Scenario: Oncology attending and fellow rounding in the hospital, seeing a new patient.
History: This patient presented with no past medical history to the ER with shortness of breath. She is a non-smoker and was found to have bilateral pleural effusion and a miliary pattern of lung nodules. She has already undergone thoracentesis and is pending results.
As the team is discussing the pending pathology results so that we can discuss cancer diagnosis, staging, prognosis, and treatment options, she gently asks:
“so how long do you think…”
How do you respond?
What is interesting to me is that as we walked back after the meeting, I asked my fellow how she thought the sentence would be complete when she heard it. How she would respond?
How do you think that the patient would complete this sentence?
Take a pause and think about it before going on to the next paragraph.
What is the question?
Some may suggest that the question is:
So how long do you think I will be in the hospital?
So how long do you think I will be “around for?”
So how long do you think I will live?
And depending on how the sentence is completed, the answer may differ. In this case, my fellow felt the question was - about the length of hospitalization. I had thought it was about prognosis. Clearly, we don’t know what the patient is asking at that moment.
Many times, when there is a pause, we complete the sentence depending on what we think the question is, rather than what the patient is asking.
So, the first thing we have to do BEFORE answering the question is to find out what the question is.
How do you elicit the entire question?
We can just pause… let the patient complete the sentence.
We can pause AND, with our non-verbal cues (intently listening for them to complete the sentence), provide space for them to complete their question.
We can also ask a clarifying question - “tell me more…” or “What do you mean?”
By understanding the patient’s question, we can avoid the mistake of answering the wrong question. Since patients are often reluctant to ask the question again, even if we answered the wrong question, it may lead to them not getting the answer that they need.
The next section assumes that the question is: “How do I have to live?” (which is what this patient wanted to know).
Discussing Uncertainty
So, how do you answer the question about prognosis? If you ask any oncologist, there is uncertainty. I was reading a substack by Dr. Rebekah Diamond, a pediatrician, and you can see a typical answer there. The oncologist responded: "Nothing is 100% percent certain in medicine... we can never predict anything exactly."
And quite honestly, it is true that we don’t know exactly how long one has, and there is no certainty. Focusing on the uncertainty, though, can seem like avoiding the question.
In our case, here is what I know as an oncologist (these are facts based on data). Depending on what her pathology and molecular testing shows, median survival is about 2 to 5+ years. Variables to this range include specific pathology and molecular drive of her cancer, how effective the treatments would be, and how toxic the therapies would be.
One way to answer is to give ranges (day to weeks, weeks to months, months to years, 2 to 5 years etc.)
Another way to answer (and this is my preferred method), is to ask the why behind the question. I learned basic skills on this when I collaborated with my mentors and colleagues in 2009-2010 and published an article in a special issue of The Cancer Journal focused on the role of the oncologist in palliative care with the title: Discussing Prognosis: Balancing Hope and Realism.
I have evolved my response in the last 15 years. One very important skill here is authenticity. It is not about avoiding answering the question but engaging in the conversation to help the patient in their journey (be their sherpa). In my experience, there is a question behind this question. It is that second layer of question that I want to know. What are their goals? Is it their son’s wedding? grandchild’s graduation? Daughter’s PhD defense? Travel the world?
The first thing I say, looking right at the patient (while I am sitting, and if appropriate with my hand on their hand) is, “if I knew the answer, how would you use that information?” or [with a dry smile] “I wonder what the real question is.”
A few things are happening in this dialogue. First, I speak only one or two sentences (and not filibuster the conversation). Second, I acknowledge either the uncertainty (by saying ‘if I knew the answer’ or head-on address their why by asking them to give me the question behind the question.
Almost always, the answer I get is. “I know you don’t know,” or “I know you are not God,” or something along those lines. This prevents me from highlighting the uncertainty and getting to the second sentence, which is their why. Most times, it is about their plans for the future; sometimes, it is about setting their family up financially, their responsibilities, or their plans.
In this case, I used the first sentence “if I knew the answer, how would you use that information?” and the patient replied: “I know you don’t know, but I think it is only a few months.”
In this case, while I didn’t get her why, I did get her worst-case scenario. Since I had that, I was able to focus my answer on that specific part. I focused on providing information about their prognosis (years 2-5) based on what we find out on their pathology.
Now, many of you may have some questions about the choices I made.
Q: Are you not avoiding answering my asking a clarifying question?
A1: Very rarely, I have gotten (more from family than patient) the response that “I mean, it’s obvious, isn’t it?” And if that occurs, I normalize by saying “I have learned over these years that many times patients ask this question for different reasons, including some feel they are supposed to ask.” This is where authenticity and the credibility earned before that question is asked come into play.
A2: more often than rarely (about 20-25% of the time), when I ask a clarifying question, patients tell me they actually know the answer. And then, they talk about what is important to them. Which is the entire purpose of asking this question.
—> even if they answer the question for themselves, I do further probe about their goals and what they want to do with the time that they have. This will be further discussed in Goals of Care post in the near future.
A3: No, never. I will always anwer the question to the best of my ability. I just want to make sure that the answer is heard. Almost always, the patient acknowledges the uncertainty, so I don’t have to.
Here is a typical interaction I have had with my patients.
Patient: How long do you think…
Doctor: [intently silent]
Patient: … I have? How long?
Doctor: [looking right at the patient] Is it ok if I ask a question before I answer your question?
Patient: Sure.
Doctor: I wonder what makes you ask that question? - or - If I knew the answer, how would you use that information? - or - I wonder what the question behind that question… what are you really asking?
Patient: [looking away] I know… I know you don’t know… It’s just [voice breaking down] its just, I had so many plans. I just retired, and we had plans to travel the world and live a little. You know, we had so many responsibilities caring for my family. This was to be our time.
Doctor: [ looking at the patient, with outstretched hand]
Patient: I read online that I won’t make it to my next birthday.
Doctor: So, that is your biggest worry?
…
Doctor: I know you told me about your worries, can you tell me a bit about what you are hoping for?
…
These skills require deliberate practice. So next time a patient starts to ask “Doc, how long do you think…” first understand the question and explore what the real question is. Be genuine and caring. This is significantly more scary for the patient than it is for us. And yes, it is terribly scary for us.
Campbell, Toby C. MD*; Carey, Elise C. MD†; Jackson, Vicki A. MD‡; Saraiya, Biren MD§; Yang, Holly B. MD∥; Back, Anthony L. MD¶; Arnold, Robert M. MD#. Discussing Prognosis: Balancing Hope and Realism. The Cancer Journal 16(5):p 461-466, September 2010. | DOI: 10.1097/PPO.0b013e3181f30e07
Additional Related Readings in this curriculum: