 Good afternoon. I'm pleased to welcome you to today's lecture in our annual series on ethical issues in end-of-life care. Today's talk is the seventh in the 27 Talk series that will run, as you know, through this quarter, next quarter, and the spring. I did want to caution everybody that we will not have a lecture next week because of the Thanksgiving holiday. We will resume our series on December 3 with a lecture by Dr. Daniel Callaghan. Now it's my absolute pleasure to introduce our speaker today, Dr. Jennifer Temmel. Dr. Temmel is an associate professor of medicine at the Harvard Medical School, the associate director of the Hemang Fellowship Program that combines the Dana-Farber and the MGH programs, and the clinical director of thoracic oncology at the Mass General. Dr. Temmel's research in clinical care of patients with lung cancer focuses on investigating palliative and supportive care and in studying doctor-patient communication about prognosis and decision making for end-of-life care. It was almost 10 years ago that Dr. Temmel co-founded a research group at the MGH, the mission of which was to reduce suffering from cancer by doing research that aimed to provide, to improve, the delivery experience and outcome of cancer care for patients, families, and caregivers. Many of you know of Dr. Temmel's distinguished earlier work published in the JCO and the famous New England Journal article of 2010. These were the basis of subsequent NIH-funded R01s to continue the work in palliative care. Essentially, Dr. Temmel examines the implications of early integrated palliative care in patients who have advanced cancers, beginning with advanced lung cancers, but now I think the new grant includes GI cancers, except for colon, except for colon. Dr. Temmel has recently been awarded an NCI mid-career development award to mentor others in palliative and end-of-life care and oncology. I'm particularly excited to learn that there is a new randomized controlled study of the effect of early palliative care that is under the auspices of Alliance, which is the old CALGB, through NCI funding. And the University of Chicago is one of the sites, one of the participants in that multi-center study looking at the impact of early palliative care. Today, Dr. Temmel will speak to us about the topic that you see up there, communication about prognosis and end-of-life care in patients with advanced cancer. This happens to be a topic very close to my heart, because the very first paper I wrote is the one that Dr. Christakis says is the first paper to talk seriously about prognostication in medicine, this paper on Pascal's wager. But now you're gonna hear something 40 years further along in progress. Join me in please in welcoming Dr. Jen Temmel. So I'm gonna be talking about communication, about prognosis and end-of-life care in patients with advanced cancer. So much of the data I'm going to show you is specific to cancer patients, and I know not everyone in the room is an oncologist. So I've sort of interspersed a clinical case and sort of some anecdotal discussion, so it won't be so cancer-focused for the people who are not oncologists. So just to start with a case presentation, this is actually a patient that I'm still taking care of. So this is Sharon, she's a 58-year-old woman who had back pain for a few months and then had some scans, which revealed that she had metastatic disease in her thoracic spine and her pelvis, which ultimately led to her having a chest CT, which is shown here, and she had a four centimeter mass in her left lower lung and also had diffuse lymphadenopathy in her chest that was consistent with lung cancer. She did undergo a biopsy of one of these lymph nodes, which confirmed adenocarcinoma, and obviously the topic of this talk is what should I be telling her about her prognosis? So just to start, what is the definition of prognosis? Today I'm gonna be talking all about why we should be disclosing prognosis to our patients who have serious illnesses, but what does prognosis mean? Well, the first definition I have on this slide, that sounds pretty easy, the prospect of recovery as anticipated from the usual course of disease. But this next statement, which is probably a little bit more accurate of what we really need to do when we're taking care of patients with an advanced illness, is a judgment about what's gonna happen in the future. Well, darn, that's kind of hard and scary. How do any of us know about what's gonna happen in the future? And that's really what the challenge of discussing prognosis is, is that uncertainty about what is going to happen in the future. And this is by no means a data-driven slide. This is sort of the Tamil method of discussing prognosis and communicating about end-of-life care to patients with advanced cancer. And I think it's important, everybody does this differently to find something that works for you and you feel comfortable with. But sort of the strategy that I take is, for me, it is important in the first couple of visits that I have with a patient who has an incurable illness, for me to disclose the incurable nature of their illness. I feel like if you don't disclose that information in the first couple of visits, you kind of never get that time back. That patients will form sort of misperceptions of their illness or sort of carry on an inaccurate perception and you never give it back. So my strategy is always to disclose that someone has incurable disease at the time of diagnosis. And while this slide is sort of written as one step to another, often you have to kind of revisit these topics and go from one to another. But the other strategy I use is basically acknowledging when someone's health status is declining. Because when someone's feeling well and doing well, it can be hard for them to focus on, for example, planning for end-of-life care, discussing resuscitation preferences. But I sort of use their declining health status as sort of a trigger to discuss future plans and then recommend transitioning to a focus on comfort. I think one thing that's really important to remember when we're talking about prognosis is even though I'm saying, I think everyone who has an advanced illness that's incurable needs to know they have incurable disease. If you are a patient who's diagnosed with an incurable cancer or other illness, you kind of can't wake up every day, stretch, and be like, I know I'm gonna die. You just can't live that way. So I think what we should anticipate is that patients sort of move out. I don't like the word denial. I prefer the word sort of less realistic hopes and acceptance as they integrate information about their prognosis. Vicki Jackson, one of the palliative care clinicians that I work with talks about opening the box and closing the box. That some days patients are comfortable talking about their prognosis and their mortality and they open the box. And then sometimes that's sort of all they can process and handle and then you close the box and put it away for another time. So how do I communicate with patients who have a new diagnosis of advanced cancer about their prognosis? Well, I always explain to patients what the stage of their cancer is and why. So in this case, if they have metastatic or stage four cancer, and the words that I tend to use is that the goal of treatment is to prevent their cancer from growing and spreading, which may, underline may, help them live longer with their cancer, but that treatment will not offer a chance of cure. And again, for me, making sure that I actually say treatment will not offer a chance of cure is just an important part of how I communicate with patients. And I never, for example, start chemotherapy or start a treatment plan without making sure that I've explicitly said that. So what happens after that? What happens after I say chemotherapy is not gonna cure you? Well, sometimes patients are so overwhelmed because sometimes that's their first or second visit with me that that's all they can handle and they don't ask any questions and sort of a conversation ends there and we have to revisit it at a future date. Sometimes patients ask about their life expectancy right away. They look at you and they say, well, doc, how long am I going to live? And again, if it's a first visit or two and patients are quite overwhelmed, I often do a double check that they're really up for that. I'll often say something like, we've talked about a lot of things today. You just found out that you have metastatic disease. Are you really up for talking about life expectancy today or would it be better to do that next week or the week after? And you have time to think about it. When we do discuss life expectancy, which I think is very important, especially on our first couple of visits, I actually try to frame the discussion and I try to be optimistic without providing false hope. So I often say something like, even five to 10 years ago, many patients with metastatic lung cancer did not even survive a year, which is true. But now we commonly see patients live a year or sometimes even two years, also true. Often the conversation ends there as patients are often shocked that their life expectancy may only be a year. So often, even though patients have metastatic lung cancer and they can look on the internet or open up a magazine that will tell them that their life expectancy might only be a year, I would say that all of these patients walk in and they truly expect that their life expectancy is five to 10 years. And they're always shocked to hear that that is, while not impossible, incredibly unlikely. Often after saying a statement like this, patients will say to me, wait, you mean I only have two years to live? And that's a problematic statement because unfortunately many people with metastatic lung cancer actually don't live two years. So I usually follow up that by sort of trying to explain in lay language what we mean by average life expectancy. And then I try to frame the discussion based on the individual patient. So I usually say something like, when we say patients live on average about a year, that means about half the patients live shorter than that and about half the patients live longer. And again, then I sort of look at the individual patient and try to frame it based on my assessment of their age and health status and comorbid disease. So if they're a patient who's younger than me, I would often say something, because you're young and healthy, I'm hopeful that you may do better. But if it's an 85 year old man with cardiac disease, I'll often say because you're older and have other medical problems, we should be prepared that your life expectancy may be shorter. What about when patients say, yeah, that's all good. Thank you for telling me what the average patient does. I need to know exactly how long I'm going to live. And obviously that's the question that every single patient wants to know. And the reality is it's what each and every one of us in this room would want to know if we were diagnosed with an incurable illness. Exactly how long do I have to live? But the reality is most of the time when patients are diagnosed with their cancer, they look like all of us. Many of them feel like all of us. And there's truly no way for an oncologist or an ethicist or a palliative care condition to tell them exactly how long they have to live. And this is when I really explain to patients that if I could tell them exactly how long they had to live, I absolutely would. And I explain that this uncertainty, this unclarity about the future is one of the hardest parts of my job. And I cannot tell a patient how long they have to live. But again, the strategy that I use, which again works well for me in my communication style, is I tell people that when I start to worry that they're getting close to the end of their life and that is the language I use, I let them know that I will let them and their family know so that we can start making plans and decisions about their future. So again, this is a question they really want the answer to. I cannot give it to them, but I assure them when I do have the answer, I will let them know. And I do feel like, although this is somewhat of a scary statement because it's patients with newly diagnosed disease and I'm already referencing the end of their life, I feel like it gives patients and families a sense of security. I'm not gonna forget that you need to know this and when I know it, you will be the first to know. So what did I do with Sharon? I did communicate to her that she had metastatic cancer and had an incurable disease and that her life expectancy would depend on her genetic testing. So for those of you who aren't oncologists, there's now a couple of different genes that patients with lung cancer have which will predict a potentially longer survival for them. I did tell her because she was young and healthy, she had never smoked, she had no other disease that I was hopeful that she would do better than other people with this disease. And again, I did assure her because she was very insistent on knowing exactly how long she had to live like many young people are that I would let her know when I was worried. So am I doing the right thing? Do patients want to know about their prognosis? What information do they want? What is the right time to discuss their prognosis? And do we cause harm to patients when disclosing their prognosis by taking away their hope or causing anxiety? And I'm gonna run through some data sort of answering these four questions now. So what do patients want to know about their illness and prognosis? And again, because I'm an oncologist, I think all of the data that I'm showing you today are from patients with cancer. There have been studies that look at broader populations, for example, patients with cancer and COPD or congestive heart failure, and the themes are generally the same. So this is a study that asked 380 patients who had advanced cancer, simply do you want to be informed about your terminal illness? These patients were sort of in all points in their illness trajectory. They were just sort of people who showed up in the waiting room and were given this questionnaire. But what you can see is that the majority of patients said yes, they want to be informed about their terminal illness. This study asked patients specifically about their communication preferences. What type of prognostic information do they want? 95% of patients wanted to know the chance that treatment would shrink their cancer. 92% wanted to know those factors that make their prognosis better or worse than average, kind of like I did with Sharon. She's young and healthy. And then if you look at the bottom box, while their percentage of patients endorsing these statements is slightly lower, you can still see that the majority of patients want prognostic information. They want the average length of time that they could live, the chance of living five years, the chance of living one year, that the majority of patients want this information. Now again, certainly patients and families do need to negotiate sort of the timing and the content of this information. We were just talking a couple of minutes ago. Most people can't handle this information all at once, but it's sort of a process letting patients and families sort of dictate what is the optimal timing and content that works for them. Similarly, this study of another, the same group of patients with advanced cancer specifically asked about physician behaviors when discussing prognosis. 98% of patients said they want their doctor to be realistic about their likely future. They want to know their prognosis. 92% wanted their oncologist or their clinician to emphasize what can be done and what can't be done. 90% wanted to know their prognosis before anyone else. So I think what you're picking up from these studies is that every single study that has asked patients with advanced cancer whether they want to know their prognosis has always been a resounding yes. This is some work that we did at Mass General of patients with advanced gastric and fat biliary and pancreatic cancer. How important is it for you to know about your prognosis? The overwhelming majority range from somewhat to very to extremely important. And when you ask patients about the details that they want about their cancer diagnosis and treatment, you can see that 76% want to hear as many details as possible. So again, very, very consistent. In fact, every study which has ever asked patients any of these types of information continue to show patients want this information about their prognosis. So what about the timing? When is the right time to disclose prognosis? Again, by telling patients pretty much the same day that they have metastatic disease, do I tell them that they have incurable disease? Is that the right strategy? So this study looked at that same group of patients with advanced cancer and asked about the timing of disclosing specific prognostic information. So they asked about treatment goals and options, symptoms and side effects, likely duration of survival and dying or palliative care services. And you can see that the majority of patients wanted to know about their treatment goals and options when the cancer has first spread. Now, certainly as you go down that row, you can see that most patients didn't really want to talk about dying when their cancer has first spread, which again, I think based on that sort of that trajectory that I told you, I think patients need more information before they're ready to talk about death and dying. So I don't find that data entirely surprising, but again, as soon as the cancer spread, patients want to know their treatment goals, again, which is curable or incurable and their options for care. This is the same study that I showed you on the first slide, which enrolled about 380 patients that patients wanted to be told that they have terminal illness. And as a follow up question, they were asked, what is the appropriate time to be informed about a terminal illness? And again, you can see that the majority of patients said immediately, some differences with family members feeling like gradually would be a better strategy. But again, I think the real message is that patients do want to know this information and they clearly don't want to wait till the very end of their life to hear this information. So I think this is pretty compelling data that patients want this information, they want it early, but do we have to give it to them? It's really hard to sit with a patient and their family and tell them that they have an incurable disease. It takes a lot of time, they cry, sometimes I cry, I don't want to do it. What's the big deal of patients with an incurable illness walking around thinking that they're gonna live longer than they are? Why is that such a bad thing? Oncologists say that all the time. So the problem is that how patients perceive their illness and prognosis impacts their decisions about care. And we in oncology have known this for two decades, dating all the way back to their support study. And that makes sense how we perceive anything in life impacts how we make decisions about that particular topic. So we know that patients who have an inaccurate and overly optimistic perception of their prognosis are more likely to choose aggressive therapy and less likely to receive hospice care. So again, in the support study, patients who estimated their chance of survival were two and a half times more likely to receive life extending therapy. In the CANCOR study, which is a large cohort of patients with advanced lung and colorectal cancer, patients who overestimated their chance of survival, which the majority of patients did, were significantly less likely to have discussed hospice care with their clinicians. And so again, this is also a slide that I showed yesterday, but the problem with this is the trend towards increasingly aggressive care, which in general isn't in concordance with patients' goals and preferences. Patients don't wanna spend the end of their life in the hospital and they don't wanna be getting chemotherapy at the end of life if it isn't going to help them. So it's incredibly important that patients have this information, have the knowledge of the goals of their treatment and what it can and cannot get them. So do patients understand their prognosis and treatment goals? And without showing you the slides, I bet you can all guess that the answer is no. So this is actually, this is one of my favorite slides because this is actually from my palliative care study. So when patients signed for a palliative care study that we did where patients with newly diagnosed lung cancer were actually randomized to standard oncology care or early palliative care, immediately after filling out the consent form, they were asked to fill out their baseline surveys, asking them about quality of life, asking them about symptoms, and also asking some questions about treatment goals. So these patients had just signed, it's at Harvard, so it's like a 20-page consent form that literally said they had advanced cancer, maybe 10-hour consent forms are really that long. Yeah, exactly. That's why no one ever reads them. And explain to them that they were, on this particular study, I'm sorry, so these patients all had metastatic disease and they were within eight weeks of diagnosis to be eligible for this study. So these patients all had new diagnosed disease. So you can certainly argue that they didn't have an opportunity yet to talk about some of these issues, absolutely. But again, I think it's ironic that they had just enrolled in a palliative care study, but when we asked them about sort of their goals of treatment, you can see that a third of patients thought their cancer was curable and 69% thought a goal of treatment is to get rid of their cancer. And again, I do think it's important that sometimes the way we ask questions can be really confusing for patients. And I can see why the goal of treatment is to get rid of all my cancer may be a little bit confusing, but the reality is at least at the time that this study was done, complete responses to chemotherapy in this disease were zero. So this is clearly a misperception about treatment goals. This is also from that GI cancer study. So again, these were patients with advanced gastrointestinal cancers, pancreatic cancer, stomach cancer, esophageal cancer, who were within six to 12 weeks of starting chemotherapy that their oncologist documented in their chemotherapy order was not being administered with chemotherapy intent. And to your point, the reason we enrolled patients six to 12 weeks after starting chemotherapy is that should darn be plenty of time for their oncologist to communicate with them what the intent of chemotherapy is. I would argue that you should not write an order for that first dose of chemotherapy through the IV until the patient knows what the intent of their chemotherapy is. And you can see that more than half of the patients rated their chance of cure somewhere between 25% and 90% with less than 10% of patients endorsing no chance of cure. So again, I don't always know, Monica and I were just talking about this, I don't always know that what patients will actually mark on a piece of paper and what they really feel are the same thing. And I try to put myself in the shoes of these patients when we give them these packet of questionnaires. And if I had an advanced cancer and Monica asked me if I knew my prognosis, I think I would probably feel comfortable saying to her, yes, I know I have an incurable disease, but when I have a pen in my hand and it's asking the chance of cure, I could see that it would be hard to mark the no-box. So I think all of these studies, including the one that I'm about to show you may have some biases. We all have a little bit of magical thinking. Writing absolutely no might be a little hard, but I think it's still representative of the fact that overall patients do have misperceptions about their prognosis. And so again, this is from the CANCOR study. So over 1,000 patients with metastatic lung and colorectal cancer that were asked the chance of cure with their chemotherapy. And again, you can see that the minority of patients stated not at all likely. The majority of patients either rated their chance of cure as a little likely, somewhat likely, and very likely. So again, even though I don't truly believe that all of the patients who endorse some chance of cure necessarily really believe that, but I think it at least gives us a glimpse that many patients with cancer have a limited understanding of their prognosis. So this is a study, I'm gonna show you a couple of slides from this study because this actually asked family caregivers. And I actually do think that's important because while certainly it would be hard for any of us to be diagnosed with an advanced cancer and hear that they having curable Z's, and it's certainly hard for your loved one, but in a lot of studies, they've shown that it's a little bit easier for the family member or the loved one to hear and recall difficult information. So this study looked at family caregivers' perception of the timing of prognostic disclosure. And you can see that the majority of family members stated that they weren't disclosed about the patient's prognosis one to six months before death or less than six months before death with only a small proportion about a third knowing more than six months before the patient's death. And 20% of the family members stated that they were never told by the physician that the patient's illness was incurable. Now that part I do have a little bit of a problem with because often we know in the room that we can disclose information to patients and they don't hear it, they don't accept it, it's very difficult for them. So I focus less on whether the MD actually told them or not and I focus more on the patient doesn't know and it doesn't matter whether we told them once or we told them 15 hot times. It's our job to tell them enough so that they know and can accept it. So again, I think one of the sort of follow-up studies to the previous study I showed you from the Cancor study was there was a lot of written in newspapers that doctors aren't disclosing prognostic information. And while certainly that's part of the problem and I'll show you some of Chris's data that suggests that in phase data, I think that is only part of the problem. So what's going wrong when it comes to patients' understanding of their prognosis and treatment goals? We know patients want this information, they want it early and they're not clearly either getting the information or understanding the information. So you here at, were you at University of Chicago when you did this? Yeah, you must have been. Really? That makes it easy for me. So surveyed hundreds of oncologists and basically asked of the following examples, choose which one summarizes your communication with your advanced cancer patients about their prognosis. And only 40% of the clinicians said I always discuss my patient's prognosis with them because they need to know it. Now I would argue based on the studies I previously showed you, this number should actually be quite much higher than it is. We also know that from previous work by Nick Christakis, who was here as well, that even when we do disclose prognostic information, we often disclose inaccurate or overly optimistic prognostic information. You can see that 16% of clinicians said they only discuss it if my patients ask about it. Why would a patient ask about it, right? I mean, this is really hard information, although we know they have to know it. They don't want to initiate these conversations. Study after study shows that the patients are waiting for their clinicians to bring it up. And then another third of clinicians said I ask my patients if they want to know their prognosis and discuss it if they say yes. So again, I would argue that although those are probably truthful answers, we really would expect that the majority of clinicians should say my patients need to know this information and I always give it to them. And again, this is from that same study where they looked at family caregivers. And you can see that even family caregivers didn't know that the patient's illness was incurable until the discussion occurred or after. You know, very few patients and families really sort of understand, even though, again, you can go on the internet and you can see that patients with advanced lung cancer have incurable disease. It's very difficult for patients and families to process this information until they're doctor. The person who's taking care of them and supposed to be shepherding them through their illness is actually the one to tell them this information. So in that same study, they also asked in telling how often do you give your patient a specific timeframe or medical estimate of the amount of time as when death is likely to occur. Now you can tell from some of those first slides I showed you. Again, you can't tell a patient who has newly diagnosed disease exactly how long they have to live. But what I can tell you is if you do not give them a timeframe, they all walk out and expect that they're gonna live five or 10 years. Because my colleagues in the lung cancer group tend to not give patients a timeframe. And when I'm covering them when they're in Chicago giving talks and I talk to their patients, I can tell you all those patients think they're gonna live five to 10 years and are shocked when they hear otherwise. But again, when you look at the responses to this, the majority of doctors really say it's not really their practice. They said usually sometimes rarely to give a timeframe with only a tiny minority of clinicians saying they always give this information. And this is my favorite part of the study, right? So they ask the clinicians, if you were a patient with advanced cancer, would you want to be told your prognosis, including a medical estimate of the time until death? And of course, we all endorse that we want this information for ourselves, but somehow we don't make it a priority to tell our patients. Chris, any other comments about that? I love this study, it's great. There's a correlation, and that makes sense. If you were a patient with advanced cancer, would you want to be told your prognosis, including a medical estimate of the time until death? And of course, we all endorse that we want this information for ourselves, but somehow we don't make it a priority to tell our patients. Chris, any other comments about that? I love this study, it's great. There's a correlation, and that makes sense. Those are the oncologists that go home after clinic and drink. About drinking after work. So again, I would argue that the response to this question should really be higher than 42%. So I think there's a lot of things that can go wrong in the room, and even the best oncologist, we can still have patients who walk around and have misperceptions of their prognosis. And I think lots of things can go wrong and do go wrong all of the time. So as I was getting at before, I don't think none of the physicians if the patients in that can-course study told them their prognosis and told them they had incurable disease. I think lots of times we say this information, but patients can't hear, accept, process it. It happens all the time. I am taking care of an 82-year-old patient who has an EGFR mutation, so it's getting treated with the pill treatment, Tarsiva or Erlatinib. And I told her when she was diagnosed that she had incurable disease and she's been on the drug for about 10 months. And initially her cancer shrunk and now it's been stable. And every time she comes in to look at her scan, so she's maybe had five scans and she's been on the drug. I like do my cheerleading. I'm so excited. Your cancer shrunk or your cancer hasn't grown. This is such great news. And every single time she is disappointed because she asks why is the cancer still there and shouldn't it be gone because the goal of the treatment is to cure her cancer. And I am a billion percent sure. I tell her every time. I have a fellow in the room who looks at me like, Jennifer, you said this last time. Even the patient's husband who's actually 90 looks at his wife and says we've talked about this every time, but she has no recollection of it. And it's not because she's 82 and demented. Her mental status is clear. She cannot accept, hear or process this information. So I think I'm a good communicator and a great oncologist, but if you ask her any of those questions about what her prognosis is, I can tell you that I'd be put in the bad doctor category that I didn't communicate. And I feel like I'm hitting her over the head with it, but she just cannot accept, process or hear it. I think another thing that goes wrong in the room and I'm sorry to insult the junior people and the fellows that are oncology fellows do this all the time. Senior attendings do too, but we think we disclose the information. We think we've said it, but we use vague words and terminology that patients don't understand. We don't use the word incurable. We don't explain that the treatment's not gonna cure their cancer. Sometimes I hear fellows use the word remission for lung cancer and it just makes my skin crawl and I seriously have to correct them because remission is not a term that has any applicability in solid tumor oncology. So even though these words incurable, life expectancy are hard to talk about, patients are overwhelmed, they're stressed and if you don't just say it, they're not going to understand it. I also think sometimes we sort of get the courage and we disclose the information and we get it all out, but like my patient, my 82-year-old patient, they're too overwhelmed to hear it, but we think we're done. We check that box and we never go back and we never bring up the conversation again and so patients forget, they don't remember it and so then they go off with some significant misperception. I think what happens a lot in the room to many oncologists is what I call the accidental collusion. So this happens to me all the time. I've spent 15 or 20 minutes with the patient and family. I've run through this all. They cry, I'm supportive, I think we're good. I'm like, I just had a really good conversation. I'm feeling really good about myself and I put my hand on the door and I'm like ready to say see you next week and the patient looks at me and says so doc, we're gonna beat this thing. And I have my hand on the door and I think, God, it would be so much easier to just go see the next patient, but I force myself to turn around and say, can you just clarify by what you mean by beating this thing? Because if you do not, you're sort of two thumbs up or see a next week, it reinforces that misperception. You're sort of sending them subtle clues that yeah, you said that they had incurable disease, but you're kind of leaving the window open that they're gonna be that one special patient. So I think that accidental collusion happens all the time and we don't often always realize it. I think the doc I'm gonna beat this thing is a little obvious, but all the time in clinical counters, patients will make very subtle, overly optimistic comments and you sort of have that internal conflict. Do I dress it? I'm really late for clinic or how do I manage it? And I guess I think it's really important to take your hand off the door, sit back down and really ask the patients what they meant by that statement. So another important question that comes up is do we cause harm by disclosing prognosis? So again, we talked about the fact that patients need this information to make decisions, but do we overwhelm them? Do we upset them? Despite the decision-making part, you can say I don't care about that. Does this information cause patients harm? There actually isn't a lot of data out there on this. This is one study from one of my colleagues at Dana-Farber, Jenny Mack, who's done a lot of work studying parents of children with cancer. Not all of these children, in fact, most of them didn't have absolutely incurable disease. They sort of had a chance of cure. But what this slide shows you is that these are the physician reports of likelihood of cure and this is sort of the odds ratio that the parents felt hopeful. And you can see that even when there was a less than 25% chance of cure, the parents were still hopeful as long as they had disclosed a prognosis. So I think it is possible that patients can maintain hope even when you have a difficult process of prognosis. We talked about this a little bit yesterday. Hope is a really complicated construct and I think depending on how these questions are answered, what does someone really mean by hopefulness? But I at least think some information that we can give this information, we can present it in a way that patients and families in this case can still feel hopeful. But I do think there are clearly gonna be negative consequences of telling someone that they're gonna die. I mean, nobody wants to die. That's really hard information to think you can say this, even if you say it very well, very empathetically, very compassionate. Of course, patients are gonna have a hard time with it. So this is a study that we did where we looked at patients who either endorsed that they understood they had a terminal illness or they did not understand that they had a terminal illness. And again, the word terminal illness is a little bit of a tough one, so that may bring up emotion in patients. You can see that patients who perceived that they had a terminal illness, they actually had worse quality of life and they did have higher rates of anxiety even after controlling, for example, for their health status. Even when you look at that Jane Weeks New England Journal of Medicine paper, and again, this got a lot of attention in the press. If patients think they have incurable disease, they rate their physician's communication much higher and better. So again, of course there's negative consequences to telling someone they have a poor prognosis. It's a really hard thing to accept. I mean, for any of you in the room, I hope none of you have one, but you probably have loved ones who do. And it's an incredibly, incredibly difficult thing. But I do not think we should interpret this as saying we shouldn't provide prognostic information. We just need to do it in a compassionate and empathetic fashion and make sure that we're supporting patients as I cope with this information. And again, remember, it's not a one shot deal. You're not empathetic and compassionate once and then it's over. I think we need to discuss it and provide support over time as patients acknowledge and process this information. So again, for me, what works, again, I'm not saying this is the right way to do it. I think we all, depending on our patient population and our communication style, need to find something that works for you. But for me, again, this closing the illness is incurable. That's a process that may take months or even years when patients have a decline in their health status, sort of acknowledge that and then use it as a trigger to eventually discussing end of life care preferences and recommending that someone not be resuscitated and recommending hospice care. So just transitioning to time communication about end of life and hospice care. So how do I initiate discussions about end of life care? Well, when patients have progressive disease, cancer that's growing on treatment, but are clearly well enough to tolerate another line of therapy, I sort of use this as an opportunity to let them know that the chance that this next line of treatment will control their cancer as well or as long as the previous one is unlikely. But when patients are becoming more ill when they're just getting sicker, they're spending more time in bed, they're out and about less, they're family saying they're not eating as much, they're not going shopping with them, they're not going out with family and friends. Whether or not they have progressive disease, I let them know that I'm starting to feel worried. So again, that message that I gave them at the time of diagnosis, I will let you know when I am worried. As soon as I start to feel worried, I fulfill that promise and I let the patients know. And the reality is sometimes I'm wrong, sometimes my worry bell goes off and it's just the patient was having a bad time with their carboplatinum and pathotaxyl and they get better and that's okay. But again, I follow through on that promise and I let them know that I'm starting to be worried that they're becoming ill and we may need to start thinking about their end-of-life care preferences. Usually patients feel overwhelmed no matter what their health status is like, that you're sort of giving them that warning shot that now we have to talk about end-of-life care. So I do sort of let patients know but say we're gonna do this next time. I think it's helpful for patients to have a little bit of a warning, maybe even talk it over with their family and then talk about it at a visit or two. I also personally think it's hard to have these conversations on a same day that someone for example, cancer is growing and you wanna consent them for a new line of treatment. I think most of us only have 15 or 20 minute appointments and this is a 15, 20 minute appointment. So I think doing it at a visit when someone's actually doing well actually gives you more time and space to have these conversations. I also think this is a very hard conversation to have if someone's feeling nauseated or in pain or feeling sick. So I actually think having these conversations during a relatively well visit is a better strategy. So going back to Sharon, she did have one of these good prognosis EGFR mutations and she got one of these pill treatments, Tarsiva or Lattinib for nine months which is actually a little bit short for someone with this good prognosis cancer type. And when she developed progressive disease she actually got very, very ill very quickly. She developed multiple bone mets one after the other that continued to require radiation therapy in the middle of her third round of radiation therapy for painful bone metastasis. She got brain metastasis and needed brain radiation. She was a little bit tricky because often these patients get treated with very specialized clinical trials and she really wasn't eligible for them. So we treated her with a chemotherapy called Pemitrexid which is actually usually very well tolerated but she became much worse. Her performance status declined. She was spending more than half of her time in bed and really not up and about much. So again, I acknowledged that she was becoming more ill from her cancer. At that time I let her know that her prognosis was likely months and probably less than a year. So again, she was just getting sick. So I really didn't have a good accurate prognostic estimate for her but I thought her life expectancy was probably less than a year. And again, this was a significant surprise for her. She still expected to live years and years and years despite all the work that we were doing. And I didn't even mention it because it's outside of the scope of this talk but she actually had enrolled in the palliative care trial and was receiving early palliative care. So she was having prognostic discussions with me and her palliative care provider. We did discuss the fact she had young children and hadn't given them a lot of information. So we also told her that again, while she was still relatively well, this would be a time to start talking to her children and we gave her the warning shot that in upcoming visits, we'd have to talk more specifically about her end of life care. Very shortly after that clinic visit, she was admitted to the hospital with uncontrolled pain, increasing shortness of breath and changing mental status. She was so ill during that hospital admission, I actually was concerned that her prognosis might only be weeks. But again, despite clear communication, despite palliative care, she later told the social worker that she had six months to live, which led to a stat page from the social worker to me. Why didn't you tell her about her prognosis? I hate those pages, we get them all the time. During her admission, I recommended that she become DNR, DNI. So again, I used her declining health status to sort of transition to recommending a focus on comfort, which she agreed to. She didn't feel comfortable proceeding with hospice at that time, but we did do a bridge to hospice with home nursing. So what is the standard practice for discussing end of life care and hospice care? Well, not surprisingly, these discussions usually happen quite late and not necessarily in the setting that we would hope for. So this is again, from that large can core study with over a thousand patients with advanced GI and lung cancer. And what you can see is the topics discussed, maybe you can't see. Sorry, on the topics discussed about end of life care, 80% of these conversations discussed hospice and only 46% discussed resuscitation, which I would say discussions about resuscitation need to happen, they need to happen early and it has to be a part of any end of life care discussion. What the second box says that you can't read is the venue of the end of life care discussion. So earlier, I argued that these conversations should happen during a good visit when things are calm, when patients are feeling okay, 50% of these conversations happened when a patient's in the hospital. By definition, if you were in the hospital, you feel sick and that is not a time that you wanna have an in-depth conversation with someone about your end of life care goals and preferences. And this is probably the hardest slide that I show. So the timing of the first end of life care discussion for the cohort of patients who died, the median days between the end of life discussion and death was 33 days, that's a month. Is 33 days enough time for you to consider all of your options and do all the things you need and want to do before you die? Absolutely not. You know, these patients have so many things that they need to do. You know, obviously for patients with advanced cancer with family, leaving a legacy, writing letters, making sure if they're not gonna be at their daughter's wedding, they write a letter that can be read. These are important things that take more than 33 days. This is again, the study of the family caregiver. So again, you might say the patient doesn't recall the conversation that the family caregiver would. 30% of family caregivers reported that the physician had never discussed hospice. And again, the majority of hospice discussions happen very late in the patient's course of illness. Again, I bet you can't read that one either. So this is a slide that I also showed yesterday. But the question is, do we cause harm by engaging in discussions about end of life care? This is from Holly Ferguson's Coping with Cancer Study. And I will just tell you, since you can't read it, patients who participated in end of life care discussions were no more likely to feel depressed or have anxiety. They were no more likely to feel depressed, nervous, sad, or terrified. And importantly, they were more likely to accept that their illness was terminal and make better decisions about their care. Specifically, patients who participated in end of life care discussions were less likely to be admitted to the ICU, less likely to be resuscitated, and more likely to have a longer length of stay on hospice. Even more importantly, this is again from Jenny Mack from the CanCours Study. Patients who participate in end of life care discussions receive care in line with their goals and wishes. Because the reality is there are some patients with advanced cancer whose goal is to die with chemotherapy in their veins or die being resuscitated. And you know what, that's okay. It's not our job to tell people how they want to live their life. And so there are patients who want that. They're not the majority of patients, but it's important that we make sure that patients are getting care in line with their goals and preferences. And what you can see is that patients who discussed their end of life care wishes were two times as likely to have end of life care that was consistent with these preferences. I know this is a little bit of a busy slide, but again, sort of as Chris's study asked the doctors about what they disclose about prognosis, this is very similar. So in the CAN course study, the physicians caring for the patients in the study were asked, assume you're caring for a patient who is newly diagnosed with metastatic cancer, but currently feeling well. You estimate the patient has four to six months to live. When in the course of the typical patient's illness are you most likely, for the first time, it's a very run-on sentence, to discuss the following with the patient or family. And again, they asked all different types of doctors, but the second row in each of these is the medical oncologist. And this is specifically about resuscitation status and site of death. And what you can see is that the majority of oncologists for DNR status or site of death said, when there are no more non-palliative treatments, only if the patient is hospitalized and only if the patient and family brings it up. So again, this is resonating from the slides I showed you about prognosis. We really tend to defer these conversations until it's very late in the patient's course of disease. So for the last three minutes, I'm very quickly just gonna run through some suggestions that we can all take away from this room about what we can do to ensure that our patients are aware of their prognosis and engage in discussions about their end-of-life care preferences and hospice in a timely fashion. Well, number one, obviously, oncologists can do it. We can make sure we have these conversations and do a better job with it. If you can't or if the patient's having a tough time, you can ask your palliative care clinicians to help and there also are decision aids to help. So again, we don't have time to go into these in detail, but I'm just gonna highlight these three topics quickly. So there are actually some great papers written about communicating with patients. And for those of you who find this hard, I'm sure you're like, yeah, an article is gonna help me have these conversations with patients. But I would almost promise you that these two articles would. So this is written by some of our Fay and my colleagues, Tony Bach and Barb Arnold. There's one for discussing prognosis with patients who are ambivalent or not sure if they wanna hear prognosis. And one that's talking about patients who are ready to discuss prognosis. And these are two amazing articles that every time I read them, I learned something new and pick up a new kernel about how to communicate with patients. Let's skip that. This is a study that we did where we actually sent physicians, oncologists only, not palliative care clinicians, reminders, like, hey, don't forget, your patient has incurable disease. You may wanna discuss their code status. And pretty remarkably, just reminding oncologists, they were three times more likely to actually document a code status discussion. So again, there's lots of little things that we can do. We can read articles, we can do training like Occantalk or Vital Talk, and we can remind our clinicians to have these conversations. This was a slide I showed yesterday we can also call on our palliative care colleagues because we know that patients who have early involvement of palliative care are more likely to accurately say that their cancer isn't curable. So we can certainly rely on our palliative care colleagues. And then there's a great deal of data written on decision aids. Decision aids are basically either written information or videos that you give to patients to inform them about their choice and help them make decisions about their care. I would say they've most commonly been used for decisions, for example, about undergoing prostate cancer screening or breast cancer screening. But there is some data for decision aids for end-of-life care conversations. There was a study where patients were given a question prompt list with the types of questions they should ask their doctor about end-of-life care and that improved communication about end-of-life care. This is a study from one of my colleagues, Angela Velandez, who actually developed a video where patients were shown what resuscitation and intubation was, physically actually shown what it's like to have a breathing tube and get chest compressions. And you can see that after watching the video, significantly less patients wanted CPR because they actually understood what it was and frankly how horrible it was. So decision aids, I understand that a video showing CPR may feel a little dramatic, but even just giving patients a list of questions, hey, why don't you ask your doctor about hospice and what does that mean can be very helpful. So in summary, patients with advanced cancer prefer to receive information about their prognosis and to discuss their end-of-life care preferences. I think that's clear. It's not every single patient that walks into my office, but it's most of them. While communicating that a patient has an incurable disease, it's time consuming, it's sad, they cry, I cry. It is absolutely essential that we do it and we support patients and families through this process. And I think it's essential that communication about these topics, it's not a one-shot deal. You don't sort of, you know, it's not diarrhea of the mouth, you don't do it all at once. You need to interweave these conversations with patients and families all throughout the course of the disease. You need to check in with them, you need to make sure they understand it, and you need to support them regardless of what is going on in their illness trajectory.