patient’. SDM may get outsourced to third parties, offered to privileged patients (and less to those who need interpreters, racialised patients, patients with cognitive and sensorial challenges, and those seen in high-volume or understaffed clinics), or reduced to the distribution of SDM tools.28

Like careful and kind care, SDM is not a luxury. And yet, it often seems as if high-quality SDM is a method of care that healthcare cannot afford to offer everyone. Clinicians and patients must play an active role in advocating and working toward healthcare that enables and supports SDM. This work can focus on reprioritising care over efficiency, advancing unhurried care conversations,21 28 reorienting healthcare innovations to advance rather than replace SDM, and on ensuring SDM for all patients.29

Evaluate and learn SDM

When done well, SDM should contribute to improve the patient’s problematic situation. Being able to cocreate and jointly revise plans of care may reduce the risk of a poor-quality decision, that is, one that does not respond sensibly to the problem, fails to support patient goals and priorities, and maximally disrupts patient lives and loves.11 Doing SDM can deepen the relationship between patient and clinician and this relationship can offer resilience to adverse patient outcomes.30 In turn, joint evaluation of how well the patient and clinician are doing SDM can motivate improvement of SDM skills and further their partnership. In this way, care and learning to care are intertwined, and are both reliant on unhurried conversations and SDM.

Evaluate beyond outcomes

It is not adequate to judge the quality of the SDM process by patient outcomes, as the link between decisions and outcomes is weak as many outcomes result from highly complex interactions, multiple decisions over time, and chance. Short of general patient satisfaction questions, to our knowledge, there are no practical means available for external evaluators to assess how well a healthcare system, a clinician and a patient implemented SDM and how well this process contributed to advance the patient’s problematic situation. A way forward may require defining a good decision by the way it was produced (evidence based, cocreated), by the goals that animated the decision-making process (advancing the patient situation in a sensible way), and by the nature of the care plan that emerged from it (maximally supportive of the patient situation and goals, minimally disruptive of the live routines of patients and their community.11

Share the evaluation

Beyond external assessments, the most important evaluation needs to take place within the patient–clinician relationship. The patient and clinician may want to ask each other how well the conversation went and to seek feedback from each other about how they went about working out what to do, that is, how well they did SDM. This may be particularly necessary early in the dyad’s decision-making experience so that their performance can improve over time and be increasingly readier to face more difficult situations. By seeking feedback, clinicians exercise their humble commitment to meet the patient where they are and to care well for, about and with the patient.

Seek shared improvement

As with every other aspect of caring for patients, this method of care must continue to be subject of innovation and improvement,31 including the preparation of both patients and clinicians (and the healthcare systems within which they meet)32 to better contribute to the joint work of making care fit.11 The shared work of SDM demands that both parties learn from their experience. Since clinicians and patients with chronic conditions face a lifetime of decisions, this learning is life long and ongoing. Few opportunities exist to improve together. Clinicians can access courses in communication, but often these courses pay limited attention to the co-creation of a plan of care, instead focusing on explaining the plan to the patient. Patients are often trained to ask questions (eg, what are my options, what are their pros and cons, how likely are these pros and cons to happen),33 but there is little training about the different ways in which they can contribute depending on the forms of SDM used. Resources to improve the performance of both patients and clinicians, including joint skill building opportunities, need to be made available to promote high quality SDM.27 34


The number of tasks assigned to clinicians seem to increase in inverse proportion to the time allotted to execute them. In this context, SDM may seem like just one more box to tick, or a skill clinicians have no time to learn or use. But SDM is not an add-on. Clinicians are already engaging patients in conversations to work through a plan of action because that is what is required to formulate the best plan. Every consultation with a patient is an opportunity to get care right—intellectually, practically and emotionally—for that person using SDM as a method of care. In these conversations, patients and clinicians can find problems that matter along with possible ways of addressing them, deciding among the possibilities, and putting it all together in a plan that the patient wants, is likely to help, and is feasible and sustainable. Within the constraints of any situation, including systemic constraints, SDM is a method of creating the best care, it is also the human, kind and caring thing to do—the sort of thing that breathes life, joy and purpose into the practice of medicine.

Ethics statements

Patient consent for publication



  • Twitter @vmontori, @MarleenKunneman

  • Contributors This article was conceived at a meeting of all the authors. VM prepared the first draft and all listed authors provided critical review and edits. All authors approved the final version.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests The authors participate in research developing, evaluating, and implementing interventions promoting shared decision making in clinical practice. These research is funded by research grants or contracts awarded by governmental and nonprofit organisations to their institutions. They do not derive personal income or have patents related to these activities. MK had financial support from the Dutch Research Council (NWO) and The Netherlands Organisation for Health Research and Development (ZonMw) (#016.196.138).

  • Provenance and peer review Not commissioned; externally peer reviewed.