What To Say When The Chemo Did Not Work Video Transcript


Achieving Communication Excellence (ACE) Lecture Series
Dr. Anthony Back
What To Say When The Chemo Did Not Work
Date: October 21, 2008
Time: 42:36

Anthony Back, M.D.
Professor of Medicine, Division of Medical Oncology
University of Washington, Fred Hutchinson Cancer Research Center

Dr. Baile:
It's my pleasure today to introduce a good friend and colleague, and distinguished educator and researcher Dr. Tony Back who is a professor of Medicine at the University of Washington in Seattle. Tony is Director of the Cancer Communication in Palliative Care Programs at the Seattle Cancer Care Alliance and the Fred Hutchinson Cancer Research Center. Tony did his undergraduate training at Stanford and his medical training at Harvard and then his medical oncology training afterwards so he's a board certified medical oncologist whose research interests are in doctor-patient communication and palliative care and he also practices GI Medical Oncology.

Tony was a faculty scholar on the project on Death in America and is a member of the American Society of Clinical Oncology Task force on Palliative Care. He is the principal investigator in a program called Oncotalk Teach which is a program to teach medical oncology faculty, how to mentor and teach communication skills. He's also an NIH-funded investigator on studies of doctor-patient communication. So we'd like to welcome Tony and really partake of his wisdom on aspects of what to tell patients when the chemotherapy or other treatments aren't working anymore. Tony, thank you.

Dr. Back:
Thank you

[ Applause ]

Dr. Back:
So wow, people don't usually clap before I start and I know that today there's a lot of competition here. I know there's always competition for great speakers here but today I understand it's the Brazilian folk dancers so I guess the good news about this talk is that I'm not going to make you get up and dance.

[ Laughter ]

But anyway thank you. It's really a great pleasure for me to be here. I feel, you know with some of the people who are here, who are, you know, amazing people like Walter and Eduardo Bruera, for instance, and you know other wonderful faculty members I've met like Dan Epner and Vinod Ravi. I feel a little bit like I'm bringing coal to Newcastle here but I'm going to do my best and some of you -- I just wanted to say that I am coming at this from the perspective of an oncologist. I know that all of you aren't oncologists but I think you can't hear me because I'm moving away from this thing okay. Maybe the guy will be here later to bring the Lavaliere-- I'm so used to wandering around. I'll try and stay here.

I'm going to go with this from the perspective of an oncologist even though I know -- some of you may not be oncologists because I think it's just a way to open up a kind of dialogue about of what happens for people when their treatment is not working. So here's a little view of us when the sun is out which is not right now but maybe in August and I'm going to see, you know what, this remote thing is not working either so I'm going to have to use the keyboard.

Okay. So here is a quick case, a 55-year-old guy that I saw in clinic a little while ago. He has a metastatic pancreatic cancer. He had first line therapy and then second line therapy with Gemcitabine and Oxaliplatin. His CT, unfortunately, shows progressive disease and the bottom line is treatment failure and I think the dilemma for clinicians, like me and all of you, is what do you do in this situation? Do you just go for the next regimen? Is a third line regimen worthwhile? Should we be talking about phase 1 studies? Or do we really talk about, you know, is it time to stop anticancer therapy?

I think, this is one of the fundamental issues in oncology right now and the reason is, is because we have more and more anticancer options and the expectations of what places like M. D. Anderson can do in a difficult situation. Those expectations have grown and grown that I think the issue of a person wanting to just say it's time to quit actually is receding from you and if you look at what's happening with Medicare data right now, more and more people are getting treated with chemotherapy closer and closer to death. I don't actually think that's necessarily a bad thing. I think it's actually just a feature of where we are in the development of anticancer therapy.

So this is what I'm going to try and cover here a little bit about, you know, "What do I say?" And then I'm going to do that by talking about observational studies about what currently happens between doctors and patients and how those empirical studies could inform what we do. So, I'm going to say I'm not going to talk so much about, you know, how to talk about phase 1 studies. I'm not going to talk about how to make a referral to palliative care hospice. I'm really going to talk about the background in a certain way. How do you get the conversation started?

So I must start with a couple studies about what patients think about these things. A patient with metastatic cancer, if you look at the literature such as it is and it's a very small literature I have to say, they overwhelmingly want to know lots of stuff so here's one of the best studies that was done in Australia by Phyllis Butow's group. 126 patients with metastatic cancer were asked what they wanted to know and you can see all these numbers are very high. You know, 95 percent of them wanted to know about treatment options and side effects. Am I doing both mikes now? 85 percent of them wanted to know longer survival with treatment. 80 percent of them wanted to know five-year survival rate. 81 percent average survival. Interestingly, when you talk about survival and dying, the numbers are lower but there's this very clear suggestion that even at one time point people wanted to negotiate when to discuss these things.

What do family members say? Here, the data that we have and this is again very limited data is from a survey of parents at Dana Farber Cancer Institute done by Jenny Mack. Of 194 parents, 85 percent of them wanted a numerical prognosis. We talked about -- I talked about this with the fellows today. We had a great discussion about what these questions about prognosis mean? And 36 percent of them, a good third, found and felt them extremely upsetting. So this is top of the scale in terms of what they ask them about, very upsetting. This group was more likely to want additional information than parents who were less upset knowing what the prognosis was -- they were equally likely as people who were not upset to say that knowing prognosis was important, it helped in decision making, for that hope for cure will get them going. So what this is telling me that is upsetting is a separate kind of stream of data from whether they want a numerical prognosis and I know that each of us have this, you know, have had conversations where talking about prognosis made somebody really upset but the reality is, is in a way the upset is going to happen because you're in a difficult situation and wanting to know prognosis is a slightly separate issue.

I think the limitations of these studies are who participates in them. You know I think if you look at that Australian study that I just quoted, you know who are the people who are willing to fill out the survey? They're probably not all the people who show up in your clinic and I think that's the limitation of them but it certainly gives you a starting point of saying that there are a number of people who want to be able to talk about the details and they want to talk about the details because they want to be able to plan in it some way and that's the point we were talking about in our open discussion today is people want to know prognosis because they want to be able to plan. They're not asking you for the exact right answer. They're trying to figure out how to move their lives forward, so you know is this just another kind of bad news? The treatment didn't work. Well, of course, it is. I feel like -- this is where I feel like I'm really bringing coal to Newcastle but, you know, treatment failure is certainly fitting the definition of something that was also in the cognitive behavioral or emotional deficit that lasts but there's a different layer that's present.

In my case, it's not present as one of these that was diagnosed which is that I have been involved in this whole decision making process. I am the one who recommended two kinds of treatment. I'm the one who's telling him that both of them didn't work and the fact that I've been part of that, I think, adds another layer of complexity because now I'm complicit in what's happened? I'm part of the bad move. I'm part of a decision that, you know, maybe he's regretting so there is this whole dynamic in treatment failure that is more than what just what happens with bad news that's about what's said and what's unsaid and here I've summarized it a little bit. You know, one of my -- you know as this guy said, "Oh no now what do I do?" That's what he's saying and what I worried about that he's thinking is he -- he's thinking "You persuaded me to do this horrible stuff and now you're telling me it didn't work. What is the story here?" And then he said, "Are you sure the new chemo is going to work?" This was even before I talked about new chemo and he's really, I think, struggling in this issue of, you know, "Can I trust you for a third recommendation since the first two didn't really work all that well and what do I do here? What kind of life line are you going to throw me?"

Oncologists, understandably, are very ambivalent about the situation. It's a very difficult situation and if you ask oncologists what they think they actually hate it. They feel damned if you do, and damned if you don't. They got a lot of negative feedback from colleagues like nurses, chaplains, social workers, psychologists that they are not being realistic and yet they get negative feedback from the patients that they're taking away their health. They think they -- they -- oncologists can see that patients who are in these situations who don't make the decisions are losing opportunities at the end of life and yet confronting patients about the limitations and what medicine can do are quite painful.

You know what? I'm going to ask if - one of you guys can go back there. Can we turn this down so it's not so echoing, thank you? Can you turn it down a little more? Keep going. Keep going. Keep going. Okay. Thank you. So oncologists have mostly negative -- is Cathy here? Can you run up and see if any of the guys are still here? I don't see him up there. Thank you. Oncologists have mostly negative emotions associated with disclosing prognosis and, you know, it turns out they have -- there is this interesting phenomenon that if the bad news is concealed, they have a more positive affect. I'll show you that graph in a minute.

How does this kind of ambivalence and discomfort influence what the doctor says? Well, you know, there's not a lot of data about this, but certainly anecdotally, there are some clear indications about this. This is a quote from a study of mine where we interviewed and followed people who are actively seeking physician-assisted suicide and we interviewed them over a period of two years and this woman was a woman with advanced ovarian cancer -- thank you up there, who described her interactions work with her oncologist who she loved. You know, that's good Cathy, I think I'm good. And she recounted how he told her bad news and described how he would stand in her clinic room and back into the corner while he was talking and she would watch this and when he got to a certain place she would say, "Okay now we're going to talk about baseball." And she didn't discuss anything about dying because she was very curious about what happened. She said, "He becomes really uncomfortable so I've learned to change the topic to baseball," which blew me away because in a sense she is so sensitized to how much support this doctor can give her that she titrates the discomfort in the interaction for him. He doesn't even do it. So, as I said, no study is addressing this but certainly it's a concern and here's the graph that I talked about with medical students. Okay so these aren't oncologists.

These are great medical students but if you tell them you got to go in and then talk to the simulated patient about bad news, you know, if they are told that they are going to conceal the bad news, it's that black dotted line so they're worried, they anticipate it's going to be hard but at the end of the consultation, it wasn't so bad because they didn't have to disclose the bad news. If they had a control task, which was not giving bad news, their affect was pretty high and if they had them doing disclosure, then the blue dotted line, you know what? They anticipated, it's going to be horrible and then it is horrible, which you know, I think, what it tells me is if you're not trained it's easier to lie. So what happens in current practice? These are four little bullets that I'm going to describe in a little bit more detail as we go along, but you know, what happens now is avoidance of the difficult topic.

There's little acknowledgment of the emotion that underlies the encounter. Their explanations that are predominately biomedical and the focus is directed mostly towards the next intervention. And as much as I think that these are wrong or that these are improper because obviously, interventions are important -- biomedical explanations are important. It's the balance of these in the conversations that we see, that I think is the thing that is concerning and would help oncologists and all the rest of the oncology staff move people along in an easier way. So here's the non-acknowledgment collusion and this was a term coined by researchers in the Netherlands. They did a very complicated in-depth study of 35 patients with small cell lung cancer. They described what the doctors and patients talked about as a recovery plot and they also documented in these conversations, they went to all the doctor's appointments with the patients. The doctors didn't want to give the death sentence. The patients didn't want to hear it. I mean, I will describe this as kind of a "don't ask, don't tell situation" in which everyone is being very polite, they are being very cordial and they will only talk about what the big issue is.

What's the elephant in the room? The patients, in this study, found out that their own cancer was fatal from their own physical deterioration as they got more frail and they found out from talking to other patients in the waiting room who were further on in their course. And yeah, there's been a lot of discussion about this in the oncology literature. There was a well known oncologist who, actually I think, is a wonderful guy, Paul Helft in the end who wrote an article called "Necessary Collusion" in the Journal of Clinical Oncology in which he made a case that a certain amount of collusion with the patient, this kind of thing in fact is necessary to maintain the doctor-patient relationship. And, you know, I think, it remains to be an open question because -- and I think the issue may not be so much how do you collude but the flip side of that, how do you engage the patient and stay engaged with them?

And so, you know, what do you need to do in the engagement. You know engagement has a couple of aspects. There's a cognitive aspect to engagement. You know, how much do you -- how do you well understand each other and there's that -- then there's this emotional aspect of the encounter and I think of when I see any patient that I'm receiving two streams of data; there's a cognitive stream of data, what do they understand and then there is this emotional stream of data and the question that these other researchers were asking and these are some friends of ours was, you know, how often the oncologists do this? So this is a big RO1 funded study done by James Tulsky at Duke and Bob Arnold at the University of Pittsburgh. 51 oncologists, this is the largest study of its kind ever in the medical literature. They have encounters with 270 patients with advanced cancer and then almost 400 visits that were audio recorded and coded by blind decoder and they look to see how often the doctor acknowledged some kind of emotional statement that the patient would make so what they call an empathic continuer would be something where the patients says "I'm really scared." And the doctor says, "You know what, I know this is a hard time." And that would be a very elaborate empathic continuer. They actually coded as positive if the doctor like grunted in response right? The empathic terminators would be where the patient said "I'm really scared" and the doctor said, "Well, how many bowel movements did you have yesterday?" You know that kind of thing. And they look at how often this happened and oncologists made empathic statements in 11 percent of the conversations.

So, 1 out of 10 times that the patient made some kind of direct referral to the emotional content of the visit the oncologists responded, and then 9 out of the 10 times they didn't respond and what it creates is really kind of a learned behavior from patients about what it's okay to talk about in the visit. And, you know, when -- oncologists tell me they get stuck in these conversations where all the patients want to talk about is what's the next, you know, chemo. Part of this is a learned behavior, actually.

So this is the project that we did with fellows that Walter referred to, Oncotalk. I just was I just wanted to acknowledge my collaborators there; Bob Arnold, Walter Baile, Kelly Freyer-Edwards, and James Tulsky. We had an opportunity to train these fellows but also look at what they do in their sort of natural behavior before we train them and so what they did was we had them talk to a patient at the very beginning of the conference where they had to tell the standardized patient that the chemo is not working anymore. There's no chemo and we had designed these situations so there was no more evidence-based chemo and we audio recorded them. We gave the oncology fellows all 20 minutes to talk to them and then we went on with the course.

So this was the pretest and so we took those audio recordings and transcribed them and said, "How do oncology fellows do it as a way of seeing what, you know, behavior happens and so this is a study that I did with a medical student who has a master's degree and she presented earlier this year. So here we transcribed 20 of these and 19 of the 20 conversations relied on a kind of biomedical logic where the doctor would go through the science of the case and then describe how the science had kind of reached the logical endpoint and say okay so that means we're ready for a change, right? 1 out of the 20 really went for the patient values and worked those to understand them before making a recommendation that palliative chemo should be discontinued and it turned out -- we've selected 20 conversations at random and this one person had the highest pretest score that we had. And so it's just shows you, I think, that the impulse and the standard practice, I think, of oncologists is to go through the biomedical explanation of why something is not working and it's quite consistent with what you just saw before about the emotional content being addressed very scantily.

Here's an excerpt from one of these quotes. Of course, we have third line chemos, single agent with different chemos but the response rate is very low, 5 to 10 percent. If we give you some more toxic chemo and your response rate is very low, you have side effects you'd be in the hospital again. At this point, the options for chemotherapy are very limited. And you know, certainly, this is a reasonable biomedical kind of explanation. What you can't see from the slide is that this is a point in the conversation where the transcript changes from the back and forth that you can see on the page to like three pages of single spaced explanations about the basic science, about the history of the chemo therapy for this treatment, about the situation the patient is in. He goes on to elaborate about the different first line, I mean, the different single agent options there were and, you know, it's well meaning and obviously the fellow has a lot of expertise that he is bringing to bear on this but it's a little, I'd argue - a little misplaced.

The reason it's misplaced is because it's not all about cognitive understanding in the patient's mind and what you see again and again, I'm sure you see it here because we see it in Seattle constantly is that emotion overrides the cognitive channel so that when both things are happening and someone gets emotional the emotion just blocks it out. And so here's an interesting example from a study that I'm doing with some people at Dana-Farber where we are audio recording first visits of people who have hematological malignancies who are coming to Dana-Farber for consult and so what we do is we approach the patients before they have their first visit. We have them fill out some surveys and questionnaires. We have them audio tape their visit. We interview them and have them filled out questionnaires after.

Here is an example. A 38-year-old woman with AML before she went to see the doctor, she rated her -- estimate of being cured by a transplant at 90 to 100 percent. The doctor, during the consult this is part of his explanation, he says the following, "In regard to prognosis, so if you had a hundred people the survival curve drops down because people die of one thing or another including relapse that tends to level off at about 2.5 years after transplant and stays level after that." It's about 30 percent in your situation and the doctor, after this was done, we say, you know, what's your estimate that this -- estimate of a chance that those patients are going to be cured and the doctor said, 30 percent. So you know the reality is this doctor here is the survival period with this patient's condition. This guy is totally accurate. He's being totally honest with this patient, right? I mean, there are some data that only about a third of doctors will do this in these circumstances but this guy totally honest, right? I mean it's a complicated explanation and you can see him describing how the survival curve drops and I'm not sure if, without this visual, the patient would exactly know what he was talking but he is trying to be honest, right? So let's give them -- I think that's important.

So what does the patient think? So two weeks later, we do a phone interview with her. Post visit qualitative interview. This person is not involved with the consult. So she thinks after he said the 30 percent, he just kept dinging along in his facts and I was stunned. Literally, my note taking was completely done. All I wrote was 30 percent, the rest of the time all over my paper and I mean I just couldn't get past that point. I don't know how to describe it. What's so interesting about this encounter, for me, is that immediately after the visit, we have her fill out another survey about what she thinks for chances of being cured are now, right? And right after the visit, she had to think 90 to 100 percent. Her understanding that she was able to represent in this little questionnaire had not changed a bit despite the fact that the doctor was being completely honest. And I think the explanation is in this interview two weeks later when she actually came to terms with this, she obviously got the 30 percent at some level or else two weeks later, she wouldn't have been able to know how to do it because she didn't have anymore contact with the physicians about prognosis in particular.

You know, she talks to the cancer center about, you know, the logistics and the transplant and all that stuff but she never had another talk with this doctor. She clearly remembers that but the point is the emotion overwhelmed her in the moment and she was stunned and she just -- there was a way in which she couldn't re-articulate it for the purpose of our -- our visit. So I -- I think that's a really powerful demonstration of what the emotional channel means for people who are coming in for a visit, even quite capable people -- even very capable people. This is a woman who is highly educated. She was a Chief Financial Officer in a company. She was -- she had done all this reading before the visit. She was well prepared. She brought a friend. She brought a tape recorder, her own tape recorder, not our study tape recorder. She was taking notes and yet, this is the -- this is reality of what happens.

So, I mean I think one of the things that doctors often ask me about is well, do these big picture conversations really matter? How much does this really matter? Am I just laying this big trip on somebody and they're going to be depressed, then I'm going to have to deal with it? Well, I mean until now, we haven't had a lot of great data about this and there hasn't been a lot of research that directly addresses this question but now there is. So this is a big study that is -- that was done by Alexi Wright and Holly Prigerson at the Dana-Farber Cancer Institute and it's in press with JAMA and I think it's going to be published this week, actually. They prospectively followed over 900 patients with cancer and looked at what happened to them over the trajectory of their illness. 332 of those patients have already died and they have data about what they said earlier in their course and one of the questions they ask was, "Have you and your doctor discussed any particular wishes about the care you would want to receive if you were dying?" Right, and they just took the people who answered yes and got people who answered no and compared them on a whole bunch of measures.

So it's a very -- it's a very interesting point. I mean, it's only one kind of discussion point but it's still a very powerful thing because it talks about the patient's awareness and their previous history with their doctors. And so, when they compared the yes and the no, distress and depression, later no difference except the illness is terminal, the odds ratio is 2.1 Wants to know life expectancy, it's going to initiate some of the those questions about prognosis, odds ratio 2.4. Those are pretty impressive statistics for a simple comparison like this. If you look at what the other outcomes are, there are even more important medical outcomes that are associated with the patient's awareness that this conversation happened. The medical care in the last week of life, ICU admission odds ratio 0.35, ventilator use 0.26, the resuscitation attempt 0.16. If you look at the literature for other kinds of treatment effects, these are huge -- these are huge treatment effects, right? Out-patient hospice, more than a week, 1.58, that's also pretty impressive. Now, those maybe pretty impressive but there are some other things that, I think, are even more important to family -- the patients because they are both family members and family outcomes for these studies are equally impressive.

So caregiver bereavement, better quality of life for the caregiver after death, you know, look at these P values. The caregiver felt prepared for this. The caregiver had regrets about the death. I mean, I think, these are important data because they speak to the experience of the whole family and how in a way an un-thought through death has lingering consequences that are very -- that are very big both for the -- the medical care system and for caregivers. Yeah, one of the issues that people are always telling me is this takes too long. I don't have time to do this, right. And in fact, you probably -- you might have seen that in the New York Times. Pauline Chen the author of that terrific book "Final Exam" had a whole blog about do doctors have time for empathy and there were like 200 responses after one day of this because, you know, people are really concerned about this.

Well, it turns out it does not take that long. Here is a study and this isn't -- it's just one but it's a vivid one. You know, 123 women with breast cancer and 87 controls, they are randomized to see one of two videos of a physician talking about the treatment that the patients going to get and one of them has 40 extra seconds of empathic language. Of course, this is an oncologist study, I think you could tell because the assumption is that an oncologist would never do something like this for more than 40 seconds, I think. The oncologist on the tape, he is not -- it's not that I mean, for you guys, since you are obviously here, interested in this, I mean he is not like way out there. He's not like the Dalai Lama. He's not charismatic. He's not talking about anything terribly personal. I mean, he's saying stuff like, "This is a lot of information, so if I'm going too fast, stop me." I mean stuff like that and yet the women who saw the empathic video afterwards were significantly less anxious so that -- and so, empathy can happen really fast but it does take a kind of preparation and a kind of facility and a kind of clinician skill to make it work and to make it work in this kind of talk.

The other thing that I think oncologists worry about is, you know, am I just going to -- am I just going to disrupt that and am I going to be the kind of person who gives bad news that ruins their life forever? I mean, I work with a lot of fellows and they're genuinely concerned about the effect of these big conversations on people and yet, you know, the other emerging data is that if you look beyond just in the moment of the clinic visit to what really translates into peace of mind in a difficult time, actually having a good conversation is one of the things that contributes to that and here's some other data that was presented at ASCO this past year by Jenny Mack -- the researcher at Dana-Farber who worked with children. It's part of the same study, 194 patients, 22 rated themselves as more than 4 out of 5 -- equal or greater to 4 out 5 on peace of mind, 22 percent had peace of mind, 82 percent had a high score for a sense of purpose. And so, they asked what correlates with peace of mind and here are the things. They trusted the oncologist's judgment. The alteration was 3.9. That's huge.

The oncologist gave detailed prognostic information, 2.1 again, huge and I think those two are interlinked. If the oncologist is willing to be frank that helps them with judgment. Failing to correlate with peace of mind were the patient's prognosis and time since diagnosis and I think the important point of this study is that just having a bad prognosis doesn't actually correlate with peace of mind.

There's other stuff that's going on, that often I think we don't know about and the other thing is that time does not necessarily heal all wounds. Time with the situation of a child and your child dying of cancer, yikes! Time with that situation actually did not improve their peace of mind and I think that shows actually the potential role of what clinicians can do in these situations because, in fact, we are the catalysts of how people think about their illness and so what we do, I think it takes on a huge importance.

Here again, the other issue is hope, you know. The patient -- doctors, trainees -- I work with lots of fellows. They're always saying, "I'm worried about taking away someone's hope." And this shows less what -- what makes you hopeful, and actually the more elements of prognosis that the doctor talked about, the more the family felt hopeful that they were being taken care of. The high quality communication, if they rated the communication as high quality, odds ratio is 6.5. I think this just goes to show how powerful your interactions are and how, in fact, shouldn't take them casually. You shouldn't take them for granted. In fact, I mean the impulse that -- that clinicians have, that how they talk to people is important is exactly right but it's not necessarily in a direction that we expect.

One of the things -- one of the other pieces of research that we're doing is, you know, we can, now we've actually done some work to show that we can teach doctors these things, well do patients know all this stuff. That's one of the things I'm interested in. And so, we're taking some of these audio tapes that we got from Oncotalk and we're playing them for patients. So we have the patient with cancer listen to the tape with an interviewer by a doctor talking to another patient. We have the patient comment on what they like and what they didn't like and we did this for -- we're just starting to do this for the transition conversations. One of our early really interesting pieces in front of this is, you know, is what do the patient's want and you know what they want is this combination between empathy and some guidance. It's something that I don't think really exists in the literature in quite this way and the -- what I mean by subject area is that this is the cancer patient who's listening to the tape. He said, "The doctor has to let him run his emotions a little bit." This is right after the bad news. "Personally, I'd say, son of a bitch, now what do we do?" It's such a great comment because it's both of those pieces in one -- one short little piece. "Son of a bitch" is acknowledging. "Yikes, this has happened to me. Now what do I do?" That's the guidance part and the people were very concerned of about how doctors frame their experience. They didn't want to be told, it's too scary because if the doctors says, it's too scary, then it's really worse for me.

Another thing is because I don't want to be told how I feel by a doctor or anybody else. So our emerging conclusion of this is that acknowledging emotion is critical but it's important not to over specify or paint the picture for them. You have to let it emerge and pay attention to what's emerging. You know, can oncologists actually learn this stuff? I don't -- I don't know how you got what's happening in terms of quality, but you probably know there's this whole sort of quality movement that basically is ignoring doctors because they view doctors as completely untrainable. And, you know, the thing I always get from these people is, you know what, you're not going to be able to make doctors change in this way and you know the reality of that is that it's just not true.

This is a landmark study done by Leslie Fallowfield of the UK where she randomized senior oncologists. These are not fellows. These are senior oncologists to a small group intervention, very much like the one we did in Oncotalk. We used it as the basis for our study or detailed written feedback on videotapes of their own clinic. They went to these doctor's clinics. They filmed them with patients. They gave the doctors detailed written feedback and the detailed written feedback had absolutely no effect. Only the small group work changed them and the reason I think it changed them is because you have to learn how to practice. You have to learn by practicing. It's not enough just to know what to say. You have to be able to say it and it goes back to the thing with the great medical students. If you -- if you can't be convinced, you can do it, it's much easier to just -- to short change in terms of what the lengths to which you're willing to go.

Here's some of the data from our study that was in the Archives of Internal Medicine. So this is -- if you look at the top row, you know, how often did the fellow acquire a new ability to assess patient understanding when they had not done that in the pretest and you can see, you know, these are pretty substantial changes for educational intervention. You know, what does this sound like, the patient, you know, we have the patients -- the standardized patients and these things happened to have required behaviors and one of the required behaviors is that the patient says, "Isn't there anything more you can do, doctor?" And here, the doctor pre-Oncotalk doctor said, "Well, you've had a lot of chemotherapy, haven't you?" which is what, you know, this is what you call it a distancing behavior. This is -- this is like anti-empathy. Afterwards, same doctor, same cue, different patient scenario. She said, "There are more things we can do. Yes, this has been a roller coaster ride for you, hasn't it?" I mean these are the kinds of responses that materially change the feel of the conversation for a patient and you know here, one of these other fellows wrote a little first person narrative in the Journal of Clinical Oncology about her experience with this in learning and I think that sounds interesting.

This is about what it's like to become more accomplished. I feel less flustered and my words are less tangled. I can focus on the person across from me and find out what they need from me in that moment and that feels like progress. So notice that she has changed her view of progress from not wanting to upset the patient to this sense that she can respond in the moment to what the patient needs. That's a very sophisticated change in what you should be doing with the patient.

So this is our new project, Oncotalk Teach that we've had the pleasure of working with some of your faculty in. And so, I was going to stop here and change directions a little bit because I wanted to show a little bit about what happens afterwards. After the optimism, there's so much emphasis on optimism, what's the -- what's the downside about it in a certain way or what are some of the consequences and this is another RO1 funded study that I did with a colleague of mine in Seattle and what we did was we studied oncologists, pulmonologists, and internists. We defined a group of patients with advanced COPD and incurable cancer for whom the doctor wouldn't be surprised if the patient died within a year. We did loads of interviews of them over time to see how they talked about open information and the thing that was interesting about -- this study is in press in the Archives of Internal Medicine, is that the patients all brought up the issue of not wanting to be abandoned.

It wasn't in our interview guide. And so, I was like, "Well, how can we keep that? How can they keep talking about this?" And the qualitative researcher said, "Well, if they keep talking about it and listening, they really think it's going to happen." And she was right actually. So here's the deal. You know, non-abandonment is all over the literature, right? But there are survey data that show that patients and families still sometimes report feeling abandoned and so, we were able to, in a way, study this prospectively because we identified all the patients before death. So, here are some of the key findings. Patients were worried about abandonment. They were worried about abandonment in a specific way about whether or not they would have continuity with the expertise of the doctor that went along and said, "I think it's important that you still have that contact with them even though there isn't anything they can do to make you better." They are so conscious of this. I mean it's -- it was eye opening to me.

Another one just said, "Just stay with me baby," when the interviewer asked him what he needed. Another one said, "I need to be able to depend on my doctor. If I call you next week, doctor, will you see me right away?" They are -- they are so conscious of what they need from your clinical expertise. The physicians, actually, were quite aware of this and they anticipated this. They were -- they tried a lot to reassure the patients about continuity. One of the doctors said, "What patients need to know is that you care about them. When they know that, it's much more important than anything." And another one said, "I want patients to know that when the going gets tough, I don't just bow out." The thing that changed after -- when the patients got close to death was that people switched to a new topic besides continuity which was all about closure.

After death, one of the patient's spouses said, "The one thing that really bothered me is that doctor so and so never called me. To me, it was almost like a slap in the face." This doctor actually had done a very responsible job of referring the patient to hospice. The patients who went to hospice had a pretty peaceful death, wasn't -- wasn't physically an issue but this was a lingering issue for this spouse and she said when she went further, she said, "I mean, I realized the doctor is busy. He's got other patients but she knew -- but he knew her husband and I, we were together through the whole thing, it would have been nice just [inaudible]." And so, it gets kind of poignant and yet, you know, what -- and then here's the other interesting part is the doctors did not describe it like that. The doctors -- what the doctors described is lack of closure. So this -- the person that we were just listening to, this person, her doctor said, "I thought, oh I should really call B." That was the spouse's name and he'd been out of town actually when the patient died and then he said, "I didn't just -- because I was too busy. Then I felt guilty. I hadn't called her earlier and I didn't know if it was appropriate. I thought maybe it would stir up -- just stir up emotional stuff if I did call." It's okay. He was really -- he was concerned about what his call might mean and in fact he never did call her.

Another one of the doctors in our study said maybe there are things I don't recognize psychologically. It probably would be good for me to have better closure and what was sounded was this kind of poignant disconnect between the patients and the doctors. You know, why do the -- why do the doctors say this? So, you know, they talk about their -- how busy they are. They talk about the lack of time. They talk about the fact that this isn't reimbursed. They talk about they're not sure exactly what to say. They -- I think they are worried. Some of them were worried that the patients -- the spouses would blame them for something, but in fact, what the spouses wanted was -- what this woman wanted. She just wanted to touch base with the doctor again. That was really all she wanted and we had a couple of instances -- we have one instance of a doctor who did this in the study with 55 patients. We had one doctor who did this and the spouse was glowing about it. She said, "You know, that's how I know we weren't just a number for him." So, where does all this take us? So I have a couple of recommendations. We're talking about treatments down here and so one is to track this emotional data. What's the patient's reaction? Acknowledge the significance to the patient's feelings, balance, you know.

As you've talked about balance, the biomedical part of your conversation with what the patient's interpretation is and what it means for them in their world. Talk about the big picture. Step back and say, you know what? I'm going to just step back from this and talk about the big picture, what's most important for you at this point." And as you can see, that's the beginning of a conversation but I know it can feel awkward because it's -- a lot of times, those conversations -- those questions don't have immediate answers but I think what you're doing is planting a seed that will germinate and grow over the next few visits that you have with this patient and then finally, don't hesitate to step forward and then take an active role in guiding the patient.

You know, I've heard and seen so many fellows offer choices about end-of-life care to the patient like the patient is just supposed to choose from this menu and the patient seemed, you know, to me often very concerned about wanting some more active end of care and one of the doctors is going to make a recommendation because that doctor knows them personally and I think that's the piece of this that comes through in terms of guidance. So with that, I just want to acknowledge my collaborators and funders and Walter and this is the situation you're trying to avoid, doctors without boundaries. "So, can we have all your stuff when you're dead?" Anyway, so thank you for the invitation. I have -- I think we have a little bit of time for questions although I don't have a clock. Thank you.

[ Applause ]