CHAPTER 9

Shared Decision-Making

In Chapter 8, we talked about motivational interviewing, a process in which we start from a place of knowing what is best for the patient. For example, virtually all clinicians would agree that patients should stop smoking to improve their health. In contrast, in shared decision-making (SDM), there may be more than one right answer. Shared decision-making is the practice of eliciting a patient’s informed preferences regarding a screening or treatment decision. Our goal is to help patients choose the answers that will work best for them.

Decision-making is complex. Think about a common decision we make such as buying a car. We usually do research before we go to an auto dealer. We talk to our friends about why they like their cars, do research online about the pros and cons of different models, and brainstorm the choices in our price range. We clarify what we care most about in making our decision. Do we want a hybrid? Do we need a four-wheel drive? What are the safety ratings? How cool will we look driving it? And then we get to test-drive the cars we are considering before making our purchase.

Making decisions about our health is even more difficult than buying a car. Patients are often asked to choose between multiple options when they are reeling from a new diagnosis that can represent bad news. They don’t have the detailed information about the choices open to them or the risks and benefits of each. And patients can’t test-drive a surgical procedure to see how well it will work for them.

That is why patients and clinicians need to work together as a team. We are the medical experts who know the evidence-based information, and patients are the experts on what matters most to them. Just as figuring out the right car purchase depends on the buyer’s preferences, many decisions in healthcare involving screening tests, surgery, and chronic disease management depend on the patient’s values and preferences, rather than solely on the clinician’s perspective.

How Well Do Clinicians Do Shared Decision-Making?

Clinicians may feel that they’re already doing effective SDM, but the evidence documents poor communication, knowledge gaps, and lack of attention to patients’ preferences.¹ In one representative study, preferences were elicited from less than a fifth of patients.² Clinicians often think that they know a patient’s preferences without directly asking, but research in end-of-life care, screening, and surgical decisions does not support this contention. In multiple studies, the majority of physicians simply recommended screening tests, rather than educating patients first.³ When clinicians did attempt to educate, most discussed the pros of screening, and only a small minority presented the arguments against screening. Less than half of patients could correctly answer even one of the five knowledge questions that experts felt were essential in understanding how to make an informed treatment decision.⁴ Yet in this era of transactional, rather than relationship-centered, encounters, we often find ourselves saying to a patient, “So you need a colonoscopy and a PSA test. You’ll get a note from the gastroenterology clinic, and go to the lab.” A lack of informed decision-making is still more the rule than the exception.

A Better Way

Care that incorporates and reflects individual patient preferences, needs, and values is a key foundation of high quality healthcare.⁵ Patients who are more engaged in making decisions about their health have better health outcomes and healthcare experiences.⁶ As we focus specifically on skills for shared decision-making, it is useful to have a structure like the Three Talk model with its team talk, option talk, and decision talk phases.⁷

SHARED DECISION-MAKING: A MODEL FOR CLINICAL PRACTICE

Although conversations with patients are rarely linear and one may need to go through the different parts more than once, the model shown in Figure 9.1 provides a useful roadmap.

FIGURE 9.1 Three Talk Model

Used with permission. ©2017 Dartmouth Shared Decision Making Center.

Team Talk

The team talk phase is where you can invite the patient to be on the same team with you. We know that even well-educated, wealthy professionals are often intimidated in a clinician’s office or the hospital. Patients worry their care will suffer if they are not deferential or if they disagree with the clinician. Real teamwork is not possible in such a setting.

With some awareness, though, it’s fairly easy to create a collaborative environment. To convey the importance of working together as a team, you might say something like:

Now that we have a diagnosis, we are facing a decision and need to figure out what to do next. Fortunately, we have options. While I know the science about the options, you are the expert on what matters to you. Different treatments have different risks and benefits. I want to understand what matters most to you. I need your help in that. Working together, we can make the best decision for you.

Option Talk

During the option talk phase, you will elicit and clarify patients’ preferences and explain the benefits and risks of the different options. Once they understand the tradeoffs of the different screening or treatment options, patients often state more clearly what is most important to them. Your job is to understand patients’ ideas, values, and unique life circumstances in order to help them make the decision that is best for them. We’ll delve more into this more complex phase in the risk communication and decision aids sections below.

Decision Talk

The third phase, decision talk, will take place once a patient understands the pros and cons of the options, and the clinician has heard their informed preferences, as shown in the following case history:

Ms. Parker was a formerly healthy Army nurse when she was diagnosed with early stage breast cancer. Medical literature confirmed that either a mastectomy or a lumpectomy with radiation would have very similar outcomes for her.⁸

After hearing her options and the risks and benefits of each, Ms. Parker was extremely clear about her choice. She definitely wanted to preserve her breast and chose to have a lumpectomy. I (Nan Cochran) sent a letter to her surgeon clearly explaining her preference to have a lumpectomy. However, he used a paternalistic approach and told her she had to have a mastectomy, which she reluctantly underwent. When I saw her again two months later, she was clinically depressed. She refused to allow her husband to see her disfiguring scar, and they no longer had a sexual relationship. She tearfully complained, “I was mutilated.” She felt like her life had been destroyed, because her preferences had not been respected.

Ms. Parker’s story painfully describes the “preference misdiagnosis” epidemic,⁹ which is unfortunately very common in U.S. healthcare. Key domains of medical professionalism include a respect for patient autonomy and a willingness to be sensitive and responsive to differences in our patients, including the diversity of their values and preferences. Yet in medicine, there is still a widespread assumption that the right decision is a matter of science alone. Instead, we need to ask patients, “What is most important to you in making this decision?” Patients often have multiple values, so follow that question with: “And what else is important?”

Decisions regarding a wide range of healthcare issues vary depending on the individual patient’s attitudes, life circumstances, and preferences. For example, treatments for hip or knee osteoarthritis are preference-based decisions. Focusing on severe knee osteoarthritis illustrates this point clearly. Patients who want to maintain an active lifestyle or don’t want to take medications regularly may be very interested in having a knee replacement to improve their symptoms. Another patient, with equally severe symptoms but who is risk averse or less active, may choose to continue taking medicines and be totally unwilling to consider a knee replacement. If clinicians do not fully elicit patient preferences, or if they fail to inform patients of the pros and cons of each treatment option, patients may feel forced into a decision, like Ms. Parker, that is contrary to their wishes. Clinicians wrongly assume that patients will speak up if they disagree with a recommendation.¹⁰ Patients must be fully informed and engaged in the decision-making process in order to make the decision that is right for them. This is where Option Talk is critical—and why it is not easy.

Risk Communication

Medical research is difficult to translate transparently, especially when working with patients challenged by health literacy. Numbers are rarely discussed with patients. However, when we use qualitative and vague terms such as “infection is rare” or “it is unlikely you will have serious side effects,” patients tend to overestimate benefit and underestimate risks.¹¹ Younger and better-educated patients desire more information and a greater role in decision-making and are more satisfied and adherent when they receive this.¹²

It is critical to provide risk and benefit information to patients in clear, easily understood language. Those challenged by health literacy will understand information better when it is accompanied by graphics and pictures.¹³ High-quality decision aids can present medical data in an understandable format using language that is less confusing to patients.

Decision Aids

Decision aids bridge the comprehension gap between patients and clinicians. These are written, video-based, or online tools designed to communicate unbiased, easily understandable, and up-to-date evidence about treatment or screening options to patients. More than 115 trials of decision aids including more than 34,000 patients, addressing 23 different screening or treatment decisions, have consistently demonstrated that these aids provide patients with greater knowledge about treatment options, more accurate risk perceptions, and greater comfort with their decisions. Decision aids also lead to high participation in decision-making and leave fewer patients undecided.¹⁴ An example of a decision aid to treat osteoporosis, authored by the Mayo Shared Decision-Making Center, is depicted in Figure 9.2.¹⁵

Let’s walk through this decision aid together. But before we do, you need a little extra background.

Think about an elderly person you know who has fallen and broken a hip, or one who cannot straighten their back and has to walk with a hunched-over gait. Both of these problems are likely due to osteoporosis, which is a weakening of the bones that is common in the elderly. The reason clinicians want to treat osteoporosis is to reduce the risk of broken bones: half of all elderly people who suffer hip fractures die within one year.

Now consider the case of Ms. Johansson, a 65-year-old woman of Swedish descent who broke her wrist in a fall several years ago. Because of her age and history of a broken bone, she has two risk factors for osteoporosis. This means she is in the “elevated risk” category for another fracture in the future.

In the leftmost panel of Figure 9.2, you can see that roughly 20 percent of women like Ms. Johansson will break a bone over the next 10 years. Clinicians typically look at this information, think the risk is really high, and try to convince Ms. Johansson she needs to take a medicine to decrease her risk.

FIGURE 9.2 Osteoporosis Decision Aid

Used with permission. ©2017 Mayo Foundation for Medical Education and Research (MFMER). All rights reserved.

A shared decision-making conversation, using the decision aid, would look very different. If clinicians review information on a decision aid that displays treatment options using graphics and pictures, patients understand their risks better. As Ms. Johansson’s clinician, you might say:

“Let’s look at the benefits and downsides of taking a new medicine to treat your osteoporosis. The left side of the figure tells us in 100 women just like you who do not take additional medicine, 20 will break a bone due to osteoporosis over the next 10 years, and 80 will not.

“The right side of the figure shows how many people will benefit if they take medication regularly to treat osteoporosis. You can see that in this group, out of the 20 who would have broken a bone, 12 will still break a bone in the next 10 years. So 8 women in 100 will avoid having a fracture.” Then pause and elicit questions from the patient.

“We also need to talk about the downsides of medicine for your osteoporosis, since all medicines have side effects. In the middle of this figure are the cost and the most common harms of the medication.”

The decision aid does not tell a patient what to do. After reviewing a decision aid, clinicians should routinely ask, “What is most important to you in making this decision?” This is probably the most important question, and it demonstrates the crux of shared decision-making. Our goal as clinicians in these instances is to be agnostic in our recommendations for these decisions to our patients, not to make the final decision for them. Having a detailed understanding of a patient’s preferences and life circumstances is therefore essential to help them make the best decision.

Conclusion

This chapter reviewed aspects of team talk, option talk, and decision talk that are important facets of successful shared decision-making. We need to learn to work with patients as members of the same team in order to understand their values and to help inform them to make the best medical decisions possible. We have illustrated the chapter with some important tools, but shared decision-making involves substantially more than just providing decision aids to patients. These tools are only adjuncts to our counseling. Many challenges persist in educating clinicians about using these techniques, and implementation of SDM in typical practice settings remains burdensome despite its clear benefits. Modifying systems of care, developing more numerous and widely available decision aids, and training other team members to help patients with preference-sensitive decisions are all essential steps in working toward providing high-quality care using a relationship-centered approach.