For Empathic Decision Making


By Victor M Montori, Ian Hargraves, Annie LeBlanc

Policymakers fashionably prescribe shared decision making for patients who face fateful decisions. These patients have two or more medically reasonable courses of action that differ in important aspects. The extent to which these aspects differ in ways that matter to each individual patient justifies patient involvement in the decision-making process. Similarly, the extent to which clinicians can accurately predict the values and preferences of informed patients reduces the value of shared decision making. Only in circumstances where the distribution of patient preferences is very narrow can clinicians correctly deduct patient preferences (e.g., analgesics vs. no intervention for moderate to severe pain). This is often the case when the pros and cons of alternative courses of action are well known, their likelihood estimates are based on highly reliable research evidence, and difference between the benefits and the potential harms and inconveniences is large and clear. In such situations the distribution of patient preferences will be narrow enough that most clinicians can assume correctly what most patients will want. At the extreme, these decisions will seem purely technical, where the right course of action is apparent to those with a good understanding of the situation. This would include professionals with pertinent training. In situations that cannot be resolved by the application of technical knowledge, patients, when informed, will exhibit a range of preferences. It seems appropriate then that patients and clinicians partner to share information, deliberate, and arrive at a decision together. We call this process shared decision making.

Proponents of shared decision making assume that most clinicians and patients, when given the tools, time, and supportive setting necessary, will be able to implement shared decision making. Reality seems to behave differently: surveys suggest that patients are not universally inclined toward shared decision making, clinicians are often portrayed as barriers to this process, and environments have electronic medical records, phone calls, time pressures, competing demands, and noise that conspire to interfere with shared decision making. What’s going on if patients and clinicians aren’t adhering to the shared decision making prescribed on their behalf?

Our group, the KER UNIT, characterizes shared decision making as a conversation – an activity in which patients and clinicians turn with one another (the etymology of conversation—versare turn; con with). In conversation, the options with their attributes or issues are in dynamic interaction as the patient and clinician consider them and experimentally try them on. This highly interactive dynamic requires the active engagement and involvement of the patient and clinician. This turning-with of patients and clinician is the dance of shared decision making.

The clinician is used to contemplating the situations of patients and making tough decisions routinely; but for this patient, at this time, the task is anything but routine. Thus, it is natural to delegate to the more experienced and emotionally detached of the two the task of organizing the decision-making conversation. The clinician, leading the dance, will identify that a decision needs to be made, the relevant options and their relative desirable and undesirable features, and will invite the patient to consider these options and features. But, to what extent are patients willing and able to engage in deliberation?

We propose that the adequate way of answering this question is through empathy. In suggesting empathy we do not mean that clinicians should empathically divine the right decision for the patient; quite the opposite. We are suggesting that the co-creation of decision also involves the co-creation of the patient-clinician relationship and the conversational environment in which each decision is made. Empathy directs attention to the clinician’s active role in finding the right relationship and stance to join this patient at this time in decision making. Clinicians are trained and are expected to exhibit empathy when interviewing and examining patients, responding to patient concerns, and delivering bad news. The role of empathy in supporting decision making has not been fully discussed, to our knowledge. In this case, empathy requires attention to the situation of the patient and to the cues, verbal and nonverbal, the patient offers as the clinician invites the patient into the deliberative process. Some patients may be able to partner fully and co-create the decision; others may engage with the information, but delegate the rest of the tasks of deliberation and decision taking to the clinician. This is the expression of a preference that is being constructed on the spot (it follows that this preference cannot be adequately assessed with a survey tool, before the encounter and therefore out of context). The appropriate stance in the conversation is available to the clinician in subtle signs that the clinician can pick up through empathic attention to the patient. Focus on who the patient and clinician are, and can be, for each other in this conversation allows us to respect that the same patient may be willing to co-create one decision while preferring a lesser role for the next. The challenge for the clinician is to correctly respond, in real time, to these emerging preferences.

Shared decision-making tools produced for use during the clinical encounter need to account for this clinical task and be designed to support empathic decision making. When encounter tools offer too much information or script a step-by-step decision process, they may inadvertently limit the ability of the clinician to empathically guide the process. When tools are used in preparation for the visit, clinicians may assume that completion of the tool and associated worksheets signals that patients are fully engaged and ready to make decisions. That a tool should enable and support empathic decision making is not currently a requirement for their design of decision aids, or a metric for their impact.

In summary, shared decision making is one of an infinite set of ways in which patients and clinicians can engage in conversation about fateful decisions without a technically correct answer. To create the environment in which patients and clinicians co-create decisions, clinicians must actively invite and support patients in the process, empathically “reading” the patient to match their evolving preference for participation. Tools to support this process need to be designed to facilitate and not interfere with empathic decision making, and this may form the basis for new measures of decisional quality.

Thus, we are not just for shared decision making. We are for empathic decision making.

Was a decision made? An assessment of patient-clinician discordance in medical oncology encounters

Oncology encounters are highly complex. Communication is suboptimal and there is evidence that patients and clinicians often fail to “get on the same page.” Shared decision making is being promoted as a means of facilitating effective and patient-centered communication in oncology. Here, Dr. Aaron Leppin and colleagues survey patients and clinicians immediately after an oncology encounter to determine the extent to which they agree on whether a cancer care decision was made during that encounter. The extent of agreement is impressively low. These findings have implications for the way we think about shared decision making and the validity of its measurement in oncology. (click here for  abstract)

by Aaron Leppin

The New Statin Choice Decision Aid

By Victor M. Montori

With a new interface that includes versions in English, Spanish, and Chinese, the Statin Choice decision aid (http://statindecisionaid.mayoclinic.org) is out. With over 70,000 uses worldwide year-to-date and new policy endorsements for its use (JAMA Article), the Statin Choice decision aid is helping patients and their clinicians have meaningful conversations about whether to use statins to reduce cardiovascular risk. It helps them adhere to the new guidelines, in a patient-centered manner. And with new work to integrate the tool into all major EHR providers, it may be the best demonstration of meaningful use.

Enhancements from the first version also include two options for printing in the office: color and black-and-white, in addition to the existing option to emailing the tool after its use to the patient, a family member, or another clinician. In terms of new content, the biggest difference is the exclusion of the aspirin component (see below).  We have also beefed up the Documentation tab, an copy-and-paste interim solution before full integration into EHR to enable documentation of shared decision making, a key step toward advancing these conversations as a measure of quality of care.

This version is the result of hundreds of notes suggesting changes and enhancements that result form the experience of using it in practice. We hope to have responded properly. And thank you.

Why was aspirin removed from the latest version of the Statin Choice decision aid?

In response to the new AHA/ACC guidelines for cardiovascular prevention, there has been renewed interest in using the Statin Choice decision aid to translate the recommendations in a patient-centered way.  With this attention, there has been interest from preventive cardiologists in using this tool. They brought to our attention that indeed the evidence about efficacy of aspirin for the primary prevention of cardiovascular disease is inconsistent: clearer effect in men in relation to heart attacks but not stroke, in women about preventing strokes but not so much heart attacks and a series of negative trials in patients with diabetes and peripheral vascular disease have made it difficult to provide a simple message to all at-risk patients: a baby aspirin can reduce your risk of cardiovascular events. Also, emerging evidence suggests that the risk of bleeding with aspirin goes up as the risk of cardiovascular events, such that those who may benefit the most are also most likely to be harmed (although most aspirin bleeds are relatively inconsequential compared to a heart attack or a stroke).

This inconsistency is reflected, for example, in the US Preventive Services Task Force guidelines: http://www.ahrq.gov/professionals/clinicians-providers/resources/aspprovider.html.

Concerns are best reflected in this FDA advisory against primary prevention with aspirin from May 2014: http://www.fda.gov/Drugs/ResourcesForYou/Consumers/ucm390574.htm

It is telling when experts are talking more about using aspirin to prevent colon cancer than to prevent cardiovascular events (to our knowledge no one is yet recommending it for this purpose).

We will continue to monitor this evidence as we, the producers of Statin Choice, thought the evidence was good enough to add to and keep in the tool, and we will have a low threshold to put it back in as new evidence emerges, both of its efficacy and harm.

Share decisions because it is the right thing to do

Written by Victor M. Montori, MD and Jon C. Tilburt, MD

Lee and Emanuel raise the profile of the shared decision making (SDM) provisions in the Patient Protection and Affordable Care Act. We concur that those provisions should spur research and development in SDM. However, their claims, that we already know how to implement SDM and that it is time for pay-for-performance for use of certified decision aids, are both premature and misguided.

Studies of decision aids implemented outside clinical visits show improvements in patient knowledge about the available options and about their risks and benefits, but not in actual sharing of decision making. Decision aids for use by patients and clinicians during the visit may work better. Video data from hundreds of recorded visits show a stark difference: patients in decision aid visits are better informed and participate more in making decisions. Patients and clinicians end up more comfortable with decisions they made together. In-visit decision aids galvanize patients and clinicians around a shared a purpose – to make the best possible evidence-based decision given the patient’s values, preferences, goals, and context. Yet, getting this degree of patient engagement does not happen with the flip of a switch and routine implementation remains untested.

Lee and Emanuel rightly point out the potential utilitarian benefits of SDM, particularly about surgical decisions. However, in so doing they jeopardize the patient-centered vision at the core of SDM. Moreover, their economic claim of cost-savings overreaches the current state of the evidence, making their SDM-linked pay for performance proposal premature. Research on SDM implementation is green, clinicians and patients are not ready, training and tools are just evolving. Thus, we support the law’s push for research and development. Their proposal is also dangerous. A focus on cost containment and pay for performance can corrupt the journey toward implementing SDM for all: we fear that the next time a clinician pulls out a decision aid, the clinician will be thinking about reimbursement while the patient wonders whether the clinician has her back.

Creating shared decision making visuals

There is quite a bit of evidence about best ways to convey risk information to help with policy or clinical decision making. Pictographs and bar graphs along with numbers and descriptions are considered best.  Some emerging research suggests that some elements will help some patients more than others (for instance people with low numeracy).

Recently, Fagerlin, Zikmund-Fisher and Ubel published their decalogue of risk communication in the Journal of the National Cancer Institute.  Their ten steps to better risk communication were:

  1.  Use plain language to make written and verbal materials more understandable.
  2. Present data using absolute risks.
  3. Present information in pictographs if you are going to include graphs.
  4. Present data using frequencies.
  5. Use an incremental risk format to highlight how treatment changes risks from preexisting baseline levels.
  6. Be aware that the order in which risks and benefits are presented can affect risk perceptions.
  7. Consider using summary tables that include all of the risks and benefits for each treatment option.
  8. Recognize that comparative risk information (eg, what the average person’s risk is) is persuasive and not just informative.
  9. Consider presenting only the information that is most critical to the patients’ decision making, even at the expense of completeness.
  10. Repeatedly draw patients’ attention to the time interval over which a risk occurs.

Online software to create pictographs can therefore be quite handy.  Some do so without resorting to giving each “person like you” an anthropomorphic shape .

Our favorite however, is one that shows the outcomes showing visual cues that are easily relatable, based on the iconic smiley face. We are impressed by Dr. Chris Cates’ Visual Rx tool, which is a free online tool that creates “smiley face plots” to depict the impact of a treatment on 100 people.

~Victor Montori, MD

What shared decision making means to the patient

With shared decision making, patients have the opportunity to discuss their lifestyle and preferences with their physician (or nurse practioner or physician assistant) to make healthcare decisions that are right for them.  This can be especially beneficial to patients with chronic conditions.  In these videos, two Mayo Clinic patients share their personal experiences with shared decision making and how it has changed the management of their own health. 

Minnesota Shared Decision Making Collaborative

Written by: Annie LeBlanc, PhD, Research Fellow with the Knowledge and Encounter Research Unit at Mayo Clinic

The Minnesota Shared Decision Making Collaborative hosted, in collaboration with Health Partners, a one-day conference on Optimizing Shared Decision Making in Health Care. Their effort to gather healthcare providers, patients and policy makers interested in Shared Decision Making was well received as there were around 70 attendees.

Internationally recognized speakers and local experts shared their experience on Shared Decision Making and its implementation in practices and then engaged the audience in discussion around the place and future of Shared Decision Making in Minnesota’s communities.

Their take home message:

Practice: Make Shared Decision Making a habit by practicing everyday;

Facilitate: Facilitate Shared Decision Making by (i) making explicit the decision to be made, (ii) providing balanced information about options, (iii) asking patients what matters most.

Process:  The implementation of Shared Decision Making, facilitated by patient Decision aids, in both primary and specialty may be at least a partial solution to the problems of poor decision  quality and unwanted practice variation in the United States.

Succeed. A daily, communitiy-wide effort is our best chance to make Shared Decision Making a National success

Educational resources for patients

Mayo Clinic patients have access to a broad array of healthcare education and counseling services through the Barbara Woodward Lips Patient Education Center, which enables them to be partners in their health care.   The Center makes efforts to reach diverse audiences, as seen in its new design for teen patient education material. 

In this video, patients and staff discuss the Barbara Woodward Lips Patient Education Center: