Whose opinion matters in medical decision-making?

Caring for patients involves the formation of multiple opinions. Traditionally, the patient's physician generates the first opinion, with other clinicians offering second opinions. However, an opinion at least as important must also be recognized in this traditional rubric—that of the patient and family members. Just as the conscientious health care practitioner is constantly reformulating his or her opinion based on many factors, so too is the patient.

 --Recognizing an Opinion: Findings From the IOM Evidence Communication Innovation Collaborative JAMA. 2012;():1-2. doi:10.1001/jama.2012.13369

How important is the opinion of the patient and/or family in medical decision-making? And how can these opinions and preferences be better integrated into clinical care? A discussion paper published this month from the Institute of Medicine (IOM) entitled Communicating with Patients on Health Care Evidence attempts to tackle these questions and other issues related to improving the quality of medical decisions. The impressive list of authors, who are all thought leaders in areas as diverse as academic medicine, public policy, and consumer empowerment, were participants in the "Evidence Communication Innovation Collaborative of the IOM Roundtable on Value & Science-Driven Health Care." (Now, there's a title.) Props to the IOM for releasing the full text of the paper as a freely available PDF and to JAMA for posting the Viewpoint summary article from Novelli et al also with free access to full text. 

A few key takeaways from the report:

• 8 in 10 people want their provider to listen to them.
• 8 in 10 people want to hear the full truth about their diagnosis.
• 7 in 10 people want to understand the risks of treatments.

BUT:

• 8 in 10 people want their health care provider to listen to them, but just 6 in 10 say it actually happens.
• Less than half of people say their provider asks about their goals and concerns for their health and health care.
• 9 in 10 people want their providers to work together as a team, but just 4 in 10 say it actually happens.

The report defines the three components of informed medical decisions as medical evidence, clinician expertise, and the patient's goals and concerns. In Figure 4, survey data are shown demonstrating that patients place nearly equal weight on each of these, with a slight edge to medical evidence as the leading component. Interestingly, women were somewhat more likely than men to say that their personal goals and concerns were "very important" (64% vs. 50%).

In addition, the following language was deemed in patient interviews and focus groups to "resonate best" in regards to medical evidence:

Making sure you get the best possible care starts with you and your doctor making the best decision for you. Your doctor can help you understand what types of care work best for your condition, based on medical evidence. Because there are always new treatments, doctors use this evidence to keep up with which work best. Your doctor's experience helps him or her evaluate and apply the evidence to your situation. The doctor also needs to listen to you so he or she understands your values, preferences, and goals. This is important because every patient is different, and when there are options, it is important for the doctor to know what is important to you.

In the conclusion, the authors call for the following actions:

1. Cultural changes - providers should recognize existing gaps and adopt decision aids when available and other tools, including mobile technologies, to facilitate the presentation of knowledge and care coordination
2. Incentive alignment and infrastructure support - patient-centered medical homes and EHR's that promote "meaningful choice" are singled out
3. Quality standards and accountability - including a suggestion for professional licensing/accreditation bodies to incorporate criteria encompassing patient-centered medical decision-making

So is there anything new here? And what are the lessons for oncologists in particular? As a busy clinician in a breast cancer practice at an academic medical center who every day tries to guide patients through difficult decisions - whether or not to receive adjuvant chemotherapy for an ER positive cancer, whether to enter a Phase I trial when conventional options are exhausted for refractory disease, whether to be hospitalized for IV hydration or pursue a purely palliative approach at home with hospice - my first reaction, and admittedly not my best reaction, is to go into Defensive Mode. Of course, I always consider patient preferences. Of course, I always give people the full truth about diagnosis and prognosis. Of course, I explain risks adequately. And so on. But based on this paper and numerous other examples, it doesn't take much self-reflection to acknowledge that my colleagues and I probably aren't hitting the mark nearly as often as we would like to give ourselves credit. The one concept from this paper that resonated the most with me was the emphasis on the idea that patients want to know all of their options for a given decision, not just the ones that I as the provider favor. Using the example of trying to decide whether to administer adjuvant chemotherapy for an ER positive, node negative breast cancer (let's make it more complicated and say the Recurrence Score is 22, and there's no trial avaiable), what options might I present to a patient, who is otherwise fit for chemotherapy and willing to receive it? Well of course I tell them that tamoxifen or an AI alone may very well be sufficient treatment. And then after trying to put the use of adjuvant chemotherapy into an historical perspective, I discuss with them the option of a standard regimen like TC x 4. I do usually tell them that no systemic treatment is also an option, although not one I advise. And within the limits of our knowledge and predictive ability, I try to estimate the risk of recurrence and the potential benefit of the intervention in numerical terms, trying to emphasize absolute benefits instead of relative. (I confess, I have been known to start the conversation by saying that an AI will "cut the risk of recurrence by half." I'm not immune to sound bites.) But along this vein, do I give them the option of TAC x 6, even if in my opinion the potential benefit is not high enough to justify the greater toxicity of a 3rd generation regimen? Not often. Do I mention regimens that I just don't use, like CMF or M->F, knowing that there are a few places that still do use them, and there is some evidence to support their use? No, I don't. Or do I regularly discuss a possible role for adjuvant bisphosphonates, recognizing they are not FDA approved for this indication, and I am personally not convinced that we have enough data, even in postmenopausal women, to demonstrate a net benefit? No, not usually, unless asked.

Why don't I do these things? I suppose I partially justify it by saying that I simply can't recommend and discuss as a viable option a treatment that I don't feel passes my own evidence test - not necessarily my own experience. And, being realistic here, time constraints do play a role. In a perfect world, a new consultation would last 2 hours, 3 hours, or however long was needed to convey all of this information, or would extend over multiple visits. But if I did that with everyone, my new patient slots would dry up, and it would take weeks and weeks to get an appointment with me, which means the next patient needing consultation would be kept waiting in frustration. Furthermore, no matter how skillful a presentation of options is in a clinical encounter, a recitation of treatment choices rendered without context or hierarchy is frequently meaningless and confusing to even highly sophisticated patients. So does this mea culpa make me a non-patient-centered oncologist, and guilty of the failings described in the IOM report? Like most of us I would hope, I have an infinite capacity for self-criticism and second guessing, so I suppose the answer is that, yes, at times I indeed need to be more mindful of the alternative approaches and considerations illustrated in this piece. But I offer this reflection to illustrate how complex and nuanced the physician perspective is in the spectrum of informed decision-making. 

Lest my comments be interpreted as critical of IOM or suggesting this discussion paper is some ivory tower assessment devoid of the realities of practice in the modern era, far from it. I salute the IOM for focusing on this issue with the right combination of balanced analysis and practical suggestions. I will try to reflect on the recommendations they put forth in all of my patient encounters this week and put the principles of patient engagement front and center. I challenge all of my clinician colleagues reading this blog to do the same. And to our patients in the difficult situation of being on the receiving end of these discussions, I salute you and recognize you are the ones thrust into this situation unwillingly and often with little time to prepare for such weighty decision-making. Help us your caregivers help you. Tell us what you need and when we have failed to include you.