Making Decisions With, Not for, Patients

Selma F Mohammed, MBBS; Victor M Montori, MD


May 19, 2015

Editorial Collaboration

Medscape &

This feature requires the newest version of Flash. You can download it here.

What Is Shared Decision Making?

Selma F Mohammed, MBBS: I am Dr Selma Mohammed, advanced cardiology fellow at Mayo Clinic. During today's video, we will be discussing shared decision making. I am joined by Dr Victor Montori, professor of medicine and director of the Knowledge and Evaluation Resource Unit at Mayo Clinic. Victor specializes in how knowledge is produced, disseminated, and taken up in practice and how this leads to optimal healthcare and delivery of patient outcomes.

Let's start with the definition—what is shared decision making? Why do we care as providers, patients, or society at large?

Victor M Montori, MD: Shared decision making is an approach to care through which patients are helped to recognize that there may be more than one reasonable way of proceeding, of going forward with treatment or tests, and how to choose between those reasonable approaches. It depends on which approach is best for this patient, on the patient's situation, and what the patient values. As clinicians, we can't be experts on every patient's situation and values, but the patient is the expert on his or her situation and values; and the clinician can be the expert about the nature of those options, strategies, and approaches—what are the pros and cons; the relative merits. We have two experts coming together and having a conversation and deliberating together about what is the best course of action, eventually coming to a consensus about how to proceed. That is shared decision making.

Why should we care about it? Because we don't want to make a mistake, and the mistake that we might make is giving people tests and treatments that they wouldn't choose if they had our expertise, given their situation and values. It helps us prevent that error. It's the ethically correct thing to do because, at the end of the day, patients will live with the consequences of these choices, and it is difficult to impose those consequences on them without their input.

Dr Mohammed: It is critical for patients because we empower them, but what about society?

Dr Montori: The large-scale version of shared decision making is democracy. The liberty of the democracy in which people are activated to choose the best course of action for their communities, at the level of the individual patient, takes the form of shared decision making.

Do Patients Really Want to Make Decisions?

Dr Mohammed: Are we not using shared decision making now?

Dr Montori: We have thousands of video recordings at Mayo Clinic and elsewhere of patients and clinicians having discussions about treatments, and we have used checklists to determine the extent to which patients and clinicians are engaged in shared decision making. It is a very rare situation in which we see shared decision making in full representation. Patients and clinicians are trying, sometimes, to engage each other in making decisions together, but they lack the tools and the skills and the opportunities to do it properly.

Dr Mohammed: It is a new science to us, and it isn't taught in medical school.

Dr Montori: No, even when there are courses about how to communicate better with patients, a lot of effort goes into setting the agenda ahead of time and then providing explanation and education. But there is very little education about how we can engage patients as partners in making decisions together and also in learning to what extent should we engage patients for this decision and in this circumstance. Not everyone will be interested in full participation, and the clinician has to be very empathic with the patient to figure out the maximum level of participation that is appropriate and desirable for this decision at this time.

Dr Mohammed: Do we know how much participation and decision making patients want, and how can we tell or get an idea of how much they want?

Dr Montori: Surveys have been done for many years now that tell us that patients want a lot of information—98% to 99% want information about the options—but a much smaller proportion would like to be involved in decision making. The problem with those surveys is that most patients have not had any experience with shared decision making. It's like asking people if they would like more pistachio ice cream. If they have never had it, they have no idea whether they want more of it. The same concept applies to clinicians. Surveys are beginning to appear in which patients have been given an opportunity to, for instance, watch a video of shared decision making and then are asked, have you had that experience? And the proportion of people reporting that they have had it is low—approximately 20%. When asked whether they would like it, the proportion goes way up. In our studies, after we exposed people to shared-decision-making experiences, we asked them, "Would you like shared decision making? The proportion of people who say yes exceeds 70% to 90%, depending on the decision that we have facilitated. The same is true with clinicians—70% to 90% of clinicians endorse the idea of having more shared decisions.

Dr Mohammed: It's amazing! Both sides want to learn about it and engage in it.

Dr Montori: They do, if they have tools designed for them to do it efficiently and effectively in practice, which is a development that has appeared only recently.

All Patients, or a Select Few?

Dr Mohammed: Which patients should we offer it to? Is it appropriate for all patients and under all circumstances?

Dr Montori: The biggest mistake we can make is to assume that someone who has difficulty speaking the language, is of low socioeconomic status, or has had limited education is not appropriate for shared decision making. In Hennepin County, north of inner-city Minneapolis, homeless patients who have been given the opportunity to participate in shared decision making (for instance, about choosing their diabetes medication) have been able to effectively participate. In a meta-analysis[1] of 10 years of work that we have done here at Mayo Clinic, we have noticed that patients who come in with low socioeconomic status benefit the most from their exposure to shared-decision-making tools. Instead of introducing more disparities in care by offering shared decision making only to those who we think are going to be able to partner with us, we have the opportunity to advance the notion of equality in healthcare by using these tools as an equalizer for the engagement that patients with different socioeconomic status bring to the table.

Dr Mohammed: We shouldn't limit it. It should be widely applicable.

Dr Montori: Yes; at least, we should not use socioeconomic status in deciding whether to offer shared decision making. But what about illness? If a patient is very sick or very emotionally distraught, it would be abusive for the clinician to say, "You need to make this decision." That's not caring. We actually do not advocate for shared decision making. What we advocate is for empathic decision making. The clinician is constantly reading the patient and deciding with the patient how much or how little involvement is right for this patient right now; and of course, many patients bring family with them who help make decisions. That skill needs to be expanded not only for including the patient but also in appropriately including all the family members—a big challenge.

Dr Mohammed: It is and it's more than science. It's science and art, and it requires a lot of experience and expertise.

Dr Montori: It all starts by caring enough to engage the patient in the decision making. Many people argue that we don't have time for this in the consultation. I would rather just tell people what I think is best for them and then they can decide whether they want it or not. Well, that's not very caring. It might be efficient, but it does not reflect the best that we can offer.

The Time Factor

Dr Mohammed: Maybe with more practice, we will be able to integrate this approach in our limited time for patient encounter?

Dr Montori: We have tested this again, in over a decade of work. Our tools are designed for the context in which they are going to be used, and most add two to three minutes to the consultation. They are efficient. They can be used. But this is not something that you can just give to the patient and have the patient decide on his or her own. This is something that requires an investment of those two to three minutes in being present in the consultation and engaging the patient. The tools that we have designed are efficient and can be used, and we have used them in the emergency department (which is a fast-paced environment) as well as in primary care, specialty care, and in the hospital. They are not identical. Each tool is designed for the context in which it will be used by the professionals who will use it.

Dr Mohammed: It's time worth spending. It's of high value during our clinical encounters. Do you think applied shared decision making would make clinicians less likely to adhere to guidelines?

Dr Montori: It's quite possible. One thing that we have discovered with these tools is that clinicians are able to perceive very effectively that adhering to the guidelines may not be the right thing to do for this patient because of the patient's situation and the patient's values at this point in time. The tools almost always will make the clinician more patient-centered. The question is—what happens when patient-centeredness is in contradiction with being adherent to the guidelines? Those are issues that need to be resolved. We are working with guideline-development teams for them to begin to appreciate those issues and recognize that where there is a class I indication for something, there are some patients for whom it is not a class I indication. This is one situation in which patients may want to be involved in shared decision making, and that should be the class I indication for that particular case. In other words, a strong recommendation should be for shared decision making in that particular case. We have seen this all the way from statin use to, for instance, atrial fibrillation and anticoagulation treatment, to whether a stent is appropriate for the management of stable angina.

Dr Mohammed: Some of our recent guidelines emphasize a team approach to decision making, but perhaps an integral part of this team should be the patient, and that would be shared decision making.

Dr Montori: Yes, the guidelines are beginning to use that kind of language, and they are also beginning to specifically ask for shared decision making. But what is unfortunately lacking in those guidelines is the support that is necessary for patient and clinicians to actually do it. It is assumed that if you say "shared decision making," it is a conversation that any clinician can have. What we are seeing is that it's actually quite hard to convey the evidence to the patient in a way that will not overwhelm the patient and will invite the patient to participate.

Tools to Ease the Process

Dr Mohammed: What tools are available today for shared decision making? Can you elaborate on those?

Dr Montori: There are more than 500 different tools to support relevant decision-making situations that are common in practice. However, most of those tools have not been tested in the clinical encounter to help those conversations. Most of those tools are patient-education materials that are given to the patient with the assumption that if the patient were to be "in the know," they will come to the consultation more prepared to have a discussion. However, that doesn't work. The consultations are still scary places, and patients can read everything; but when they come to the consultation, their minds will be blank. So it's better to support the conversation as it happens during the visit.

In terms of tools for the visit, there are quite a few, and many of them are beginning to appear in cardiology. The kinds of tools fall into two categories. One type are tools that help us understand the issues that matter to the patient and then decide which interventions are more consistent with those interests. There are also tools that communicate risk and help decide how we to reduce those risks.

Let me show you the tool that we have developed for deciding whether a statin is right for a certain patient. This particular tool is for a 67-year-old man with no cardiovascular risk factors, a relatively benign lipid profile, and blood-pressure risk. The way we will use it is to say, "Mr Jones, you are in a room with 100 people like you. As you can see, if we close the door for 10 years and we open it 10 years later and we count, we will see 78 people in green who, during the 10-year period, would not have had any coronary events or cerebrovascular events. You will see that there will be 22 people in orange who, during that 10-year period, would have experienced a cardiovascular event. This is very important. We say, "Mr Jones, we do not know whether you are one of the green ones or one of the orange ones. If we were to give a statin to 100 of you, then close the door and open the door 10 years later and count, we will see that the 78 people who were not destined to have a cardiovascular event would have taken a statin. Their cholesterol levels would have gone down, but they would not benefit because they were not destined to have an event that would be preventable by statins. Of the 22 people who were destined to have a cardiovascular event, these five in light blue would avoid their destiny by taking a statin regularly. Then there will be 17 people here in orange who would still have a cardiovascular event despite taking statins." As you can imagine, many clinicians would look at this and say, "Well, there must be something wrong; statins are much more effective than this," when, in fact, this is the 25% risk reduction that is seen across the board for cardiovascular events with the primary-prevention dose of statins. And we will tell Mr Jones, "We do not know whether you are one of the green ones, one of the light blue ones, or one of the orange ones." We now have 10 years of experience with this tool, and its use is picking up because the guidelines finally became risk guidelines rather than low-density-lipoprotein [LDL] target guidelines.

This tool is being used about 10,000 times a month all over the world, most of it the United States. We have a video and clinical-trial data that show that these tools help clinicians and patients have discussions. Patients can participate in discussions about this, but not if you tell them, "Mr Jones, your LDL cholesterol is 101 mg/dL. According to the guidelines, it should be 100 mg/dL. We are going to start medicine and check in 3 months whether your cholesterol is lower than 101 mg/dL." The patient can only say, "Whatever you say, doc," and then will come back and see what his or her cholesterol level is. If it's lower, the patient is going to say, "I'm benefiting from statins," which as we have seen from the decision aid is actually not the case for most people who take it for prevention. So, we add transparency. We add clarity. We calibrate risk information for the patient and also for the clinician. As a result, people are making informed choices, the dogmatic "You must take a statin! Your cholesterol must be this!" has been eliminated, and patients can participate in the choice.

Dr Mohammed: Fascinating. It is taking the discussion to a whole different level about patient preferences, values, and goals and involving them in their own decisions.

Dr Montori: The hope is that it will improve adherence and improve outcomes; and unfortunately, we are yet to demonstrate that that is true. But more studies are being done at the moment about that.

Dr Mohammed: Thanks, Victor, for these very important insights, and thanks to our viewers here on on Medscape Cardiology.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.