Editor's note: Case 1 below was edited to clarify that John chose the FIT-DNA test.

Shared decision making (SDM) is an essential component of modern primary care. At its most basic, SDM is a communication strategy in which the clinician informs the patient that they have a choice about a screening or treatment, the patient tells the clinician their thoughts about the options, and the two make a decision together about how to proceed. The bidirectional nature of the conversation is what makes it unique: The clinician provides knowledge of the scientific evidence, makes a recommendation, describes potential risks and benefits, and then asks the patient about their preferences, values, and goals. This sharing of decision making is relatively new in medical care and is an improvement over the previous process in which the clinician made a recommendation and hoped that the patient followed it.

Data suggests that when clinicians use SDM, patients are more knowledgeable about their health decisions, more confident in specific decisions they make, and more likely to adhere to the plan.¹ (See “Why use shared decision making?”) 

SDM obviously does not work for medical emergencies or situations when there is one clear best medical decision (e.g., insulin for diabetic ketoacidosis or antibiotics for sepsis). But for a broad range of decisions, where there is no clear “right” answer or there is more than one medically reasonable option, SDM is important.

Research has shown that SDM does not have to be time-consuming to be effective. It can be performed in about 2.5 minutes even when discussing complex decisions such as whether to have surgery.² This article provides a simplified model for SDM, resources to help patients make decisions, and three case studies to show how the model works.

THE COD MODEL OF SHARED DECISION MAKING

There are different models of SDM,³ some more complex than others. Most include requirements of joint decision making, information about the choices that could be made, potential harms or benefits of each choice, and a guide for completing the decision. These models can be excellent but are often complex, with long acronyms listing many steps (e.g., the SHARE model by the Agency for Healthcare Research and Quality⁴). But in 2012, Elwyn and colleagues described a simpler SDM model made up of just three steps: choice talk, options talk, and decision talk.⁵ (They later changed “choice talk” to “team talk” to reflect the team-based nature of health care.⁶)

Our team adapted this into what we call the COD model of SDM:

• Choice talk. Tell the patient something like “This is your choice,” or “You have a choice.” Saying this up front is important because many people do not understand that the decision is really up to them. As the clinician, you can provide information, guidance, and support, but the patient gets to choose.

• Option talk. Provide information about each choice, potential risks and benefits (e.g., “If you choose to do low-dose CT scans to screen for lung cancer, we may find nodules that will require more frequent screening”), and potential outcomes of the choice (“This may cause you anxiety about nodules that are actually harmless”).

• Decision talk. Put the focus back on the patient. What do they want to do? Why? Do they have questions? What is important to them (e.g., avoiding medication side effects, having reassurance, avoiding costs)? How can you help them make a choice?

Patient decision aids can help make this process go more smoothly.⁷ Decision aids are paper or online tools with information specific to a screening (e.g., abdominal aortic aneurysm screening) or treatment (e.g., taking an antidepressant medication).^8,9 Some can customize risks and benefits using a patient's specific information, such as age and disease history. Decision aids improve patients' knowledge and increase their confidence in their decisions.⁷

Patient decision aid resources are available online and several different companies also make decision aids compatible with many electronic health record (EHR) systems. One benefit of integrating decision aids into an EHR is that you can automatically upload patient data into them to make it easier to calculate a specific patient's risks. Also, you can save a record of the conversation in the patient's chart.^10,11

THE IMPORTANCE OF PATIENT VALUES

While decision aids can help explain the evidence to patients, evidence is not the only factor. The other key aspect of SDM is getting your patients to talk about their values. Patient values can be related to a myriad of factors including current health status, previous experiences with and trust in the health care system, race, ethnicity, socioeconomic status, and religious beliefs.^12,13 Patient values may fall into the following broad categories:

• Life philosophy (e.g., fatalism, desire to not be a burden on others, desire to avoid suffering),

• Life priorities (e.g., health, career, finances),

• Sociocultural values and personal background (e.g., religion, family history).

To help facilitate conversations about values, draw on what you know about the patient from previous interactions (e.g., “Remember when we talked about what kind of critical care you want and don't want when we filled out your advance directives?”).

You can also simply ask, “What is important to you?” If they say, for example, “Knowing whether I have cancer,” that helps you decide together whether a screening is an appropriate choice. If they say, “Not wanting to suffer from side effects of medications,” that helps you decide together whether a certain treatment is worthwhile.

You can also lay out specific options, with risks and benefits, and then ask, “What are your thoughts?” Maybe they will say, “I am scared about that procedure,” “My friend had a terrible time with that medication,” or “What if the test is positive?” Any of these responses open the door to a conversation that tells you more about the patient's values. (See “Patient value prompts.”)

Patients who are truly at a loss as to how they want to proceed may ask, “What would you do, Doc?” Many clinicians are wary about sharing their opinion as it may overly influence what a patient decides, and their values may not be the same as ours. One way of ensuring that the patient is making a decision that is in line with their own values is to detail the potential harms and benefits of any decision first, and then share your opinion using your medical knowledge as well as your knowledge of the patient based on your longitudinal relationship.

THREE CASE STUDIES

Let's put the “COD” model into practice using three case studies.

Case #1: colon cancer screening. John is a 45-year-old man who presents for a physical. He is healthy and has no relevant family history. He takes no medications. You tell him that the general recommendation is to start screening for colon cancer at his age.

“It is your choice whether to be screened or not for colon cancer,” you say, “and there are several different screening options with their own pros and cons.” (Choice talk.)

“I'll do whatever you tell me to do, Doc,” John says.

“I'm happy to make a recommendation but first would like to tell you about all the options,” you say. “Would you like to hear about the different tests?”

John nods, so you proceed to tell him about colonoscopy, virtual colonoscopy, fecal occult blood testing, fecal immunochemical testing (FIT) and stool DNA fecal immunochemical testing (FIT-DNA) — what they can do and what they cannot do, what the procedures entail, and their sensitivity and specificity. (Options talk.)

John says, “I am leaning toward getting a colonoscopy because that seems like the best test. I especially like that if they see a polyp during the procedure they can remove it and prevent colon cancer. But I have heard a lot of bad stuff about the prep, I can't really take time off work, and my wife would also have to take time off so she could drive me home.”

You tell John more about what he can expect from the colonoscopy prep. It is clear he is weighing his options carefully.

“I think I will start with the FIT-DNA testing,” he says. “If it is positive, I know that I will need a colonoscopy, but if it is negative, then we can have this conversation again in three years.”

“Okay, it sounds like you have made a decision,” you say. “I will go ahead and order that test for you. I hope it is negative. Remember, even if it is positive, it doesn't mean that you have colon cancer. But, yes, you would then need a colonoscopy.” (Decision talk.)

“I understand,” John says. “Thanks for the help making this decision, Doc.”

Case #2: hyperlipidemia management. Peggy comes to see you to discuss her lipids. She is 63 years old and has a past medical history of hypertension that is controlled with one medication. She recently had high fasting lipids. You entered her information into an online decision aid and found that her 10-year risk of having a heart attack or stroke was 9.4%.

You tell her, “You have a choice about whether to start on medication to lower your cholesterol.” (Choice talk.)

Peggy says, “I've heard bad things about statins and would like to avoid taking one if possible.”

“Let's talk about different treatment options,” you say. You proceed to discuss different options including lifestyle changes as well as taking medications. (Options talk.)

“It is important to me to take as few medications as possible,” Peggy says. “I am worried about side effects, and my quality of life is really important. I am planning to retire this year and don't want to be in pain due to medications. I think I want to try to bring down these numbers by exercising more and eating healthy. If I can lose 10 pounds, even better!”

“Okay,” you say. “You are making a decision based on the data and your priorities, and I think it is reasonable. How about you work on exercise and diet and we can recheck your lipids in three or four months?” (Decision talk.)

“Great,” Peggy says. “Thank you for being supportive of my goals.”

Case #3: GLP-1 medication. Eric is a 46-year-old man who is seeing you about his uncontrolled diabetes mellitus. His last A1C was 8.3%, and he has been unsuccessful losing weight. He presents today to discuss medication. He has tried metformin in the past but didn't tolerate it well due to gastrointestinal issues.

“There are new treatments we can try if you choose to,” you tell him. “The GLP-1 medications work really well for diabetes and may also help you lose weight.” (Choice talk.)

“Really?” he says. “That sounds great. Do they have any side effects?”

You explain that the GLP-1 medications work differently than other treatments for diabetes, that some are injected and some are oral. You outline what is known about the potential side effects, and you explain that GLP-1s can be quite expensive compared with older medications and his insurance company may not cover all of them.

“Whoa,” Eric says. “I don't like needles. And I have a high-deductible plan, so I might end up spending a lot on these medications. Is there anything else we can try?”

“You could continue to work on losing weight and exercising more,” you say. “Or we could try metformin again or a different medication. I can also have you talk to a nutritionist.” (Options talk.)

“Thanks, Doc,” Eric says. “I think I'd like to continue trying to lose weight. If in a few months it's not working, I would consider trying metformin again or looking into the cost of the GLP-1s with my insurance.”

“That sounds like a very reasonable plan, Eric,” you say. “I'll put in an order for another A1C test in three months, and I encourage you to talk to your pharmacist to get more information about your out-of-pocket costs for a GLP-1.” (Decision talk.)

PATIENT-CENTERED CARE

SDM is not always as easy as these case studies illustrate. Patients may sometimes struggle to grasp the risks and benefits of various options or have trouble weighing their competing priorities. But using the COD model, which is easy to remember and implement, ensures that you're helping patients navigate these decisions in a comprehensive way that centers their values and priorities.